AlastairThomson.com

Many people think numbers are “objective” and therefore represent some pure form of decision-making. Nothing could be further from the truth.

For now, I’ll skip over the question of whether objective decision-making should be regarded as the best kind of decision-making, as I have some questions about that. But even if it is, equating numbers with objectivity is just baloney.

There’s nothing less objective than a number – a number on its own means nothing without the full context.

Here’s a quick example.

Let’s imagine we did a survey of 30 people in a room and asked how many of them had heard of Einstein’s famous E=mc² formula.

And let’s imagine none of them had.

That’s a real indictment on the UK education system, isn’t it? Possibly the world’s most famous equation ever, and nobody out of 30 people had even heard of it, never mind knew what it meant!

Shocking, right?

Well it is. Until I give you the context that this imaginary experiment was done with a group of 30 under-5s at a nursery school.

A number without context means absolutely nothing.

Politicians are the best at this

Just before we get into the meat of this article, it’s worth saying that I’m not taking a political stance here. This is just about how to interpret numbers – I’m not supporting one interpretation or the other, just pointing out how people use numbers to mislead others.

If you know me at all, you’ll know my views that, irrespective of party, I regard all politicians as sitting on a scale that varies between disdain and outright contempt. So I’m not supporting any party’s political views here.

However, exactly the same tricks are deployed by people in business every day of the week -generally less expertly than the average politician, but the themes are broadly the same so, by the end of this article, hopefully you’ll have some pointers about what to look out for.

It’s no exaggeration to say that being able to spot flaws in the numbers that are presented to me has been at the core of the career success I’ve had. and it comes from something I learned at university, where I studied Law.

A professor in a second year law class told us that, in every legal dispute, there were always at least two sides – sometimes more – and, whichever party we represented, it was our job to uncover all the different sides to the argument, and examine whether they were persuasive or not, in light of the available evidence.

It was not our duty just to parrot our client’s case as they relayed it to us, he said – at least not unless we wanted to lose a lot of cases, because our clients would generally only have one view of the matter. Their own. They were blind to all the other ways of looking at the situation

Seeing the other side of the argument was drummed into us then and it’s a skill that has carried me through my career.

So, whenever someone sticks a set of numbers in front of me, they are essentially making a case like a lawyer would in court, wanting me to agree with their view of the situation. Which is why I always ask myself “what might the other side of the story be here?”

If you get nothing else from this article, whenever anyone presents you with a set of numbers, ask yourself that question. Don’t be railroaded into agreeing with whatever the person presenting those numbers wants to do.

Otherwise you might end up unintentionally raging about why a group of 4-year-olds doesn’t know all about E=mc².

Comparison is the thief of joy (and also thinking)

US President Theodore Roosevelt supposedly come up with the expression “comparison is the thief of joy”.

Whether he did or not, isn’t the point (online sources differ), because all you need to know here is that comparison is often the basis for someone using numbers to trick you.

What they are trying to do is short-circuit your thinking processes by presenting something outrageous, or controversial (or sometimes, seemingly uncontroversial, like a generally-accepted truth) to give some superficial rationale for you taking the action they want you to take.

Comparisons are a good way to do this.

Stimulating your emotional thought process with the aid of a set of “objective numbers”, then presenting a simple solution on a place, (ie the action the person presenting those numbers wanted you to take all along) is a surprisingly effective way for the unscrupulous to get what they want.

When I worked in higher education, league tables were a great way of doing this. (League tables, by the way, are just about the worst way to manage anything, which is why politicians like them – it gives a superficial illusion of oversight, while actually lets politicians abdicate their responsibilities to make the country better.)

The pitch usually went something like this (deliberately choosing a ridiculous example to protect the guilty): “Institution A is 10 places higher in the league table than us. They recently painted all the walls in their student canteens purple. So we should repaint our canteens purple and we’ll climb up the league tables”.

I won’t distress you with the number of times a group of otherwise intelligent people were moved to action on the back of a pitch like that.

So every time someone starts their pitch with a comparison, be very, very suspicious. The odds are high they are trying to trick you into agreeing with a position they have already decided beforehand.

Despite appearing to consult with you, the last thing they want is for you to ask any awkward questions.

Which is precisely why you should.

Three recent examples

As it happens, I’ve seen three great examples of this phenomenon recently, which attracted quite a degree of press coverage and online attention.

The key thing to remember here is that all the numbers presented are factually accurate, insofar as they go. They are the truth, they are just not the whole truth.

I find it’s rare, albeit not unheard of, for people to completely make up a set of numbers. Although it happens.

More often, they don’t understand the context for those numbers well enough, which is why so many proposals don’t stand up to detailed scrutiny.

Or alternatively, they understand the context perfectly, but it suits their case to ignore the wider context, so they just cherry-pick the numbers which best support the case they wanted to make all along.

I’m not sure whether incompetence or malevolence is a more attractive trait, but I have very little time for either of them. Which might explain my dim views of politicians…

Anyway, here are three stories which have consumed countless column inches over the past few weeks to illustrate how the process works, and what to look out for:

1-Pensions

If you live in the UK, you’ll be aware that we have something called the “triple lock” on old age pensions. Introduced by a Conservative-led government in 2011, the intention is to ensure dignity in retirement by increasing state pensions each year by the highest of the rate of CPI inflation, the rate of earnings growth, or 2.5%.

And lately, you can’t help but notice the number of commentators who say that the 4.8% increase pensioners will be getting in 2026 is unaffordable. It is, for example, “ruinously expensive” according to one Daily Telegraph commentator.

Yet the UK still has one of the least generous state pensions in the developed world – which should give us all pause for thought about how bad things must have been before 15 years or so of triple-lock.

A report for the House of Commons (linked below) highlights that 20-odd, developed nations have a lower percentage of pensions living in poverty than the UK. The same report also highlights that the UK spends a much smaller percentage of GDP on pensions than 27 other advanced nations.

I’m not a pensions expert, so I don’t know the right answer. But I do know that stoking outrage about increases in state pension is a great way to dodge the question about how badly-off the average UK pensioner actually is.

If the triple-lock was an attempt to do something about that, albeit is very slowly and over 15 years or more, then surely trying to alleviate the number of pensioners in poverty is the bit we should focus on, not the annual rate of pension increase announced for 2026.

(Full disclosure: I have major issues with how “poverty” is defined in this government statistic, but that’s not to detract from the point made above – it’s the comparison designed to stoke outrage I’m encouraging you to be alert for.)

2-Taxes paid by the rich

Some right-wing commentators are claiming that the top 0.1% of taxpayers pay more tax than the bottom 50% of taxpayers, generally as a justification for not taxing the highly-paid any more than they already are.

Now, I don’t have access to all the information I would ideally like to unpick this, but I’m prepared to believe that the numbers quoted above are mathematically correct. It’s the conclusion we need to challenge.

There are some listings around, such as The Times’ Tax List, which takes the methodologically questionable basis of adding things like the corporation tax paid by their companies into an estimate of how much total tax the super-rich pay.

I say “methodologically questionable” because if those super-rich left the country, but kept their companies operating, they would pay just as much corporation tax as they do now.

But looking at HMRC’s own stats on income tax specifically, we get a slightly different picture.

Firstly, their stats on taxpayers show that the income cut-off point for the bottom 1% of taxpayers is an annual salary of around £12,900.

Now, given that the annual personal allowance (the amount us Brits get tax-free before having to pay tax) is currently £12,570, the average person in the bottom 1% of taxpayers will be paying tiny amounts of tax over the course of the year.

The top 1%, on the other hand, which starts at a salary of £219,000 but clearly moves up into the stratosphere from there, will clearly pay vastly more tax. Assuming they don’t pay into a pension for the sake of simplicity, someone on a salary of £219,000 will pay £79,082 a year, according to MoneySavingExpert’s online calculator.

Someone in the bottom 1% would, on the same basis, pay about £64 a year, assuming they all earned £12,900.

Put another way, it takes about 1,235 people in the bottom 1% to pay the same amount of income tax as one person earning £219,000 a year.

When you understand that context, it’s much less of a surprise that a small group of very wealthy people at the top of the income scale pay so much tax that, collectively, they pay more tax than a large number of relatively poor people at the bottom end of the income scale.

Frankly I’d be more surprised if that wasn’t the case, given that the tax take starts at £64 a year in the bottom 1% and shoots up to almost £80k in the top 1%.

Now, that’s an entirely separate argument from the question about whether we are over- or under-taxing the super-rich (or, more accurately, the people paying income tax, which isn’t quite the same thing).

Someone on a gross salary of £219,000 a year is paying about 36% of their income in tax at MoneySavingExpert’s estimate of £79,082.

You might think that’s too low, too high, or about right. I’m not expressing a view here.

And you might think that someone earning £12,900 and paying £64 in tax (about 0.5% of their income) is being taxed too low, too high, or about right.

My problem with the argument as it’s usually stated is that the fact that wealthy taxpayers pay a lot more than poorer taxpayers is exactly what I’d expect to see.

It’s not necessarily an argument for or against increasing tax on high earners – although the people writing opinion pieces would like to you uncritically accept their view that “people at the top of the income spectrum are already doing more than their fair share”, based on the way they present the data.

That said, according to OECD comparisons, the UK is generally at the lower end of tax rates across major OECD economies for someone on average earnings.

My purpose here is not to suggest one solution or another. Just to say that the “right” rate of tax is a lot more complicated than tweets and soundbites from interested parties would have you believe.

3-AI adoption

There’s only one group of people I trust less than politicians, and that’s people selling AI solutions.

Recently this chart has been doing the rounds, showing the speed of AI adoption relative to the speed of internet adoption.

This is often used to promote the idea that AI is hugely successful and we should integrate it into our everyday lives at every opportunity.

Of course, this is precisely what AI folk would like you to do, as it’s in their direct financial interests. It’s no great surprise that OpenAI, a leading AI company, is quoted as a source for this chart. It’s hard to imagine a group of people with a greater interest in getting you to think AI adoption is inevitable.

But in this case, while it might not be the thief of joy, comparison is at least misleading.

Think back. In the first couple of years of the internet it was just three people scattered around the world. Internet adoption in the 1990s was slow because you couldn’t do anything with it – you needed to buy a computer, hook up a modem, find a telecoms provider who, if you were lucky, could install an ISDN line for you.

There were no games, no emails, and no social media. Of course adoption was slow.

By the time AI came along, you could download an app to your phone in 30 seconds, so of course more people “adopted” it.

And AI started being baked into all sorts of products (earning their makers my enduring hatred, as a rule), so of course more people “adopted it”. They didn’t have a choice.

Thanks, if that’s the right word, to AI, ordinary people, with no training, can do things more or less out-of-the-box with AI, like make videos of their dead cat from an old photograph, whereas in the early days of the internet you needed a degree in computer science to send an email.

The fact that most of the things you can do with AI are pretty pointless, and generally offer little-to-no real-world RoI for businesses is a separate issue.

But the speed of AI adoption is only possible because the internet was invented before AI was. If the internet hadn’t been invented first, AI would still be something three people did in a lab somewhere. It’s “rapid growth” is no more of a surprise than the fact that Angry Birds or Wordle went from being nowhere to being everywhere in a short period of time.

The speed of AI adoption is not collateral in the arguments from AI folk that their technology represents the future.

But that’s what the people running AI companies would like you to believe. So they are more than happy to try their best to mislead us by promoting charts and graphs like the one above.

I’m sure the numbers are true, as far as they go. But think about the context for longer than a second or two and you’ll realise AI companies are trying to manipulate you by making an unfair comparison.

Check out the other side

It’s easy to get swept along by someone with a bit of enthusiasm, backed up by what appears to be “objective numbers”.

Every time someone turns up with a heady blend of enthusiasm and mathematics, your initial realisation should probably be that someone is trying to bounce you into making a decision you’ll probably come to regret, but which is very much in their own self-interest.

So, stop a moment, and take the advice of my old law professor.

For everyone presenting a case, there’s always at least one other side to it.

Take a few moments to think those options through before saying “yes”.

Every decision you make will be better for checking out the “other side” of a proposal, even if you end up taking the action proposed to you with that blend of enthusiasm and maths, because you’ll have weighed up the options.

And, every now and again, this process will either save you from disaster or open up a bigger opportunity that had been missed by the person making the proposal in the first place.

For those reasons and more, always check out the other side. You never know what’s there until you take a look.

There are always at least two sides to everything.

When it comes to managing the bottom line, very few people are the real deal. Mostly it’s people talking about slashing costs or “cutting out the nice to have’s”.

With rare exceptions, those people are idiots. They’re not the real deal – they just talk tough and hope they don’t get caught out.

A few of the more delusional ones actually think they know what they’re doing, even though every time they unleash another bout of their strategic genius on a business, things get worse instead of better.

Some people never learn.

In the interests of full disclosure, I’ve seen people from just about every profession do this. People who have learned to talk the talk without ever learning how to walk the walk.

But talk the talk well enough – whilst making sure to move to another company before the somewhat awkward chickens come home to roost, of course – and this can be the foundation for a lucrative career for the hard of thinking and/or people who are too cynical to care.

Some days, I wish I was enough of an idiot and/or enough of a sociopath to join them. It certainly sounds like a much easier life than working on your craft for long enough to get good at it.

At least the vast majority of the time – nobody who’s any good has a perfect batting average. Occasionally, the best laid plans of mice and men don’t quite work out the way you hoped. And sometimes life has to teach you a lesson to keep you humble. That happens.

But anyone with an “infallible recipe” or a “standardised process” who is trying to manage your bottom line for you almost certainly doesn’t understand what they’re doing well enough to be allowed out without a responsible adult present.

Tech people and engineers are particularly susceptible to this sort of thinking, I’ve found, because they are used to working in controlled environments where they can manage all the variables.

Not good ones, obviously – really good tech people and engineers can get beyond that very restricted thinking. But an amazing number of third-rate ones think they know how to manage a bottom line with their box of magic tricks. And yet they get it completely wrong, because they don’t really understand what they’re doing.

Nobody is immune

Before tech people and engineers get too defensive, this is not something only they do – I’ve just found it’s more common there than in most other places.

Accountants suffer from this too. I’ll happily acknowledge that very few people can destroy an otherwise successful business more comprehensively than an accountant who doesn’t know what they’re doing running rampant through the business, cutting costs wherever they find them.

But today, I’m not talking about those professions – albeit I think an accountant or two might well have been involved in the story below somewhere along the way.

Today I’m talking about marketers.

Normally, I think marketers get an unfairly hard time. They perform one of the few roles in most businesses where they need to reach out to the outside world, with no guardrails or safety net, and try to do something new to have a positive impact with customers of the business.

Whilst that doesn’t always work, of course, at the point they press “send”, nobody knows for sure one way or the other, no matter how much testing has been done with sample groups in advance.

It could be a great success. It could be a giant flop. Nobody knows for sure until the campaign runs.

But somehow, marketers have to develop a constant stream of new ideas, and put themselves out there every day, hoping they get more clicks and likes than they did the day before, or that their brand-scoring metrics at the end of the month will show an uptick based on whatever campaign they launched at the start of the month.

That’s hard. Really hard. So, I usually have a big dose of respect for the people who even try. It’s a tough gig.

The big picture

One thing marketers are almost invariably good at is keeping the focus on the big picture. The mission. The key attributes. The ICP. The brand values.

But occasionally, marketers lose the plot too.

Sometimes, they’re just having a momentary episode of some sort.

Other times, they start behaving like those engineers and tech people and end up making a whole host of short-sighted decisions which end up damaging the bottom line.

In some organisations, this is quite the badge of honour internally. At least until they get found out.

It can be a big career boost if the CFO gets wind of the fact that someone in the marketing department is taking an axe to the cost base and (so they think) managing the bottom line more purposefully than it was managed before.

In those environments, frankly the CFO is every bit as much to blame as the marketer. It’s a sure sign that, not only doesn’t the marketer understand their business, but their CFO doesn’t understand it either.

That’s unlikely to lead to a positive outcome.

The holidays are coming

This dynamic has played out recently in the Coca-Cola commercial for Christmas 2025.

There has been a lot of online commentary about this ad, with the word “backlash” making a regular appearance in articles written about it, because Coca-Cola has decided to make this year’s Christmas ad using AI.

My longstanding position as an AI sceptic notwithstanding, it’s a great illustration of how some unholy alliance of tech people, marketers, and accountants can spend $millions making everything worse, while completely missing the point in the process.

Of course, the tech folk and marketing people involved are talking up what hard work this all was, and what creative geniuses they are.

Which is nonsense, of course.

While it would be unfair to call the ad terrible – there are plenty of other, equally bad ads around – the finished product is neither creative nor genius.

But it took 100 people to refine 70,000 video clips, according to the hagiographic PR put out by those involved, in a vain attempt to make us think this was a good piece of work. The truth is, it’s an extremely average digital re-hashing of themes which had already been developed in past generations of Christmas Coca-Cola commercials over several decades.

I’m sure there were some tricky technical issues to overcome in the process, but whatever they were, when it comes to managing Coke’s bottom line, this commercial achieved just about the worst possible outcomes.

Seeing the wood for the trees

You see, creating an ad isn’t like negotiating a cheaper price for printer paper, where all the paper you might buy has identical technical characteristics, and all you’re doing is trying to buy it for the smallest amount of money possible.

When you’re buying printer paper, that’s a perfectly decent strategy for managing your bottom line.

Running an ad, however, is different.

For a start, none of your customers care where you buy your printer paper from. You can buy it from Mars, for all they care.

With an ad, though, whilst you might want to push the envelope and all that, the first thing you have to make sure of is that you don’t completely cheese off the customers you already have now.

In that context, given the number of times the word “backlash” appears in reports and commentaries about Coke’s Christmas ad this year, I’d say this campaign fell at the first hurdle.

If Coca-Cola is your favourite soft drink, you’re probably not going to stop buying it tomorrow. Equally if almost any brand irritates you enough, for long enough, you’ll switch.

For example, as a Windows user for 30-odd years, I’m now in the position where Microsoft are cheesing me off so much with the AI nonsense they’re packing into all their products, that I’m seriously thinking of switching my allegiance to a MacBook.

Rule number one of advertising: don’t cheese off all your customers.

Outcomes, not inputs

The second rule of advertising is that what matters are the results you get, not the inputs that go into the process.

If I could sell $100million of products on the back of a $2million commercial, but you wanted me to cut my commercial budget to $1million, on the back of which my ad only shifts $50million of product, who’s the winner here?

Not the customers, not the company, not the brand.

And, in the long run, not even the CFO or the investors either. Provided they keep their eyes on the prize.

See, there are very few circumstances, with the possible exception of impending bankruptcy restricting access to cash, when even the CFO and investors wouldn’t prefer to generate $98million in net revenue, after ad costs, than $49million.

The cost of the ad here is secondary to the additional revenue it brings into the business.

And if one of the outcomes from your ad is that a significant number of your customers hate what you’ve done to your once-iconic Christmas commercials so much, to the point where some of them are organising consumer boycotts of all the brands you own, that’s not a good outcome.

No matter how much you’ve saved in production costs.

The Real Thing

You’ve also got to keep the brand in mind at all times.

For a business like Coca-Cola, their brand is incredibly valuable to them. After all, the cost of mixing water with some sugary syrup is pennies a bottle.

At the end of Q3 2025, Coca-Cola had a market capitalisation of $285billion, despite only having a balance sheet a third that size.

While there are some technical reasons for that difference I won’t go into here, the fact that Coca-Cola is worth 3x as much on the stock market as the value of its assets is based on – in the widest possible sense – the value of its brand.

Remember, there are no subscription revenues here. Nobody is locked into buying Coke for the next 5 years. There’s no recurring revenue. No certainty as to the income stream.

Every time someone wanders into a shop, they decide whether to buy Coke or some other brand of fizzy, sugary water.

And ironically, for a lot of people of a certain age, one of Coke’s more memorable tag lines over the years is “The Real Thing”. Coke put a lot of time, effort, and money into convincing consumers that some other bland, average, sweet, fizzy water that looked a bit like Coca-Cola was a third-rate fake of “the real thing” – ie Coca-Cola.

Against the backdrop of using “The Real Thing” to promote and protect your brand, using AI to create your ads – and being publicly self-congratulatory about it, to make things worse – essentially says that “the real thing” doesn’t matter any more.

If any sort of fake garbage someone throws together in an AI studio can represent the real thing in the world of Coca-Cola’s advertising, so can the bootleg Coke I brew up in my bathtub and sell by the case-load off the back of a van in the dodgier areas of town.

That’s every bit as real as the advertising Coca-Cola puts out.

I might be in a small minority here, but if I was Coke’s CFO or CEO, the last thing I’d be wanting to do with my valuable intellectual property and brand positioning in my customers’ eyes as “the real thing” would be to encourage anyone else to think that large-scale fakery was OK.

As a way to damage my own brand in real time, there are not many better ways of doing that, as the fake soft-drink syrup taking the enamel off the inside of my bathtub at the moment can corroborate.

And it’s all so pointless

Industrial-scale stupidity like this is bad enough.

What makes it worse is that this was all so pointless.

Nobody buys Coca-Cola because they use AI in the commercials. That’s not going to bring in a single new customer, and risks cheesing off a significant number of existing customers.

The whole point about Coca-Cola’s holiday commercials is that they speak to a tradition. They’re supposed to bring back warm childhood memories. They’re not trying to sell you the latest version of a video game. All this tech-enabled advertising nonsense is an irrelevance.

Actually, more than an irrelevance. A pointless irrelevance.

Had I been Coke’s CFO or CEO (open to offers, folks) I’ve got a really easy way to cut costs, flatter the bottom line, and not risk cheesing off a significant number of my customers, while still evoking every bit of the warm childhood traditional feel that so many people look forward to in the Coca-Cola ads.

I just run “a classic ad from the archives” each year.

Sure I still have the same media costs. A 30-second commercial costs the same whatever imagery I put on there – even if it’s imagery from three or four years ago.

But my production costs each year are pretty much zero (and, possibly, actually zero).

It wouldn’t even take a huge amount of work to spin all this as speaking to the traditions of yesteryear, which are never far below the surface at Christmastime.

So, I’d get all the benefits of reminding people of my brand, I’d keep my position front-and-centre as one of the western world’s major Christmas holiday traditions, and I don’t have my customers organising boycotts of all the brands I own.

At the same time as I’m doing all that, instead of spending $millions on a new campaign (whether AI generated or not), all my savings from re-running “one from the archives” flow straight down to the bottom line.

And not only does nobody care that I’m running “one from the archives”, people actually look forward to it, if I do this the right way, and I enhance my brand at a minimal cost.

The bottom line lesson

In the words of the great Peter Drucker, there is nothing so pointless as doing something efficiently that doesn’t need doing at all.

To run this season’s AI-generated ad, Coca-Cola might well have thought they were pushing the envelope and developing some cutting-edge brand collateral, while boosting their bottom line.

Sadly, they have probably achieved none of that. And they managed to cheese off a significant number of their customers in the process.

Not what I’d call a victory – either for that business, or for common sense more generally.

This happens to be a particularly pertinent example, but the purpose of this article is to make this point.

When you’re managing your bottom line well, you’re not just managing your cost base.

You have to manage a much more complex series of variables, including the impact on your customers, whether you increase or reduce your average customer lifespan, whether your decisions will adversely impact your productive capacity, how motived your staff will be, and a range of other, often subjective and hard-to-pin-down, variables.

But the alternative is that you spend $millions to end up with your customers organising boycotts of your products.

However smart you think you’re being, that’s particularly dumb.

If you employ anyone in your finance, technology or marketing teams who think otherwise, I’d recommend firing them now.

No matter how good they are at talking the talk on a technical level, if you let those folks walk the walk, they’ll walk all over everything good about your business, and you’ll end up regretting it.

I promise. That’s the real thing.

A long time, when the Industrial Revolution was getting under way, just having some new invention was reason enough for people with enough cash in the bank to try it out.

“So you’ve invented a steam-powered horseless carriage, have you? I must try one of those out!” was probably something members of the landed gentry said back in the late 1800s.

Of course, some of them were dead against giving up their actual horses, as those had been a mainstay of the lives of lords and ladies of the manor for hundreds of years up till that point.

But a few of the younger, or crazier, members of the landed gentry with more money than sense were prepared to give it a try. And out of that, the automobile industry was born.

For the first couple of decades, automobiles were for the rich. They were hand-built from the ground up, and required the sort of skills that were hard to come by in the middle of the Industrial Revolution.

Then Henry Ford came along and invented mass production.

Now cars were affordable for just about everyone – including the people who worked in Henry Ford’s own factories. They were many more vehicles on the streets and, over the next few decades, automobiles became ubiquitous.

Now we’ve got more of them than we know what to do with, and many cities are clogged up pretty much 24/7 with motor vehicles of one sort or another.

What started out as a niche pastime for the idle rich became something that almost everyone owns.

The early years

In the early years, automobile manufacturers focused on refining their vehicles, and making them work better.

We went from steam-powered monstrosities through to the internal combustion engine fairly rapidly and, by the time the first Model Ts were rolling off Henry Ford’s assembly lines, the internal combustion-powered motor vehicle existed in pretty much the same format we use it today.

Sure, it wasn’t nearly as aerodynamic as a modern vehicle, nor could you plug in your phone to fire up Google Maps, but no adult today could look at a Model T Ford and be confused about whether or not it was a car…albeit a prehistoric one.

But the key thing is that a large part of the development of the automotive industry took place when there were less than 10,000 vehicles on the roads in the United States.

While there were, of course, engineering and technical challenges in the process, it’s a lot easier tinkering with one car at a time in a specialist workshop full of skilled mechanics than it is to try to fix Model T Fords when they’re coming off the assembly line at the rate of one every couple of minutes.

You just can’t fix things fast enough to put right a fundamentally broken production line process at the end of the process, just before despatch to your customers. You have to fix everything before you crank up the production line, not after.

But once you’ve got a template that works, you can roll out thousands of near-identical models of a Ford Model T without too much difficulty.

Smart people know you fix the system first, then crank up the volume. Not the other way round.

Features first, benefits second

Another feature of the early years of the automotive sector was that, in the beginning, just having a “horseless carriage” to sell was exciting enough – even if it wasn’t terribly practical and broke down a lot.

There was an infinitesimally small number of incredibly wealthy people who could afford one and they acquired significant personal pleasure and, I’m sure, status amongst their peers, just by owning a horseless carriage.

Even if it required coal to be shovelled into a boiler on a regular basis, and had a top speed of only 4mph.

The point here wasn’t that it was a more practical proposition than a horse. Because, in the early days, it wasn’t.

Rather, the attraction was that horseless carriages were new and novel. The sort of things only rich, sophisticated people who fancied themselves at the cutting edge of science and technology would consider buying.

When hand-built motor vehicles cost 5x an average annual salary, only the super-wealthy could afford one.

So, in the early days, just putting a sign outside your workshop saying “Horseless Carriages For Sale” would bring in the customers.

99.999% of the people passing your workshop knew they couldn’t afford one.

And the tiny number of very wealthy people in your neck of the woods would go to the only place around where horseless carriages were sold, if they wanted to buy one.

So selling wasn’t all that hard.

And, at a time when the motor vehicle industry was predominantly comprised of locally-based artisan workshops, realistically very few people were going to make the trip from, say, Los Angeles to Cincinnati to check out the new model of motor vehicle being built up there. Not least because there the car that could get you there hadn’t been invented yet, and the interstate highway network hadn’t been built.

So, in those days, cars were sold mostly on their features. “Look! We have a car for sale. And it’s got a front windscreen and a spare wheel.” (Or a roof for the passengers. Or a handbrake. Or something else that sounded cool in the early years of the 20th century.)

All of those were novelties to a group of people who had previously ridden horses everywhere, because there was no other way to get around.

When the novelty value is high, and you’re appealing only to a tiny group of very wealthy people, this features-led approach will work perfectly well.

It’s largely how luxury goods are still sold today.

Hermes scarves, Louis Vuitton handbags, and Dolce and Gabbana dresses are sold on their features (high-grade silk from artisan Cambodian silk worm farms located high in the mountains to ensure the very smoothest weave, for example – which I’ve totally made up, by the way).

And possessing one of those scarves, handbags, or dresses is enough to cement your status as one of the cognoscenti.

You can imagine Patsy and Eddie gushing about the mountain-top Cambodian silkworm farms on Absolutely Fabulous.

All they want is something to gush about that sets them apart from the hoi-polloi.

After that, they’re not too bothered.

And the product doesn’t even need to be particularly functional. I’ve no idea what the insulating qualities of a Hermes scarf are, because I’ve never owned one.

But I do know that the people who buy a Hermes scarf are not buying them because they’re planning to climb Everest and they’ve heard what great insulation your neck gets from a Hermes scarf.

People buy Hermes scarves because they are Hermes scarves and very few people can afford one – the Cambodian silk farm story, or whatever they say about their products, is mostly incidental but it helps to sell luxury goods, so who am I to complain?

Going mass market

For as long as your target market is rich people with time on their hands and money in the bank, you can get away with doing very little actual selling. The features alone are usually enough.

People come to you.

And, on the flimsiest of precepts – which may, or may not, include stories about mountain-top silk farms – they’ll give you $1000s to buy a product not that different in a functional sense from something you can buy for $10 at a mass-market outlet.

When you go mass market, however, features alone aren’t enough to sell.

You need to major on the benefits. Both benefits compared to other people who sell similar products or services to the ones you sell. And, ideally, benefits compared to doing nothing at all.

Nowadays, this is not usually done too overtly, although back in the early days of mail order retail, for example, the benefits were usually rammed down your throat whether you wanted them or not.

Let’s take toothpaste as our example.

In my local supermarket there is regular toothpaste. Extra-whitening toothpaste. Gum health toothpaste. Fresh breath toothpaste. Stain removal toothpaste. Toothpaste where the effects last more than 24 hours. Toothpaste 8 out of 10 dentists recommend. And dozens of other claims of one sort or another.

Now, I suspect 95% or more of any toothpaste is much the same as any other toothpaste, but there might be small percentages of active ingredients which are different between, for example, tooth-whitening brands and gum health brands.

My point here is that, for a mass market product, you have to sell the benefits, because finding toothpaste on a supermarket shelf is not difficult. My local supermarket has probably 20 brands or more vying for each shopper’s attention.

But depending on which solution I wanted – better gum health or whiter teeth, say – I now have only two or three possible options and I’ll pick one based on whether I want the cheapest, the most long-lasting effects, the freshest breath or whatever other criteria I might consider important.

Where it goes wrong

Where things often go wrong is when a mass market product (or one with aspirations to become a mass market product) is sold on features, not benefits.

Not many people would wander down the aisle of my local Sainsbury’s and respond positively to a big sign saying “Look! A tube of toothpaste!”

We all know what a toothpaste is and what it does, at least conceptually. What we need is information about why we should by Brand A or Brand B based on what criteria we have for selecting a toothpaste, be it gum health or whiter teeth.

Trying to sell a mass-market product, which requires strong benefits, as if it was a luxury good, when you can just witter on for a while about features like mountain-top silkworm farms for a while, is generally a mistake.

A great place to see that at the moment is the pitches for AI products.

I mean, it’s great that you can use AI turn an old photo of the cat you had as a teenager into an animated video of Snuggles walking around your back garden today. I’m sure that’s an impressive piece of technical work, much like inventing the internal combustion engine or pneumatic tyres was for the motor industry.

And if you’re selling to exclusively wealthy people with time on their hands and plenty of cash, knock yourself out. That’s probably enough. All they want is the exclusivity of being able to do something the hoi-polloi can’t do.

When you’re trying to sell a product to the entire planet, though, that doesn’t work any more.

Firstly, not that many people would want to create a video of a dead cat anyway. And very few of those who do are going to spend $1000s to see Snuggles walk around their garden again.

So, if you want your mass-market product to be commercially successful, you need some benefits.

And frankly, however interesting and clever the underlying technology is, there are virtually no benefits to reanimating Snuggles to the vast majority of the population.

All things being equal, therefore, they won’t.

Sure, people might mess around with it for a while, for fun, if it’s free or very cheap.

But the modern consumer has more than enough things to waste their time on already. Beyond the initial endorphin boost of seeing Snuggles again for the first time, how many videos of their dead cat does any sane person want to make?

Not many.

So, without a meaningful benefit, people stop using your AI video service and move on to a different, more beneficial (to their way of thinking – not necessarily yours) way of wasting their time.

I saw a pitch for an “AI-powered” accounting system recently, which very excitedly told me all of its features. The thing the tech folk who developed it didn’t seem to have understood was that their whizzy new technology didn’t give me anything I couldn’t already get from my £20 per month Xero subscription.

There was no discernible benefit to me at all, with the exception of being able to boast that I was one of the cool kids using AI.

You could sell horseless carriages in the late 1800s like that, when your audience was a very niche market comprising exclusively extremely wealthy people, on novelty value alone.

You can’t sell an aspiring mass-market accounting product in 2025 with a feature (“Look! AI!”) which, in the use-case of every accountant in the land, delivers no significant tangible benefits beyond what we can already get from Xero for £20pm.

It got worse

When I gently pushed back against this less-than-compelling pitch, it got worse.

I was told that it would allow me to run 100s of reports which would help me “stay on top of my business”. Doh! What do they think Xero does now?

And also that it would allow me to manage risk better, even though none of the features they claimed to deliver would allow anyone to manage risk better, AI-powered or not.

This exchange just demonstrated how little the people selling this product knew about accounting software. Which, if I’m honest, did not increase the chances of me wanting to buy what they had to sell – if a software company doesn’t understand their own target market, and hadn’t done even a rudimentary competitor analysis, they don’t deserve to make any sales.

Outside accounting, this is also a characteristic many of the “It’s X…but with added AI!” pitches I see at the moment.

AI is generally delivering little or no tangible benefit over and above what existing software products are doing perfectly well already.

So AI companies are piling in more and more features in the hope that enough of the planet will go “hey, that’s cool” and shell out $000s a year to use that product or service.

But ultimately the AI companies are wasting their time. Particularly in business, where nobody is going to invest in anything without a cast-iron business case. (The B2C market is slightly different, albeit significantly less lucrative.)

And for a cast-iron business case you need benefits, not features. Yet, there are remarkably few of those around.

Which, I suspect, is why AI companies are struggling to generate a viable revenue stream.

The truth is that, beyond an initial “hey, that’s cool” reaction, like the cat videos, nobody cares.

And certainly nobody cares enough to shell out the enterprise-level $millions a year those businesses need to stay afloat when their seed capital runs out.

Frankly, I’m suspicious

This is all so obvious that I can only form one view of the AI industry.

And it’s this – even they are acknowledging, internally if not externally, that their products offer little or no real-world benefits.

And I can be pretty sure of that because if there were any benefits, they’d be shouting them from the rooftops by now and we’d all be handing over $000s a year for the privilege of using their services.

It’s such an obvious route to monetisation at a level beyond every venture capitalist’s dreams that the fact they’re not doing it can only mean there aren’t any real-world benefits to the technology at all. Which means, when the investor cash runs out, they don’t have a viable business.

Why would any sane person take the much harder route of finding hundreds of new and novel features in the hope that at least one of them captures the public’s imagination, if a really simple route akin to “buy this toothpaste for whiter teeth” was available without breaching all the “truth in advertising” legislation.

Actually, I’m being slightly unfair there. The only benefit AI companies talk about regularly is “we can do what you do now, just a bit cheaper”.

Frankly, that’s just about the laziest benefit there is.

And it’s a particularly stupid “benefit” because, however cheap Company A does something, I can guarantee there’s a Company B out there somewhere who can do it a little bit cheaper than Company A does.

Before you know where you are, everyone is playing off all those companies off against one another and they’re all supplying incremental customers at a price below their marginal cost of production.

I say that with confidence because that’s what happens in every sector which can only deliver a price-based benefit. The industry competes itself down to margins of pretty much zero…which isn’t the world’s most long-lasting strategy for running a successful business. Especially if you want a business worth $billions.

And that’s even assuming AI companies deliver tangible cost savings at all, which I haven’t seen yet.

There’s been a recent MIT report saying that 95% of AI implementations haven’t delivered the business benefits originally claimed, for example.

And IBM have reportedly had to hire back most of the people they let go on the basis that AI had made their jobs redundant, because AI could not, in fact, replace those human beings on a like-for-like basis. If a massive tech company could be hoodwinked like that by the false promises of AI technology, there’s probably no hope for the rest of us.

So, my working assumption – which I’m happy to revise in the face of evidence to the contrary – is that AI delivers no significant business benefits. If it did, we’d be hearing about them every day of the week and we’d all be happy to pay AI companies $000s a year just for the privilege of using their products.

Sell the benefits, not the features

If, fundamentally, your product doesn’t work, and you’re about as ethically-challenged as the average AI company, by all means sell the features of your product, because that’s probably all you’ve got.

It’s like structuring your entire sales process around a “Look! A Squirrel!” strategy.

But if you deliver a real-world benefit you’ll scale faster and become more profitable if you sell those benefits.

(Unless the “benefit” is “we do X, but cheaper” in which case you and your competition will drive your margins down to almost nothing because that’s what happens with commodity products.)

Selling the benefits is much easier because who wouldn’t want a tangible bottom-line boost to their business, provided they believe that the product you’re selling will do exactly that on a sustainable basis.

As you deliver more and more products, and more and more people appreciate the tangible benefits your business brings, word gets round and it becomes easier and easier to make the next sale…and the next…and the next…

If AI companies actually delivered real bottom line benefits, we’d be hearing all about it.

And we’re not.

Which is why I don’t think they deliver any real-world benefit at all.

But, when thinking about real-world benefits, you have to do something that tech folk are very poor at (in fairness, so are many engineers and accountants, in my experience) which is to look at what you think a benefit is through your customer’s eyes to make sure it isn’t just a feature really.

This requires you to realise that you don’t know everything about the world…and to accept that there is a real world outside your lab/workshop/finance department which might not conform to the neat and tidy rules you can implement inside the closed system that is your lab/workshop/finance department.

That’s where people – especially in tech and those other professions – go wrong.

They think just because they’ve come up with some cool idea that it’s automatically a benefit. But usually it isn’t. At best it’s a feature. Often, it’s an irrelevance.

For a business, being able to animate videos of dead cats, for example, is a feature, not a benefit. Or more likely an irrelevance, in a business setting.

So let’s look at a few real world examples to make a point – mostly non-tech based, just for balance, because other industries get this wrong too.

Some benefit examples (or not)

When you look at a problem through your customers’ eyes, you get a very different perspective from “look, I developed this cool thing in a lab”.

Here are three quick examples of something that might sound like a benefit to the people trying to shift products, but which, in fact, aren’t really benefits at all. If I was being kind, I’d say they were features. Some aren’t even that.

1-Cartonboard

Many years ago I worked in the printing industry. We made printed folding cartons (the brightly coloured boxes you see in the supermarket with breakfast cereal inside them, or tea bags, or a frozen dinner).

One of our suppliers wanted us to buy their super-duper carton board to make those products, which was well-engineered and super-smooth. It would, they claimed, allow us to run our presses faster and give us a better finished product with less wastage.

We tried it on a job alongside our standard carton board and the claims the manufacturer made were true. We could run our presses faster and we did have a better finished product with less wastage.

The problem was the price they wanted for their carton board meant that the best economic decision for our business was to run our presses a little slower and tolerate a slightly higher wastage in production.

The well-engineered, super-smooth product was a feature. It delivered no bottom line benefit to the business, so we never used it again.

2-Cars

The motor trade generates lots of examples of not taking a customer view of your selling messages.

For example, for some time I’ve been targeted by online ads telling me I should buy a new hybrid car because it delivers just over 50mpg.

That sounds like a benefit from the manufacturer’s perspective, I’m sure.

Except the car I have now does 41/42mpg. So they want me to spend £40-50k buying a new car “because it’s more fuel efficient” for about another 8 or 9pmg.

Factor in that, at the moment, I’m driving less than 3,000 miles a year and there is no economic benefit whatsoever to me spending £40-50k to get another 8 or 9mpg.

Part of me would like a swanky new car. But whatever else is going on here, that 50mpg car delivers no bottom line benefits to me.

Whereas the same pitch targeted at a high-mileage driver might well be seen as a good reason to spend upwards of £40k on a new car.

What might be a benefit to them isn’t a benefit to me. Not every customer is the same. Not every “benefit” is seen as a benefit to every customer.

The art of sales is knowing which is which.

3-Customer service

Many organisations parrot a line you’ll have heard before.

You find it on packaging sometimes, and also in the recorded message you get when you call a call centre, shortly after the recorded message saying, for the 8,000th day in a row, that the company is experiencing “unusually high call volumes” and “you might need to wait longer than usual to be served”.

Once they’ve softened you up with that, the next recorded message says something like: “for a faster service, please go to our website where we can serve you 24/7”.

While I would concede that my experience of most organisations’ call centres is variable (some are excellent, some are terrible), my experience of customer service delivered via a website is uniformly terrible.

First, we have the hunt for our customer number on the invoice that we don’t have any more, or we need to remember the email and password we used to log into the darned website originally, which can take some time.

Next, we get the dreaded automated chatbot which spends 10 minutes not understanding our problem, before directing us to the same FAQ pages we’d already checked out before calling the call centre.

Then we might get a human chat agent.

Finally, we might get to speak with a human.

And finally, finally, finally…if we’re really, really lucky…we might get our problem solved.

24/7 customer service sounds like a benefit, but it isn’t.

For two reasons.

Firstly, it’s invariably terrible.

Secondly this “benefit” has turned a 3 minute phone call into an hour-long extravaganza (a particular thanks to Norton anti-virus products here, who kindly gave me the inspiration for this article a couple of weeks ago).

From a customer point of view, making me spend 57 minutes more than I have to for the privilege of doing something simple is not…by even the most fevered stretch of imagination…a benefit to me.

It’s the exact opposite. It’s a perfect reason to shop somewhere else, quite frankly.

While 24/7 customer service sounds like a benefit, if it doesn’t actually solve your customer’s problem in a timely manner, it probably isn’t even a feature. It’s more like an irritation or a turn-off, if anything.

The moral of our tale

So, there are three morals to today’s tale.

Firstly, it is so much quicker and easier to sell when you focus on the benefits you offer, rather than the features of your product.

If you only sell the features, your market is limited to people who think the same as you on a topic – that videos of your dead cat brought back to life are cool, for example.

That’s likely to be 5% or less of the total potential market for your products, so this is a very short-sighted approach.

Secondly, don’t get confused between what a feature is and what a benefit is. Not everything you put into your product is a benefit to anyone, however cool you think it is.

Finally, a benefit is only a benefit if customers see it as such, otherwise it’s just a feature and isn’t going to help you make a sale – that £40k-plus car that was going to get me another 8mpg for example. Or 24/7 customer service that doesn’t actually serve any of your customers in a timely manner.

Fundamentally, if you’re not leading with a strong list of benefits, my working assumption is that there aren’t any benefits in buying from your business compared to buying from one of your competitors, so I’m likely to shop around and just buy the cheapest because you haven’t given me any good reasons to prefer your business over your competitors.

While there will always be some people why buy the most expensive product (the Hermes scarf) or buy something because it’s cool (animated videos of dead cats brought back to life) or buy on the back of some grifter’s sales spiel about changing the world (most AI products), if you have to depend on those techniques to make a sale, your business is not built on solid foundations.

So you might want to think again.

And if you can’t think of any genuine benefits from your customers’ perspective, you might want to think about doing something else before the cash runs out, because you don’t have a viable business.

Imagine you want to understand a subject you know currently nothing about.

You start off in complete ignorance. Knowing nothing whatsoever.

Then you pick up a book, and read a chapter or two. Or do a search for the topic online and read a blog post or two.

Or, heaven forbid, read one of those tiresome AI summaries which are popping up all over the place at the moment.

Congratulations. You have just had the highest RoI learning experience on that topic you’ll ever have.

You’ve gone from knowing zero, to knowing a tiny amount – a fraction of 1% of all the available knowledge in the world on that subject.

Starting from a base of zero, that’s an exponential return. So enjoy it – you’ll never get as high a percentage return on anything more you learn about that subject.

That’s because of the Law of Diminishing Returns.

To get your head around that concept, imagine someone else who has read 100 published books on a topic.

How much more are they going to learn about that topic from reading their 101st book?

Well, possibly not absolutely zero, although the chances are high that the 101st book is just a rehash, to a greater or lesser extent, of things already said in the other 100.

A little bit of maths

Let’s take this example a little further and imagine that the first person went from knowing zero to knowing 0.1% of the available information on that topic.

And the second person went from knowing 99% of the available information on that same topic and they now know 99.1% of the available information on it.

So they’ve both increased their knowledge base by 0.1%, haven’t they?

In absolute terms, that’s true.

But in relative terms, the second person has increased their knowledge by slightly over 0.1%, whereas the first person increased their knowledge by infinity % (don’t come at me maths geeks), which is a lot more than 0.1%.

Let’s make this a little clearer. And the maths a little easier. And get the maths geeks off my back.

Imagine they both increase their knowledge by a further 0.1% by reading another book.

Person A has gone from 0.1% to 0.2% – that’s a 100% improvement on their base of 0.1%. A doubling of their knowledge on a topic.

Person B has gone from 99.1% to 99.2% – that’s an improvement of a tiny fraction over 0.1%.

So there’s a curve somewhere between knowing 0.1% about a topic and knowing 99.1% about it that diminishes steadily the more you know – each incremental bit of effort you put into learning more pays back less and less the more you learn. In relative growth terms, at least.

What do you need to know?

Why is this important in business?

Well, unless you want to waste endless amounts of time and resources, you need to work out what you need to know to make a decision. And then make a decision.

In other words (well, actually in Michael Jackson’s words) you’ve got to stop when you get enough.

Any other outcome is seriously detrimental to your bottom line.

Yet, the mindset in most businesses is that you can never have too much information.

That’s dangerous. Really dangerous. To your bottom line.

I’ve seen organisations invest in six figure software products because it “gave them more information” and “allowed them to produce more detailed reports”.

I’ve seen other organisations invest in data analysts, at huge expense, to analyse all manner of data.

And nowadays, organisations often implement some sort of AI solution to “manage all their data and reporting needs”.

With rare exceptions, all of those are terrible decisions. That sort of thinking will destroy your bottom line faster than almost anything else you might be tempted to do – and that includes telling your biggest customer to get lost.

So what’s my point?

Here’s my point.

Up to a certain point, more data and more analytics produces more bottom-line enhancing insight than it costs. A bit like Person A above – they rapidly get better because they start out knowing so little.

Beyond that point, every further investment in people and/or technology brings more cost to the organisation than the thin layer of extra benefits would justify.

That’s like the second person above. It’s almost certainly worth precisely zero to your business to go from having 99.1% of the available information to knowing 99.2% of it.

Except in the unlikely event that extra 0.1% unlocks some transformational knowledge that you were previously completely unaware of – and the odds of that are worse than the odds of you winning a million on the lottery – any spend which gets you from the 99.1% level to the 99.2% level is almost certainly stealing money away from your bottom line, not adding to it.

Of course, no software salesperson is ever going to admit that. If they even recognise the phenomenon.

In the world of data and analytics sales, any outcome that gives you less than 100% of all the available information means there’s some potentially untapped opportunity for tech folk to sell their software.

I don’t particularly blame tech folk for that. It’s the way their brains are wired. And they’ve got software to sell. So good luck to them.

That doesn’t mean you have to swallow their sales pitch, though.

But it starts in your head

You see, it’s never been easier for data and analytics software companies to sell their software.

And that’s because most people have bought into the mindset that “you can never have enough data”. This is the insidious message that all sorts of people drip feed to you every day – the giant consulting firms, the tech folk with software to sell, the people convincing you to hire another data analyst “to understand your metrics better”…and thousands of other people.

Believing the message that the combined PR efforts of thousands of interested parties with a product to sell has tricked vast swathes of people into believing that anything less than 100% knowledge of all the data in the world affecting a particular topic is a dereliction of your duty as a business owner.

However, that’s completely untrue.

Rather, as a business owner, your mindset should be that every time you spend a pound, you should expect to get back more than a pound.

If you do that consistently, you’ll build a great business.

Now, sometimes you’ll invest a pound today and not get more than that pound back for a year or more – for example, if you buy a new machine for your factory. But the principle remains that you’re looking to spend money where you are most likely to get a multiple of your original investment back over a sensible period of time.

And also that, all things being equal, you should rank the RoI on everything you might potentially do so that you implement the highest RoI activities first and only get to the lower RoI activities after all the better uses of your money have been taken care of.

To give a tangible example, there is almost certainly a higher RoI on investing in a couple of days of induction training with a new member of staff so they can learn all about your business than it is to invest in a new data tracking system to take your understanding of the data from your company’s petty cash system from 99.1% of all the available data to 99.2%.

Taking your new member of staff from knowing zero about your business to knowing even just 10% or 20% of it is a phenomenal RoI for the cost of only a couple of days’ salary, in terms of how they can get up to speed faster, do their job more efficiently, and serve your customers better.

If you learn anything useful at all after getting that extra 0.1% of data out of your petty cash system, I’d be astonished.

Think incrementally

To counteract the mindset that tech companies’ PR systems have been borrowing deep inside your head for years, you have to think incrementally.

What I mean by that is thinking about what extra return, over and above what you have now, you get from making a new investment, and comparing that to the additional cost (both up-front and ongoing).

The people selling tech solutions will try very hard to convince you to buy their products, so it’s in their interest to muddy the waters around all this, in order to make it seem like a no-brainer to buy their product.

In particular, there are three aspects to look out for which often creep into tech pitches…often subtly, so they can be hard to spot:

1-Project RoI

A common tech pitch asks you to give them £X and get £Y back again.

Even if that’s true (and it usually isn’t, by the way), that’s not really the question you should ask yourself.

The question you need to ask is whether, out of the dozens or hundreds of potential things you could invest in for your business, that’s the number one candidate for investment in terms of incremental bottom line performance.

If it’s going to make more of a difference than anything else (which is unlikely), of course you should invest in that product.

More likely it’ll make a marginal-at-best impact, in which case you should invest your time and money on whatever your number one RoI project is instead of messing around the margins with comparative irrelevancies.

Saying “no” to pointless or marginal-at-best projects is one of the most powerful things you can do for your bottom line. Instead, focus your energy on projects that will make a transformational difference to your business – unless you work in tech, it’s extremely unlikely that’s going to be a tech product.

2-You’ll find something

I’ve been on the receiving end of a version of this pitch twice recently.

It goes something like this: “Look, currently you only track 40 datapoints in your business and we’ve identified another 1200 potential data points you currently know nothing about. You’re bound to find something in there that will make a huge difference to your business – you’re only scratching the surface at the moment.”

This is a slightly more disingenuous pitch than the one above, where there was at least a theoretical project RoI, even if it wasn’t the biggest potential RoI in your business. But it preys on the several decades of “you can never have enough data” messaging that’s lodged deep inside your brain.

I can save you a lot of time here.

If you’re already tracking 40 datapoints and managing them well, you’re probably at 99.1% of all the information you need to know about your business. And the extra 1200 datapoints will get that up to 99.2%.

Armed with that knowledge, how much time, money, and effort are you willing to spend on this proposed tech project?

Think very carefully if your answer is anything other than “zero”.

3-What’ll you do different?

Information alone is pretty much worthless.

The worth comes from what you would do differently, if you had the information, that you’re unable to do now.

And if you wouldn’t do anything different to what you’re already doing, more information is worthless. So spending even a penny on it is seriously detrimental to your bottom line. There is no upside to getting this information because it won’t make any difference to what you’re already doing.

A couple of years back, I had a client who wanted to invest in a tech solution to track all sorts of KPIs for the entire sales team where there were a couple of underperformers who, they weren’t generating enough leads to hit their sales targets.

So I asked, given that we already know what the problem is (lack of leads), what would we do differently on their lead generation if we knew that, for example, we could organise their in-person client visits more efficiently by better grouping together of in-person sales calls?

Of course, the answer was “nothing”. However interesting the software might have been if we had a “you don’t use your time efficiently when you’re out on sales calls” problem, knowing that information would not help us to increase their lead flow in the slightest.

Thankfully, that idea was shelved and we spent the time helping the salespeople generate more leads instead, which was the actual problem we were grappling with.

So be clear in your own mind that any proposed solution isn’t just a nice idea in and of itself. For it to have a bottom-line benefit to your business, you have to do something different to whatever you’re doing now to solve a current problem in your business.

If the proposed investment either isn’t going to do that (ie it solves an entirely different problem), or it might do something to help with your current problem, but there are other, better, higher RoI ways of solving that problem, then don’t invest.

You SHOULD stop when you get enough

Your ambition shouldn’t be a dispassionate “nothing less than 100% of all the data is acceptable” approach, no matter how much time and effort tech folk put into convincing you how important that is.

Rather, set out to find the difference between having enough data to set a course of action – up to and including the point at which the information benefits of more data or analysis are outweighed by the costs of producing them.

And when you find that point, stop spending money. Anything you spend from that point onwards has a negative impact on your bottom line.

Michael Jackson almost got it right.

He sang “don’t stop ’til you get enough”.

The bit he left out is that “but when you’ve got enough, for the sake of all that’s holy, stop!”

I know Behavioural Science is a bit of a buzzword at the moment (can you have a buzzword made out of two words…? I’m not clear on that, but that isn’t my point, so we’ll press on…).

The concept behind Behavioural Science, to horribly over-simplify a complex topic, is that understanding how humans behave in the real world can help businesses understand their needs better and thereby get them to do something the business would like them to do, whether as customers or employees.

In marketing and advertising, people like Rory Sutherland and Richard Shotton showcase how the way we talk about products and services can make it more or less likely that someone is going to buy the products a business is selling. The differences are often subtle, but with out-sized returns if you make the right choice.

My pal Christian Hunt has done a tremendous job implementing the concepts behind behavioural science into the seemingly unrelated field of compliance – but it turns out that skills of a marketer or advertiser are very similar to the skills an organisation needs to get people to follow the rules, especially in areas with significant regulatory oversight, like in financial services, for example.

Yet behavioural science is not a concept I hear people talking about in accounting all that often.

To be fair, that’s partly because accountants have a (generally well-deserved) reputation of not engaging with any airy-fairy hippy nonsense that you can’t capture on a spreadsheet.

But I think that’s a mistake because understanding human behaviour is a shortcut to much better, bottom-line-boosting business decisions.

And if your CFO isn’t wanting to make a lot more of those, frankly I don’t know what else they think they ought to be concerned about.

Traditional financial analysis

Somehow, there’s an expectation – both of accountants by non-accountants, and of accountants themselves – that to make any important decision you need a 90 day-long project to extract mountains of data which someone is going to produce enough charts and tables from to explain every possible outcome and variation.

Let’s be up-front about this.

That’s always an expensive process. However well-intentioned it might be, tying up senior people in weeks or months of meetings, analysing spreadsheets, charts and graphs until the numbers give up their secrets is horrifyingly expensive.

Sometimes people tell me that doesn’t matter because these people are paid a salary “so it doesn’t cost any more to get this work done”.

When people tell me that, I know they’re either not an accountant or, if they are an accountant, they’re a particularly third-rate one.

Internal projects are a huge time suck…and therefore cost fortunes, although generally nobody joins the dots to work that out.

And they often lead to very equivocal conclusions – “well maybe this might work, but equally it might not”.

Frankly, you don’t need half a dozen very well-paid people spending 90 days together to work that out. I could probably have come to that conclusion myself with 10 minutes thinking time and a calculator.

If you didn’t do the analysis project at all, and fired half your finance department because they were no longer needed, you’d make a guaranteed bottom line return on Day 1. And your business would probably be no worse off because the project’s conclusion was only ever going to be “meh” anyway.

So what do you do?

Now, at some point in the process, you do need to do some thorough analysis. I’m not suggesting for a moment that you commit $ millions to an idea someone had in a fever dream without checking it out thoroughly first.

However, my suggestion is that you don’t knee-jerk your way into an analytics project at the start.

I kick things off with two other tools long before any serious analysis takes place.

The first of these, I call “a rough cut”.

That really is me spending 10 minutes with a calculator and a bit of paper (I wasn’t joking about that).

Doing rough cuts is not the point of this article – we might do that another time. But, briefly, I’m trying to work out “what’s the prize?” in this stage. Put another way, is this project likely to be worth the time and effort required to explore it further?

You’d be amazed at the number of projects which have floated across my desk over the years which can’t even meet that hurdle rate.

In my home life, I once had a salesperson try to convince me to switch my utility provider because they could save me something like £2.73 a year compared to what I was paying now. There is almost no amount of work I’m prepared to put into something that costs less than a cup of coffee in Costa, and spending an hour on a phone with the utility company’s salesperson didn’t sound like the best use of time to me.

At work, I’ve often had people try to flog some system or other which requires me to spend £1million up-front in return for a £50k-£100k annual cost saving, after factoring in the cost of the system.

But pause for a moment and cost in the disruption, training, and general inefficiencies in implementing any new system…together with the fact that you can be pretty sure that anyone flogging a system to do anything will, at most, deliver half the benefits the salesperson claims…and that project is a breakeven project at best. Certainly not one I’m going to commit hundreds of hours of highly-paid people’s time to analyse in excruciating detail.

There just isn’t enough upside in it to make it viable.

Behavioural accounting

To be fair, this is a term I’ve probably made up. (If it is, I hereby claim the copyright. If it isn’t, the expression is the property of its current owners.)

But what this is about is looking at what people actually do, and extrapolating a decision from there.

Now, it has to be what people actually do. That’s really important.

It’s not what they say – whether that’s people answering a survey or a business claiming to have thousands of happy customers.

It has to be what they do.

And you’ve got to be especially careful when the proposal in front of you suggests a course of action which is not consistent with what people actually do. The cost of behavioural change on an organisation-wide level is prohibitively expensive and while I wouldn’t say there is never a business case for it, the number of times you are likely to get a positive RoI on an investment of that kind is no more than once or twice in the course of your entire career.

Looking at what people actually do is a great guide to the decisions you should take – and, often more importantly, the decisions you should avoid.

It’s been said the eyes are the windows to the soul.

Well, the decisions people make are the windows to their soul too.

Perhaps a couple of examples would help…

1-Taxis

If I was in the market for a relatively inexpensive, good value, reliable, easy and cheap to repair sort of car, I could conduct an extensive piece of research into every model offered by every manufacturer on the market, compile huge cross-model comparison spreadsheets, and go on dozens of test drives.

Or…I could just look at what taxi drivers choose.

If there is one group of people who are optimising for exactly the attributes I’m looking for in a car, it’s taxi drivers. (Not London-type black cabs, but the normal cars which regional taxi operators tend to use.)

Everywhere I go, taxi drivers overwhelmingly choose diesel Skoda Octavias for the job. And if they don’t drive one of those, they almost certainly drive a Toyota Corolla Estate Hybrid. While there is the occasional other model, 80% or 90% of the taxis I see are one or other of those models.

So I don’t need to conduct extensive research over the course of several months to choose a good value, reliable car. I just buy one of the two models that the overwhelming majority of taxi drivers actually drive on a day-to-day basis.

Equally, if I ran a taxi fleet and someone pitched me on buying 100 of some other brand for my fleet, I’d be very suspicious because if it was such a good idea, I’d see a vastly higher number of Brand X on the roads working as taxis than I do.

Decision made. I’m buying 100 Skoda Octavias, probably.

2-Inexplicable inconsistencies

The time I’m most sceptical of a proposed course of action is where the people who are proposing it are not acting consistently with the opportunity they are pitching.

A great example of that at the moment is people selling AI solutions.

People claim that AI will save businesses $ billions. But if that’s true why are all the AI companies so focused on helping you make animated videos of your dead cat?

Frankly, the real world business case for videos of people’s dead cats is pretty much zero. Sure, people might enjoy playing around with that, and posting their videos on social media. But there is not a billion-dollar market in consumers ponying up thousands of dollars a year to make videos of Muffin, their much-beloved, sadly-departed cat from when they were a teenager, brought back to life.

Pretty much no-one is going to be handing over more than pennies a month for that.

On the other hand, AI companies claim to have technology solutions which will drastically reduce companies’ operating costs through automating business processes.

That market genuinely is worth billions of dollars a year.

Yet, in recent weeks, MIT have published a study suggesting that businesses see no real world benefits from AI in 95% of the projects they analysed. And there are also some stats suggesting that most organisations have not reduced headcount even when they have implemented AI solution. That’s because a large number of humans are required to check that AI is doing the job properly…because, by and large, it doesn’t.

Now, I am quite convinced that there are some excellent real world applications for AI – running datacentres, perhaps, or automating low-end computer programming. But those are vanishingly small areas of operation for a typical business.

So, my question is, if huge tech companies who claim to have a magic solution to problems worth $ billions to businesses around the world, actually spend most of their own time and money perfecting videos of dead cats, why would any rational seller of tech solutions do that?

And the answer is that AI doesn’t work all that well outside a computer lab. Maybe it will one day, but right now tech companies are turning their back on a business market worth billions of dollars to service a consumer market worth pennies on the dollar.

That’s only a rational decision for tech companies if AI solutions don’t actually work as well for businesses as the people pushing AI solutions claim.

Otherwise I have an inexplicable inconsistency between what tech companies say and what they do.

Faced with an inconsistency like that (in my new, made-up discipline of behavioural accounting) I look at what tech companies actually do, and pay very little attention to what they say.

At least for the moment, AI is an easy “no”. If they people pushing the solutions demonstrate by their behaviour that they are more concerned about perfecting cat videos than automating credit control processes (or whatever other business activity), you can be pretty sure that their business solutions don’t work, or they’d be pursuing a market worth $ billions over a market worth pennies.

3-“Playing what’s not there”

Celebrated jazz pioneer Miles Davis once said the secret to a great performance was not in playing “what’s there” (ie the notes on a page) but playing “what’s not there” (ie how you play the notes).

Another of my behavioural accounting techniques is to look for what’s not there…but should be.

This draws a little from my early career as an auditor – if a company claims to have banked £10 million in sales this year we didn’t see £10 million or so being deposited in the company bank account, we were taught to immediately become highly suspicious of everything that organisation told us, and to make sure we confirmed every piece of company data with independent sources in case the organisation, or its officers, were lying to us.

To apply this in practice, what you do is ask yourself “what would need to be true for X to be true?”

In auditing, you quickly discovered that people tend not to pay you until you submit an invoice, for example. And also, they don’t pay you if they don’t think you’ve done any work for them requiring payment.

So if one company make a payment to another company – evidenced by a payment flowing into your client’s bank account – you can be pretty sure a genuine piece of work was carried out, and duly invoiced. (You do need to check that the second company doesn’t somehow funnel the cash back to the first company again, directly or indirectly, but I’m trying to keep this example as simple as possible.)

When I worked in the printing industry, we didn’t have a product without buying paper or carton board to print on. So if sales were high, but purchases or paper were low, on the face of it (adjusting for any stockholding) either the sales number is wrong, or the company hasn’t recognised enough cost for the paper it must have bought to make the products it sells.

Because buying paper was a necessary precursor for making a sale, so you would expect those two numbers to move more or less in lock-step.

In the part of the printing industry I worked in, you could even prove the sales made to each client if you wanted because we used a lot of special colours of ink (think M&S green, Sainsbury’s orange, or Cadbury’s purple).

If we claimed to be making lots of sales to M&S but were not buying much M&S green, on the face of it, our claimed sales to M&S are unlikely to be accurate. Buying the right shade of ink was a necessary precursor to making products that M&S were going to buy, because they were printed in M&S’s house colours.

Behaviour first, numbers second

Where a lot of decisions go wrong is that there’s a (generally well-intentioned) drive inside organisations to launch a huge analysis project of some sort when faced with a big decision.

Let me be clear, there will times you do need to do a full analysis. But that’s going to happen less than 100% of the time you’re faced with a decision. Significantly less.

What you should do first is triage the problem or opportunity.

First, a quick rough cut to make sure the project is worth doing at all – that there’s enough of a potential upside to make all the analysis, delivery, and ongoing operation worthwhile. If it’s not looking attractive at that level, no amount of data analysis is going to make that into a good project – generally things only look worse when you dig into the detail, so if it doesn’t look good at a “headline level” it’s never going to look good “down in the weeds”.

Then check the behavioural accounting. Look at what people do and ask yourself whether the decision you’re being asked to make is consistent with what you see people doing.

If the answer is “no”, then the quicker you shelve that idea, the faster you’ll stop wasting money on it. Your chances of making that work at all are 100-1 against – and likely to take vastly longer than you think and cost a lot more in the process.

The best decision you can make for your bottom line is to bail out early and do something more productive with your time.

As Charlie Munger (Warren Buffett’s long-term business partner) said: “It is remarkable how much long-term advantage people like us have gotten by trying to be consistently not stupid, instead of trying to be very intelligent.”

Make sure you’re the right side of the line on that, and your business bottom line will thank you.

There is a big problem with being efficient.

It’s a particularly big problem amongst tech folk, accountants, and engineers (but there are plenty of people in other professions who fall for this too). And it comes about from not thinking deeply enough about issues.

Often “being efficient” just means skating over the surface of an issue, doing the very minimum required to chalk up a “victory”, and moving on to the next thing selected for a completely superficial make-over.

That’s why people who try to “maximise efficiencies” often end up delivering outcomes which cost the company more than it needed to…and often more than the option of doing nothing at all.

Ultimately, this all becomes completely pointless.

The notes

A great example of this phenomenon at the moment are those AI note-takers people take to meetings with them.

Recently, I was in a meeting with three other people, each of whom had a notetaking app “joining in the meeting”. I had a notebook and a pen.

While we were talking, those apps were busying away, transcribing every word everyone said (sometimes hilariously, but that’s not my point here). And at the end of the meeting I received 3 sets of AI note-taker transcriptions together with a list of action points from each.

Now, first of all, those action points were all slightly different. This is a problem in itself, potentially, but also not my main issue with this process. (Although I should point out that having 4 people all running off to do slightly different versions of the same action point is a recipe for chaos, not a recipe for efficiency.)

But my main issue here is with the process.

In the name of efficiency, we all now had 3 sets of verbatim notes to read if we wanted to make sure we had been correctly quoted – these notes ran to about 10 pages of A4 from each app.

So that’s 30 pages of A4.

Google tells me that would take about half an hour to read. Times 4 people, in this instance. So 2 hours overall.

The people in that meeting were all well-paid professionals, and while I don’t know the day-rate of everyone involved, let’s imagine that’s several hundred pounds-worth of time, collectively.

That’s the cost of auditing the output of a notetaking app for a single meeting to make sure it’s not taking you (or anyone else) out of context.

“Ah, but I only check my own!”

Sometimes people tell me that’s not true because they only check their own notes and don’t read everyone else’s.

Firstly, even if that’s true, it’s still costing you 10 minutes of your own time. If you have 8-10 meetings a day, that’s about an hour and a half of your time every day. On the assumption you’re being paid a salary, that time isn’t free. It has a cost.

Secondly, that’s unlikely to be of much help if it comes to a fight. If someone else’s notetaking app recorded that you said “X”, and you didn’t challenge it, the person you’re arguing with has a pretty decent argument that you did, in fact say “X”.

So if you read all the notes, it’s costing you and your company £100s.

And if you don’t read all the notes, at some point you’re going to get sued so hard you might end up wishing you had read all the notes.

Whether you acknowledge it or not, there’s cost and risk aplenty here.

However, from what I’ve seen in practice, a large number of people don’t read the notes at all. They just rely on the action points generated by the notetaking app.

That’s great for as long as you can get away with it, without being sued. But if anyone ever takes legal action against your business, you can be absolutely certain that the notes you never read will be used by the “other side” in evidence, and you might end up looking like a complete prat who didn’t do their homework properly when the original notes were circulated.

Win, lose, or draw the legal action, getting a reputation as someone who doesn’t do their homework diligently is not going to be good for your career prospects with your current employer.

But it’s worse than that

However, it’s even worse than that.

Let’s say you don’t read any of the notes, perhaps even including your own, because you are “being efficient”.

That means the whole industrial infrastructure of the AI note-taking ecosystem is a complete waste of everyone’s time and money. All those millions of gigawatts of electricity. All those super-powered microchips. All that time and effort in software development.

All to produce a product nobody even glances at.

Because people are too busy “being efficient”.

This is precisely the situation Peter Drucker had in mind when he said: “There is nothing so useless as doing efficiently that which should not be done at all.”

Nobody in their right mind would employ a staff member to take verbatim notes of every meeting you attend.

So using tech to “do it more efficiently” is not an advance for society. On average, it’s a net dis-benefit. It makes us all collectively poorer when we do things that take our time, energy, and money, but which don’t move us forward in the slightest.

“But the tech is really clever!”

I’m prepared to bet the tech in a notetaking app is very clever.

But a technology which is very clever, yet ultimately produces no discernible benefit for your business or society at large, is still worthless.

There are plenty of very clever people stuck in dead-end jobs because they were incapable of producing something of value.

Cleverness, of itself, is not something which has economic value – it’s the application of cleverness which, potentially, can create economic value.

There is, however, no economic value in producing sets of verbatim notes nobody reads.

That’s almost a textbook description of a completely pointless activity.

Yet it’s a completely pointless activity that a remarkable number of people are paying $200-500 a year to use.

Here, it’s very clear what’s happening.

The notetaking apps are skimming value out of your business to put into their own pockets. They are considerably richer. Every other business is considerably poorer, to an equal and opposite amount of money.

Across society as a whole, precisely zero value has been added. Money has just been moved out of your pocket into someone else’s.

Another way

I don’t use a notetaking app. I use a pen and notebook.

And I don’t take verbatim notes. I only record action points.

I tend to type them up (I’m a fast typist so it doesn’t take long). But I used to know a guy who would take a picture of his handwritten action points with his phone and send the photo to everyone who was at the meeting so they knew what they had to do before the next one.

I also do something else which AI doesn’t do, can’t do, and never will do.

As we go along, I make a habit of repeating the action point I’ve written down to get everyone’s agreement to that being an accurate reflection of our discussion.

So there are, hopefully, no post-meeting misunderstandings about what people agreed to do.

Now, you might say, that’s what the action point summary of the notetaking app records too.

But firstly, nobody has agreed those action points because nobody knew what they are until the action point summary was circulated. So there is some process, however, short of reading them, assessing them against your own memory of the conversation and/or your own verbatim notetaking app.

So that takes time. And that costs money.

My approach isn’t free. I get paid a salary and I try to always add value.

But 5 minutes to type up some action points I’ve agreed in advance with all the participants doesn’t cost much. And there’s no downstream “that’s not quite what I said” or “I thought you meant this” arguments – so there’s a saving there too.

Factor in the monthly costs of your notetaking app of choice over and above this and the economic return for your business from this whole process is, at best, flat and, more likely, negative.

All from doing something efficiently that needn’t be done at all.

Technology can be efficient – but isn’t necessarily efficient

Amongst the simpler folk in society – people like politicians and tech people – there’s a belief that technology necessarily makes everything more efficient.

I can’t say this strongly enough. People who think those two statements follow one another as unerringly as day follows night are idiots.

Tech can make things more efficient, but it can also make them a lot worse than they were before.

Tasks that used to be as simple, and low-cost, as picking up the phone and speaking for two minutes now take ten times as long by the time I’ve located my account number to log into something, verified my two factor authentication, given the 8th and 14th letter from my security phrase, and entered the PIN I last used six months ago and can’t remember off the top of my head (necessitating a password reset process that takes several minutes more).

That’s not efficient – that’s a process that converted a 2 minute phone call for 2 people (ie 4 minutes in total) into a 10-20 minute process that’s all on me as the customer because the business I’m trying to do something with has decided to “be efficient” and make everything worse for all their customers in the process.

The only thing that “more efficient” process has done is make me determined to find another business who sells what yours does and buy from them instead.

Set against that, tech can lead to improvements.

Moving to computerised bookkeeping was a definite tech improvement, as was the introduction of Word instead of typewriters.

In industry, CNC machines have been a boon, and when I worked in the manufacturing sector, we had equipment which performed activities using automated processes which had previously required human intervention. This saved us time and money, and made the process safer from a health and safety point of view too.

So technology can be transformative.

It just ain’t necessarily so.

It can also be a completely pointless, value-destroying activities like notetaking apps, which might make someone in Silicon Valley wealthy, but all they’ve done is take money out of your pocket and put it into theirs without delivering any benefit.

(Or at least without delivering any meaningful benefit – while it might be clever tech, producing verbatim notes that nobody reads has precisely zero economic benefit to anyone beyond the software developer.)

When we put an automated process into the factory, we had a business case that said “this activity costs us £X per unit now – installing this new machinery to automate part of the process will cost half of £X per unit instead”.

I’m prepared to bet that nobody using a notetaking app had someone sat next to them taking verbatim notes before. And people have got by pretty OK without that since the dawn of the industrial revolution, until now, seemingly.

If we could operate perfectly well before without notetaking apps, introducing a way to produce verbatim records of conversations (even assuming the apps record them correctly) is by definition not adding any value because there is no corresponding saving.

And if it’s taking people’s time to read verbatim notes they never had to read before because they didn’t exist, that’s a reduction in bottom line performance, not an increase due to efficiency.

Granted, it sounds more efficient if you don’t think about it for long – that’s the surface-level thinking tech people are good at.

But dig deeper to really understand what’s going on, and digital notetaking apps drain resources from a business. They don’t add value to the bottom line.

That’s true of most tech nowadays. The times it delivers true bottom line value are few and far between once you understand what’s really going on.

If you’re serious about your bottom line, ask yourself not just about the tech, but about the business processes you’ll need that go around whatever tech solution you choose.

Add up the cost of the subscriptions, the cost of operating all the new business processes, and the impact on your customer experience. Then decide if the new tech tool you want to introduce really adds value.

When you do that rigorously, you’ll find remarkably few tech “innovations” are worth investing in for your business.

They are, at best, often just a distraction from the job in hand. At worst, they destroy value.

Make sure you know which one it is before you sign the order form. After all, there’s nothing so useless as doing something efficiently which need not be done at all.

The other day, I briefly stopped by a raging debate on Twitter about the Laffer Curve and “trickle-down economics” and got frustrated for the umpteenth time about how readily people mix those concepts up.

So much so, that I suspect many of the people in the debate don’t get the point on purpose because it would put a hole below the waterline in some of their predetermined conclusions.

Before I get into the meat of this article, though, let me say two quick things.

Firstly, I’m an accountant, not an economist. Professional economists may feel I’ve over-simplified some of these issues, which I probably have – both in the interests of space and also in the interests of not having too many people fall asleep while they are reading this article.

Secondly, I’m not taking a political stance here, for or against any side in the debate. I’m just trying to explain the issues the best a humble accountant can. Admittedly, partly in the hope that I don’t end up arguing with you on Twitter about it at some point in the future.

But in popular culture, if not amongst professional economists, nearly everybody gets this wrong and thinks the Laffer Curve and trickle-down economics are the same thing. So I’m trying to untangle the nonsense you’ve probably been fed over the years, in the interests of both your sanity and my Twitter feed.

The Laffer Curve

Famously drawn on the back of a restaurant napkin by economist Arthur Laffer, the Laffer Curve is usually drawn something like the image at the top of this article – essentially a bell curve.

Sometimes people talk about the napkin-drawing episode in derogatory terms: “it’s just nonsense some bloke drew up on the back of a napkin”.

But it’s a big mistake to think that.

An early boss of mine (different times…) used to say “if you can’t explain something on the back of a fag packet, you don’t understand it well enough”. (For non-Brits, that’s a packet of cigarettes, back when people routinely smoked in the workplace. Although, ironically, not my old boss, who was a confirmed non-smoker.)

For me, almost the purest form of explanation is something that can be reduced to a single image. It’s one reason I’m such a big fan of newspaper cartoonists who sum up the biggest stories of the day in a single image that forces you to think more deeply about the headlines. The Dilbert comic strips and the Alex cartoon in the Telegraph are also great examples of this style of impactful storytelling with just three or four images and a handful of words.

There’s a big difference between “simple”, ie reduced to its essential elements for impactful communication – like a newspaper cartoon – and “simplistic”, ie delivered at the level of a 5-year-old by someone who doesn’t know what they’re talking about – like when a professional politician of any party speaks, for example.

So, the Laffer Curve is simple. It’s not simplistic.

And if you think about it for more than 2 seconds, and you’re even a tiny bit smarter than the average politician (to be fair, that’s not hard – I own pencils for which that statement is true), the message of the Laffer Curve is inarguable.

What’s more, it’s inarguable because pretty much every person in the world applies the principles which underpin the Laffer Curve whether they choose to recognise that or not.

Its principles are simple. If taxes were set at 0% across the board, Arthur Laffer illustrated that a government would collect no tax revenues.

And if the tax rate was 100%, government tax revenues would also be pretty much £0, because nobody would have any incentive to work, so the government wouldn’t collect any tax.

In between the 0% and 100% tax rates, however, there is a point at which a government will maximise its tax revenues. Below that point, they are “leaving money on the table”. Above that point, there is an increasing disincentive to work, so people choose not to, leading to a reduced tax take.

It’s the real world

Many people would have you believe this is some sort of right-wing free market messaging, but the concept behind the Laffer Curve reflects the real world almost perfectly.

If you have ever employed a tradesperson who insisted in cash for the job, they were applying the Laffer Curve. By the time they had declared the income and been taxed on it, they would rather have been sat at home watching the telly than fixing your blocked sink.

At the moment in the UK, there is a significant cluster of business reporting revenues just under £90,000pa because if they go over that level they have to start accounting for VAT and pass on an extra 20% tax hike to their customers, which (so those businesses believe) would reduce their income. (I accept this might be more a regulatory compliance cost factor, rather than the tax itself, however it illustrates the principle, so I’m leaving it in.)

And if you are lucky enough to be in a salaried role in the UK which pays £100,000 a year, the last thing you want is a pay rise, because between £100k and £125k a year, the government currently takes 60% of your salary in extra income tax. (Which reduces to 45% once you get to £125k – see my comment above about how smart politicians are compared to pencils.)

So, people who would otherwise earn, say, £110k a year often lock away that extra £10k in a pension fund they can’t touch until retirement, instead of taking the extra pay and spending some of it, after taxes, in the real economy and thereby boosting economic growth.

While people might disagree on what the precise tax rate at which government tax revenues are maximised might be, very few people think that suddenly paying 60% tax on your income at £100,001-plus would encourage people to work harder and earn more money, given how little of it ends up in their pockets.

Love it or hate it, the Laffer Curve reflects a real dynamic in ordinary people’s lives every day of the week. It may be politically inconvenient to some, but we all make decisions based on the principles underpinning the Laffer Curve.

The “back of a napkin” thing is simple. But it isn’t simplistic. It’s an insightful summing up of an essential truth.

What it isn’t, though, is an endorsement of “trickle-down economics”, as some people like to claim – either in ignorance or as a deliberate attempt to mislead.

The Laffer Curve is purely a way to talk about how to maximise the government tax take, and puts forward the proposition that there is an optimum tax rate at which government tax revenues are maximised. Taxing at rates either below or above that point leads to a reduction in tax revenues.

You might like the Laffer Curve, or not like it, but like the famous Winston Churchill quote about the truth, in the end, there it is.

Everyone in the whole world – including you – applies the principles behind the Laffer Curve every time the question of how much tax they pay comes under consideration.

Precisely where the optimum tax-maximising point is, is another matter and it’s not something Arthur Laffer built into his curve.

It’s likely to be different between countries, and even between different groups in the same country. But overall, if you aggregate all the decisions in the country, the Laffer Curve reflects the reality of human decision-making.

The trickle-down

I don’t especially like the expression “trickle-down economics” because it always suggests someone with a bladder problem to me. However, I’ll use it here as it’s a commonly-used term.

The principle behind trickle-down economics is also simple…albeit possibly simplistic.

The principle here is that if you cut tax for high earners, they will go and spend more money in the economy at large which, in turn, will create more jobs – and, in time, higher incomes – for everyone else too.

A bit like the “just a scribble on the back of a napkin” critique of the Laffer Curve, “trickle-down economics” is a deliberately slightly derogatory term for what might more properly be called supply-side economics.

Popularised by the influential Chicago School of economists – and much loved by political leaders on both sides of the Atlantic in the 1980s – supply-side economics suggests that low taxes and reductions in government regulations will encourage companies to invest, thereby boosting economic growth and bringing prosperity to all.

Now, at some level, this isn’t the craziest concept in the world either.

Some of the applications of it have, arguably, been crazy, but the concept is sound.

To give a bit of a real world example (and I accept this isn’t strictly an issue of trickle-down economics) it’s been quite fashionable of late to let businesses pay lower taxes “to stimulate investment” by giving generous allowances against their corporation tax for capital investments businesses make.

The theory is that the effective reduction in tax (that’s where the trickle-down but comes in) will motivate business owners to spend more – and specifically to spend more on capital equipment which will help grow the economy and provide jobs.

That’s a reasonable enough theory. But it’s just theory. In the real world something a bit different is going on.

A good recent example is when small business owners in the UK were able to fully expense capital equipment purchases, up to a limit, in the year in which they bought their equipment. At least some small businesses bought more items of capital equipment because they got an immediate 100% allowance against their tax bill for doing so.

So far so good, right? Working exactly as planned.

Where the principle goes a bit iffy is that, for a while, some of the capital investment was in things like a new electric BMW for the company’s owner. Now, I have zero objection to company owners buying themselves BMWs, or anything else for that matter, but the principles of supply-side economics fall away somewhat when a UK business owners buys a new BMW.

In that scenario, if there is any economic benefit to businesses at all (and that’s pretty marginal if you’re buying a car, however cool it might make you feel), it is flowing mostly to the good folk at Bayerische Motoren Werke in Munich, together with the people across Germany who make up their supply chain.

While a tiny amount of economic benefit will stick with the UK-based dealer who sold you that new beemer, the dealer element of the total bill you get stuck with for a new electric BMW is a tiny proportion of the whole.

Not, by the way, that I’m advocating protectionism. Far from it.

I’m just illustrating that not all capital spend you get allowed as a write-off against your UK tax will necessarily deliver an economic benefit to the UK economy. The same dynamic is true if, for example, you buy a new assembly-line robot from Japan, or a bunch of microchips for your datacentre from Taiwan.

That’s why, under the principles of trickle-down (or supply-side) economics, you can have high levels of tax-allowable capital investment, whilst still having a moribund UK economy which isn’t seeing the benefit of that tax allowance in jobs or incomes.

While the loss of tax revenues is borne by UK taxpayers, the benefits of economic growth from those investments flow mostly to the citizens of Munich, Tokyo or Taipei. So substantial capital investment in the UK economy can be significantly less beneficial to UK citizens as a whole than a simplistic view of trickle-down economics might suggest.

There is some benefit to the UK, of course. Those new machines are presumably making a UK business more productive to some extent as well. But the nature of capital investment is that you accrue the benefits over a long period of time – perhaps 5 or 10 years or more.

So all the up-front tax loss might take 10 years or more to come back again – and factor in the time-value of money (which I’m not going to do here) it’s probably more like 15 years before the UK taxpayer is back to the point they started, following a decision to use reductions in tax, through stimulating capital investment, in the hope of growing the economy.

Where it goes wrong

Where this goes wrong (especially on Twitter… 😉) is when people run those two concepts together and rail against both the Laffer Curve and trickle-down economics in the same breath.

They are two entirely different things, based on different principles, measured in different ways.

Now, it is true that, broadly, if you’re a believer in supply-side/trickle-down economics, you are also likely to believe that the Laffer Curve is true. In fact, people who promote trickle-down economics often use the Laffer Curve to illustrate how reducing taxes will actually lead to an increase in government tax revenues, which, they believe, bolsters their case for tax cuts.

There is an important pre-supposition here, of course, even assuming trickle-down economics works. And that is this assumes the tax rate is currently above the “maximum amount of tax collected” point.

I’m not expressing a view here on whether it is or not. And what might be true in some countries might not be true in others.

I’m merely pointing out that there is a potential flaw in thinking that a tax-cutting, supply-side economics agenda will necessarily result in increased tax payments flowing into the government’s coffers. If the “optimum collection point” was at a 40% tax rate, say, and proponents of supply-side economics pushed through a reduction to 35%, then the government’s tax take, according to the principles of the Laffer Curve which they used to justify the tax cuts in the first place, would be less, not more.

And, as I said at the outset, I’m an accountant, not an economist. There are, I am sure, a wide range of economic arguments for and against supply-side/trickle-down economics that I’m not qualified to express a view on.

But the point I want to make here is that the Laffer Curve is a pithy reflection of an economic phenomenon which accurately reflects real world behaviour by individual taxpayers. I say that without judgement as to what their behaviour should be – merely that those judgements are the judgements pretty much everybody in the whole world makes as they draw on the essence of human nature.

You might not like that answer, but it’s the truth.

Supply-side/trickle-down economics is conceptually fair enough, but has some problems when it comes face-to-face with the real world. I don’t have the technical expertise to say whether trickle-down economics works or not. I just know that, on its own terms, and applying the Laffer Curve which most people promoting the trickle-down economics agenda use as part of their justification, there are scenarios in which it might not be true that tax cuts automatically increase government tax income and/or boost the economy as a whole, along with increasing employment opportunities.

But whether you love supply-side/trickle-down economics or hate the concept to your very core, please don’t – on Twitter or anywhere else – make the mistake of thinking that the Laffer Curve and trickle-down economics are the same thing.

They are not.

What this means for business

This newsletter is supposed to be about business, not economics, so what’s the business insight here, you might ask?

Well, there are a few:

• Simple isn’t the same as simplistic. It really is true that if you can’t explain something on the back of a fag packet (or a napkin) you don’t understand it well enough. You don’t always need that 400-page report to make a decision. Sometimes the back of a napkin is plenty.
• Watch the concepts you lump together – people will often try to get you to link A and B because B is what they really want, but A makes the argument for B more plausible, even if they are not really connected at all. The Laffer Curve can be used to make trickle-down economics seem more palatable in the same way that convincing your boss that because people matter in your business, you ought to have 17 people in the HR department. In both those cases, one statement can be true without the other.
• Reality beats theory every time. The Laffer Curve reflects reality. Supply-side economics is a theory which may or may not be true in every scenario. To butcher one of my favourite business quotes, from Jeff Bezos: “when reality and theory throw up different results, usually the reality is right”.
• You don’t need to like it – you may not like the implications of the Laffer Curve shows, but the real world doesn’t care whether you like it or not. It just is what it is. Try not to let your emotions and beliefs get in the way of big decisions. You can argue against reality all you like, but in the end you can’t out-run it. At best, you can hold it off for a little while, but you’ll never escape it.
• “Directionally right” is usually good enough. For most business decisions, the nth degree of precision doesn’t matter – and the cost of working it out to six decimal places, even where you can, is generally prohibitive anyway. A common criticism of the Laffer Curve is that Arthur Laffer never said what the perfect tax-maximising % was, but that’s to miss the point. The lack of a precise % doesn’t invalidate the concept – and it’s almost certainly different in different countries anyway. In business, when a decision is directionally right, just make the decision and fine-tune is as you go. For example, there’s no need to wait three years for the £million study from a fancy firm of consultants to tell you that the reason all your customers are unhappy is because you don’t employ enough call centre agents to answer the volume of calls you receive on a daily basis. Just employ more people in your call centre, and when all your calls get answered in a reasonable amount of time, consider fine-tuning that a little.

And please, please, please – if we ever cross paths, on Twitter or elsewhere – please don’t confuse the Laffer Curve and supply-side/trickle-down economics. They are two entirely different concepts and one can be true without the other necessarily being true.

OK. “Saviour of the universe” might be putting it a bit strongly.

But it can certainly pull you out of some sticky holes. And, most importantly, give you extra room for manoeuvre when you need it.

What am I talking about? Well Flash Reporting, of course.

Flash Reporting

In case you’re not familiar with the term, flash reporting is the technique whereby a “first look” of the monthly financial results is available within 24/48 hours of the books closing at the end of the month.

In a lot of organisations, it takes 2-3 weeks to produce the previous month’s accounts – and I’ve seen some where “last month’s accounts” were not available until almost the end of the following month, nearly 4 weeks after the month they relate to had ended.

Especially in this day and age, that really isn’t good enough.

Ask yourself: if there was a big problem with your business finances, would you rather know about it in 24/48 hours, or are you happy waiting the best part of a month before you even find out there’s a problem, much less start to do anything about it?

Well, unless you’re insane, the only credible answer to that question is 24/48 hours, isn’t it?

But there’s a problem.

For very good reasons, a lot of checking and double-checking goes into preparing a full set of monthly accounts. After all, this is the sort of thing you want to be as close to 100% accurate as possible.

But checking and double-checking takes time. And before you think that AI can do that, it can’t. “Checking and double-checking” doesn’t just mean doing the maths – accounting systems already do the maths perfectly well without the aid of AI.

Rather, checking and double-checking is more of an investigative process where your CFO makes sure that all the elements of the story “add up” and there are no inconsistencies which might need further investigation.

This often means comparing performance in several different areas of the business and unpicking what might have happened, so the CFO can satisfy themselves the accounts are an accurate picture of that month’s performance.

For example, a record month in sales, but the quietest month on record in the factory is, on the face of it, inconsistent. And that should trouble your CFO.

Although, on investigation, it might just be that you shipped more product than usual from stock this month, or that a large part of the sales were some sort of “pass through” charge where you bill for a third party’s product or service which forms part of the “complete package” you sell to clients, even though it takes no time in your factory to make.

However no self-respecting CFO would close that month’s accounts until they were happy that they could explain why those inconsistencies had arisen and were very comfortable that the accounts would stand up to scrutiny by the board, the auditors, and the investors.

Time isn’t on your side

At some level, of course, everyone running a business understands the need for the checking and double-checking. But, in the past, the idea that financial results were not available for several weeks after the month-end meant the Finance Department sometimes acquired a reputation for being “unhelpful” or “not responsive enough to commercial pressures”.

It’s one of those “lost in translation” scenarios. For an accountant, accuracy is, like cleanliness, next to godliness. And it’s pursued with the same zeal you can expect from someone pursuing a holy mission.

In its own world, there’s nothing wrong with prizing accuracy above all else. It’s wise. Commendable, even.

But outside the Finance Department it’s probably the least welcome trait of a top-notch Finance Team.

So, in the last dozen years or so, a fashion for “flash reporting” has crept in so that the wider business can get a quick fix on the financial results for last month and get on with the current month, even while the Finance Department continues with its usual round of checking and double-checking prior to the final “official” accounts being released.

The quick fix

There is a trade-off, though. A quick fix on the monthly results will never be as accurate as a final set of accounts. The trade-off is that you get some slightly less accurate information almost instantly, instead of having to wait 3 weeks for the 100% accurate stuff.

When flash reporting first came in, most accounts teams thought this was crazy. Why wouldn’t you prefer more accurate information over less accurate information?

Well that’s because they tended to miss the wider business benefits of having 95% accurate information three weeks earlier – information that helped the rest of the business outside the Finance Department stay light on their toes and respond in the best possible manner to news, both good and bad, more or less as it happened.

Outside the Finance Department, speed matters more than accuracy (as long as the flash report isn’t out by much from the final reported accounts). That’s because the rest of the business tends to manage itself directionally, not accurately.

The sales team needs to know if it’s a busy month or a quiet month, for example.

If sales were 65% of target, the sales management isn’t going to do anything much different to what they’d do if sales were 60% of target of 70% of target.

That +/- 5% is irrelevant to the course of action the sales management team will be taking, even though their Finance Department will be obsessed with determining that the definitive number is 63.5% of target.

By getting a flash result 3 weeks before the definitive 63.5% number becomes available, the business can take action 3 weeks sooner, and almost certainly get the results 3 weeks earlier as well.

I’m a fan

I’m a big fan of flash reporting because it helps the business take action faster, and that’s almost always a better idea than taking action later.

But for flash reporting to work, it needs to be pretty accurate or you risk the business working on entirely the wrong problem for 3 weeks, and then flip-flopping back into fixing the right problem after that month’s accounts have been finalised. That is not a recipe for a well-run business.

To do flash reporting well, you need really good systems in place.

Not so much accounting systems, although you need those too.

But systems so you know what’s really going on in the business, because that’s what enables you to do at least an element of the double-checking you would normally do after the month-end close in time for the flash report.

And by systems here, I don’t necessarily mean IT-based systems, although that might be part of it.

But you need to be talking with the sales team regularly – not just the sales managers – to get some idea of the deal flow, long before it hits the formal reporting systems.

You need to walk through the factory regularly and assess how busy is it. Are machines cranked up to full speed or idle? Has the same four pallets of half-finished goods been in the same spot for the last three days, implying there’s a bottleneck in production, perhaps? In the loading bay a hive of activity with trucks going in and out more or less constantly, or is it like a wasteland in there most days?

That’s the information you need to do a reliable flash report, to within a 5-10% accuracy. And while systems can tell you some of the information, fundamentally you need a feel for business activity you’ll never get from a spreadsheet or a report from your MIS.

Your “feel” is doing the sense-checking job for the flash report that represents the equivalent of the “checking and double-checking” does for the formal monthly accounts. While those four pallets of half-finished goods might not, in themselves, make a difference to your flash report, noticing them means your eyes will be peeled for other signs of bottlenecks in production which might mean the factory is running inefficiently this month, with possible consequences in terms of sales volumes.

Of course, you can just prepare a flash report mechanically from the numbers available to you, but they tend to be wildly inaccurate and not that helpful for business decision-making.

We have seen a version of this recently in both UK and US government statistics, where their equivalent of a business’s “flash results” have been subject to huge revisions a few weeks or months later.

That’s because government statistics are, perhaps inevitably, prepared purely from numbers on an MIS report or a spreadsheet.

That’s “objective”, because it’s factual. But I would argue that being factually accurate about a set of inevitably inaccurate numbers is not really a positive move in financial reporting.

And that’s true at a company level as well as at a national level.

Even though, as a business, you might be prepared to trade a little bit of accuracy for a whole truckload of immediacy when it comes to financial reporting, that doesn’t mean that a wildly inaccurate “early peek” at the month’s results has any benefit to your business.

Fast, but inaccurate reporting is no better – and probably worse, on balance – than slow, but accurate reporting.

Even if the report is “objective”, because it was prepared from the numbers which were available at the time.

Done well, flash reporting can be the saviour of your universe.

Done badly, it can propel you towards a financial black hole faster than you can say “gravitational pull”.

If you’re reading this article, odds are you’re trying to grow your business and put more money on your bottom line.

And if you are trying to grow your bottom line, there’s one very important concept to get your head around: being purely rational at all times is not the best way to build your bottom line, despite what lots of people tell you.

There’s a place for rationality, of course. But if “rationality” appears above the title in the movie of your business, odds are you’re missing huge opportunities. It should appear somewhere alongside the list of this week’s guest stars in a long-running TV show.

All the very best ideas in any business are irrational, at least at first. Generally the role of rationality in business is to smother those ideas at birth just because overly-rational people don’t understand how the world’s greatest opportunities are created.

The other day, I was having a chat with a pal about the role of branding and how good branding built businesses better, and faster, than just about anything else.

That lead us onto a conversation about the role of intangible assets and the complexity surrounding brand valuation, which you’ll be relieved to hear I’m going to save for another day.

But it did lead to us having a chat about intangible assets.

Intangible assets

Intangible assets are not the most fascinating subject in the world for non-accountants (but really fascinating for accountants…).

Basically, “intangible assets” is an expression which covers the assets in a business you can’t touch.

Your office building, the machinery in your factory, the stock in your warehouse, the trucks you use to deliver to your customers…you can touch all of those. They are tangible assets.

Patents, trademarks, goodwill and so on are intangible assets. While you can’t touch them – apart, perhaps, from the bit of paper which certifies that you own the assets concerned – these are some of the most important assets in many businesses.

To illustrate, imagine you run a pharmaceutical business. Of course you need a factory, machinery, and trucks – tangible assets – to run your business. But without the original patent for whatever the medicined you developed, most pharmaceutical businesses wouldn’t be worth much.

However, when a pharmaceutical company starts the development work on a new treatment, the decision to press “go” is not rational.

Most new drug developments fail to make it through clinical trials for any one of hundreds of reasons. According to the National Institutes of Health, the success rate for drug development is just 10-15%, a number that has been broadly steady for many years.

Of course, in the early days of drug development, pharmaceutical companies aren’t spending billions of dollars.

Usually they start with an idea which might have some promise, a small team, and a small budget. As each part of the development process is completed successfully, a little bit more investment goes in, the team gets a little bit bigger, and the bet gets a little bit bigger.

But, however well-informed it might be, every penny spent in the development phase is a bet. Sure, an experienced, well-resourced pharmaceutical company might be able to tilt the odds in its favour a little by drawing on the combined expertise and experience of their staff.

It’s still a bet though.

The nearest equivalent is perhaps an expert card counter at a Las Vegas blackjack table. For them, no matter what they do, the odds are skewed in favour of the house, just as a new pharmaceutical product getting the OK for development is more likely not to work than to make it through clinical trials.

At the time the decision to go ahead with the development is made, there’s at least an 85% chance of it not being successful.

When you look at it in those terms, it’s not rational to develop anything with those odds of success. Yet businesses worth trillions of dollars have been created as a result of taking decisions which could not be considered rational at the time they were made.

Every one of those trillion dollar businesses got to be worth that much by navigating a series of leaps into the unknown, each with a high chance of failure, even though it’s not rational to bet against an expected 85% failure rate.

Balance sheets

Balance sheets are rational places too. They are a collection of assets and liabilities, all verified by the auditors as being valued correctly under the appropriate accounting standards.

Broadly speaking (and I’m wildly over-simplifying here) physical assets are valued at the lower of cost or net realisable value. Put another way, what you paid to buy it, or what it’s estimated to be worth now.

For example, a 10 year-old machine probably isn’t worth what you paid for it. With 10 years’ wear and tear, even excluding the possibility that a change in technology made the machine obsolete in the meantime, most machinery is worth pennies on the dollar against the original purchase price.

Because balance sheets are completely rational – the auditors can trace back every item on the balance sheet to proof of its original cost or evidence of a liability to a third party – you’d expect them to control how a business is run.

Yet, as non-accountants are often surprised to learn, almost no decisions about running a business are made based on its perfectly rational balance sheet.

A balance sheet serves some purposes for technical calculations like Return on Capital Employed (ROCE) which get accountants and investors excited, but for most day-to-day purposes a balance sheet is not a huge factor in company decision-making, despite it being incredibly rational.

That’s especially true when it comes to valuing a business, where scarcely any attention is given to the balance sheet even though, in theory, the value of the company’s assets, minus its liabilities, is what most non-accountants think a business ought to be worth.

To take an extreme case, Apple Inc had a balance sheet worth $330 billion at the end of June 2025, but at the same date it had a market capitalisation (what the total of all its shares were worth on the stock market) of $3.5 trillion.

Put another way, only 10% of Apple’s valuation was “rational” in the sense that it was represented by assets, less liabilities, on its balance sheet which the auditors could check back to source documentation.

The rest was “irrational” in the sense that it’s based on people’s guesses about what a unit of Apple stock might be worth.

There is a slightly technical answer to that difference which I won’t bore you with here, but the non-technical answer is that irrational thinking about Apple’s prospects was worth 90% or so of its stock market valuation at the end of June 2025, and only 10% of the valuation is rational, based on the value of assets and liabilities on its balance sheet.

Where do you focus?

That being the case, how much of your time would you spend on the “irrational” things in a business like Apple, and how much would you spend on the rational sort of things you find on a balance sheet?

Well, rationally (sorry, I couldn’t resist) you’d split your time 90/10 in favour of irrational things, wouldn’t you? After all, that’s where 90% of the value of the business is.

Of course, most businesses aren’t like Apple. I deliberately chose an extreme case. But on the US stock market, most businesses have just 25-30% of their valuation based on their tangible assts, and 70-75% is based on intangible (or irrational, you might say) asset valuations – assets like the value of the brand, their future product development cycle, and so on.

I don’t know what the split is for your business, but it might be worth taking a look.

Odds are somewhere between 50% and 75% of the value of your business, if not more, is based on outsiders’ view of the value of things they can’t see, feel, and touch (because if they could, those assets would appear on your balance sheet for an entirely rational valuation under accounting standards).

The question for you, though, is do you spend your time in proportion to the elements of your business which generate the most value?

If you ran a property company, which are largely based on the value of their physical assets, you would spend most of your time checking the properties are in good repair, the tenants are paying their rent on time, nearby planned developments are not going to impinge on the valuation of your properties, and so on.

It makes sense for people who run physical assets businesses, like property companies, to spend most of their time working with their physical assets because that’s where by far the majority of the company’s valuation comes from.

But if, say, 75% of your company’s valuation was based on “assets” like brand value, future product development pipeline, or distribution networks, how should your time be split?

Well, 75/25 in favour of things you can’t see, feel, or touch, right?

Do the maths. What’s your split?

And how does that compare to your diary?

Are you spending your time where the value is, or are you spending all your time on things that only account for 25% of your business valuation? (If it’s the latter, don’t worry. That’s what a lot of people do. Just make a start on redressing the balance right away and you’ll be fine.)

Branding

All of this really comes home to roost in areas like branding, which I’m using here in its very broadest sense to avoid having dozens of marketing strategy purists come after me.

Many people – in particular, accountants, engineers, and software developers who largely operate in “rational mode” – poo-poo the idea of brands having a value and are very reluctant to invest in building or maintaining a brand even though, done well, that might account for 75% or more of the value of the business.

While I’m not suggesting that the 90% of Apple’s stock market valuation which isn’t based on its balance sheet is all about Apple’s skill in branding their company and their products, equally I’m sure we can all agree it isn’t 0% either.

Apple is one of the world’s most powerful brands. It has a non-zero impact on what the business is worth.

But even if the brand was worth just 10% of the total valuation of Apple, which feels very conservative, that’s still an asset worth $330 billion.

I don’t know how much time, money, and resources you would spend to protect and build an asset worth $330 billion, but the answer probably isn’t “zero”.

On a smaller scale, getting a male model to remove his jeans and stick them in a washing machine was worth an 800% growth in sales of Levi’s in the mid-1980s. Even though that was a completely irrational idea which had almost nothing to do with jeans – Levi’s even degraded their signature patch on the back so you couldn’t tell, at the point Nick Kamen was throwing his jeans in the washing machine, that he was wearing Levi’s.

How much time would you spend developing assets which could generate an 800% increase in sales?

Well, rationally, not zero. Even though the activities you are working on when it comes to laundrettes and male models are entirely irrational.

And on an even smaller scale, very small businesses who do this well can make an impact far beyond anything they could achieve if they only ever took rational decisions.

My favourite current example of this is the wonderful Michelle J Raymond (not forgetting the equally wonderful Liliane Abboud) who have structured an entire chunk…pun intended…of content on LinkedIn around Australia’s national biscuit – the Tim Tam – with their Friday Tim Tam Tips.

Think about what Michelle and Lil have done here. They have taken a product they didn’t even invent (the Tim Tam) and structured an entire fun-packed narrative around the idea of being Australian and teaching people how to build their business on social media.

I don’t know what a packet of Tim Tams cost (personally I’m much more focused on the cost of Tunnock’s Caramel Wafers…ideally the plain chocolate ones). But just a few dollars in cost brings a boatload of fun, laughs, and learning which gets spread on social media and builds Michelle and Lil’s business.

The role of Tim Tams in building a business though? Completely irrational.

And yet, in this case, somehow perfect.

The moral of this story?

Often it’s the irrational factors which create the most value in a business, not the rational ones.

If you spend all your time on factual, rational, tangible things, your bottom line is probably suffering.

50% or more of the value of your business is likely to be based on the value outsiders place on the intangible, non-obvious, “irrational” aspects of your business.

Invest your time, money, and resources accordingly.

One of the joys of social media is that strangers will stop by your profile every now and again to tell you that you’re completely wrong about something. Often something that you’re an expert in and they are just over-confident amateurs.

The area that happens most for me is when I talk about whether there is a true RoI (Return on Investment) for a particular course of action.

Bear in mind that I’m a qualified accountant, with many years assessing investment proposals across a wide range of sectors. So it’s always a delight when someone working in tech who has just asked ChatGPT what a business case is decides to stop by to tell me I don’t know what I’m talking about.

That’s partly because tech folk suffer from a common ailment (not known only to tech folk, to be fair) of deliberately “failing to understand something when their salary depends on them not understanding it”.

Tech folks’ share award vesting depends on nobody ever thinking that might not be the case. So you can forgive them for finding it easier not to ask themselves the hard questions. Their future wealth depends on nobody, themselves included, asking those.

It’s also partly because tech folk think that knowing how to populate formula for calculating something is the same as understanding what the formula means.

To be fair, that is true when writing code, which is why this is a frequent blind spot for tech folk.

You only work with surface-level thinking when you write code – you need to explain to your computer, one logical step and a time, exactly what you want it to do next. The computer can’t think for itself – a programmer has to tell it what the sequence of events, and decision points, are.

So, in the tech world, understanding the formula (or programming language, if you like) is the same as understanding how it all works, because tech has no depth to it.

Sure, some people carry out the task a little more elegantly than others, but fundamentally the only way your printer is going to print a sheet of A4 paper is if pretty much the same instructions, word for word, flow from your computer to your printer. Your computer has to communicate in the rigid, pre-determined language structures your printer understands, or no printing is going to happen any time soon.

In the interests of full disclosure, though, this is not just a disease tech folk suffer from. I’d be lying if I said I haven’t come across quite a few accountants over the years who made most or all the same mistakes that tech folk commonly do. And other professions are not immune to it either.

So even if you’re not a tech person, there’s something in this article for you.

Stay tuned, and you’ll discover what a real business case RoI looks like – believe me, this is a lot more complicated than knowing the right formula in Excel.

There are seven areas where people drastically misunderstand what a real RoI looks like. So let’s dive in:

1-It’s not the numbers

The most common mistake too many people make is to take numbers at face value.

I’ve worked with numbers for a long time and I’ve got to say a staggering amount of the numbers people throw into a business case are wrong.

Not mathematically wrong, usually. Most people can get Excel to add up a column of numbers at least semi-competently.

But wrong in the sense that the numbers are a biased view of the problem (consciously or unconsciously), or that they address only a partial view of the data, or that they sound pertinent to the case being put forward but in reality are only tangentially relevant.

My most recent interaction with an over-confident tech person was them claiming that a particular app (which they may or may not have had a hand in writing or specifying) was generating “high levels of satisfaction”.

When I checked the App Store, this app had a 3.8 rating – which isn’t terrible, but isn’t stellar either.

But this tech person missed the point: people didn’t want to use the app at all. They wanted a different solution. So they were dissatisfied. But if the app was the only way to get the service they needed, then, their experience was that the app itself wasn’t terrible.

I don’t know about you, but that’s not a set-up I’d use the expression “high levels of satisfaction” to describe. But this tech person had missed the bigger point that people fundamentally weren’t happy about their experience.

Pretty much every number presented as part of a business case for anything is either less than the full story, or capable of one or more possible interpretations which are different from the rationale being put forward.

If you think a business case is just about finding some numbers and slotting them into a model, you’ve missed the point. You need a much deeper understanding of reality to truly understand whether a proposed course of action is likely to have a positive RoI.

2-Task not system

As I often say, one of the biggest lies in business is the old mantra that to solve any problem, you break it into its component parts and solve each component separately.

This mindset tricks a lot of tech folk into thinking they are making a positive impact on the bottom line when they are doing the exact opposite.

Imagine your business wants to dramatically grow its revenue line. There are many, many components to make a sale in most businesses, but every sales process starts with finding a lead.

Now, if it’s my job to generate leads, I guarantee you I can generate cheaper leads very easily.

Over the last couple of decades I’d have switched from off-line to on-line. I’d have switched from SEO to social media. I’d have switched from text to video. I’d have switched from long-form video to short-form video. I’d have switched from Facebook to TikTok. I’d have integrated AI into whatever I was doing. And so on.

Each one of those – at first, anyway – was cheaper than whatever most businesses were doing before. (In time, they also all trend back to being about as expensive as whatever solution they replaced, but that’s not my point here.)

But, as each wave of new tech is introduced, and as costs reduce dramatically (at first, anyway) that sounds like a positive RoI to most tech folk.

Except your business objective is to grow the revenue line. Lead gen that costs half as much but converts half as well is no better than whatever you were doing before. And even reaching that conclusion requires you to ignore switching costs and training costs.

Focusing on a narrow task can make it look like a business is generating positive bottom line results at the same time as the business is actually going backwards at the “whole system” level.

It’s rare I come across a tech person who is able to do that – especially when the definition of “whole system” encompasses things that can’t be done with tech alone.

3-Visible vs invisible

Tech only works in a visible dimension – what you see is what you get.

If a few lines of code say “send a message to the printer to let it know I’m about to send a document to print and it needs to wake itself up”, then that’s exactly what happens. Every. Single. Time.

That code will never do anything else because it’s incapable of doing anything else. That’s an entirely visible process.

Where humans are involved – which sooner or later they will be, until tech folk work out a way to sell only to other spambots – not only is an exclusive focus on the visible elements less than the full picture, it’s often positively harmful.

For example, I know very few humans who are enthralled by the constant diet of AI spam which pollutes most social media platforms these days. Yet there seems to be no end to it.

The visible dimension might still be getting attended to – maybe your AI tools are auto-posting to Instagram for you. And maybe you’re even still getting clicks and impressions (although it’s no coincidence that those measures are down across most platforms as AI spam overwhelms our social feeds).

Here, the visible dimension of “still posting 3x a day” doesn’t help in the slightest when it comes to assessing how your social media followers feel about your business. The clicks and likes are circumstantial at best…and nowadays are probably mostly automated anyway, so they mean little or nothing.

When we focus only on the visible elements of whatever we’re doing, and forget about the seven-eighths of the invisible elements below the waterline, odds are we’re going drastically off-course and won’t notice until it’s too late because all the visible elements seem OK.

The invisible is a valuable leading indicator of the visible which you ignore at your peril – yet most tech folk never let it enter their thinking.

4-Logical not illogical

Tech is entirely logical. That’s how it works. Logic plus a lot of 1’s and 0’s.

The problem is that only perhaps as much as 5% of everyday life is exclusively logical. So the thinking processes deployed in tech are only valuable 5% of the time.

The tech view of the world is that to sell a pair of jeans you run lots of online ads, get people to click on them, join a sales funnel of some sort, go on at some length about features and benefits, and eventually X% of the people you shovelled into the sales funnel will buy some jeans.

Back in 1985, top London ad agency BBH got Nick Kamen to take off his Levi’s and throw them into a washing machine in a laundrette while Marvin Gaye’s “I Heard It Through The Grapevine” played over the top.

This is a classic, much-admired, award-winning ad.

But not a single element of that was logical. Not the actor. Not the laundrette. Not the song.

None of them had anything to do with jeans in general or Levi’s in particular.

The brand wasn’t mentioned in a voiceover. Only in the last couple of seconds of the ad do you discover what the ad is for. There are no long lists of features and benefits. No sales funnels. No tech.

Yet sales of Levi’s went up 800% in the aftermath of that TV ad running.

I can pretty much guarantee that nothing you do by way of AI-generated Facebook ads is going to grow your sales 800%.

That’s because we live in a predominantly illogical world. Logic only gets you so far. But, in the world of tech, logic is all there is.

5-Tech view not user view

Tech folk get very excited about the tech, but often forget about the user.

The problem with that is, ultimately, the success of your business is determined by your customers buying from you, not by how clever your tech is.

For example, I’ve yet to come across a chatbot which wasn’t complete garbage. So the minute I see one of those popping up on a website, I know the business concerned doesn’t really care about me as a customer.

If it did, either I wouldn’t have had the problem in the first place, or there would be an easier way of sorting it out than going through a series of inane questions which I had to rephrase 5 or 6 times because I didn’t use the precise language the chatbot was programmed to understand when describing my problem.

Almost every piece of tech I interact with has this problem in spades. It’s one of the main reasons I’m deeply sceptical that tech can provide a positive RoI solution to many more of the world’s problems.

However I have previously written about my experiences of trying to book my car in for a service before which I won’t repeat here.

All you need to know is that poorly thought-through tech is just about the most unhelpful way of getting simple things done that you can imagine.

6-Real world vs laboratory environment

A lot of tech works fine in theory.

In a lab, where they can control all the variables, feed a model a pre-determined case study to demonstrate how smart the tech is, and the responses are selected from a pre-determined, pre-optimised list, tech folk can make almost anything look slick.

Trouble is, the real world looks nothing like that.

In my days running a 1,000 seater call centre, while of course there were some well-worn topics people used to call up about, scarcely a day went by when a caller didn’t raise something completely random that we hadn’t heard before.

Real life humans are vastly more unpredictable than tech folk think they are. Even when we thought we had made everything simple and obvious, to a fair proportion of the people who called us up 24/7/365, they weren’t.

And that’s where the tech falls down.

Doing something clever in a lab proves nothing about how it’s going to work in the real world, with a whole boatload of extra randomness thrown in.

The thing is, in our call centre we already know how to respond to things that happen regularly. When someone called up to change their address on our records, pretty much everyone in the call centre could do that competently with little or no tech.

These calls were simple and easy and dispatched in a minute or two by a human for next to nothing, without requiring us to invest in a £10m data centre for our online AI chatbot and integrated website first.

And tech was pretty much useless for anything outside the mundane because the developers had no idea what the real world looked like for us and our customers.

While humans are really good at responding to new and unusual challenges on the fly, tech just drops into a doom loop of serving up essentially the same FAQ write-up which didn’t fit the customer’s needs 10 minute ago when it served them the same article either, but still refuses to connect the customer to a human.

7-Cost to customer, not cost to firm, is what matters

Whenever I hear a sales pitch about how much money a piece of tech will save, I silently roll my eyes. (And sometimes, not so silently.)

Of course, I don’t want to run my business any more expensively than I have to. But that’s not the dimension that really matters here…except to the tech folk who have hung their whole sales pitch on that.

What tech folk (and quite a few accountants) fail to understand is that the dimension we are really trying to manage here is the RoI to the customer of dealing with our business.

Not the cost to the firm.

Those are two entirely different – and often completely unrelated – concepts.

Put yourself in a customer’s position for a moment.

Let’s say I used to be able to ring your call centre and get my address changed in the course of a 2-minute phone call. That was a typical call duration for us in my old call centre.

But now, to save the firm money, the business fires everyone in the call centre and invests half of the savings in a new tech stack to automate all their customer interactions. That’s an impressive RoI, right…?

Wrong.

Now I’ve got to go to your website and navigate some crappy chatbot for a few minutes to find out what to do.

Then I need to enter my customer account number – which I can’t remember off-hand, so I need to search through my emails to find that information. Thankfully “in the interests of security” you only ever show a line of asterisks and the last four digits of my account number in the emails you send me several times a week.

So now I need to go back in my emails for three or four years until I find the one that has my full account number on it from when I originally set up my account.

Five or ten minutes later, I’m back at the login screen with the required information.

Except now I need my password. A password I input when I set up the account four years ago and have never used since.

So I click the “reset password” button and go back to my emails to pick up the magic link – only valid for 15 minutes – which lets me reset my password.

Now I need to guess which rules your tech person has randomly decided I need to follow in setting passwords – capitals, or not? A number, perhaps? Maybe some special characters – but obviously not all the special characters as tech folk like to give us a challenge. And, often, not a password I’ve used before on your website.

So some hacking around to find a combination acceptable to the glitzy tech website takes another couple of minutes.

Next, I’m allowed to go back to the login screen again with the details I’ve now written down on a Post-It note next to my PC. Even though your website told me I’m not allowed to write those details down “because of GDPR” or something.

Ten minutes later, after hunting through a range of menus and options, I’ve finally been able to update my new address in your system

Except now I need to go back into the email system I just signed out of to click the acknowledgement that it really was me who changed my address on your system.

Which takes me to the screen which forces me to set up two-factor authentication in order to be allowed to purchase the notebooks I buy from your business once or twice a year.

The net effect of all this from a customer perspective is that this business has converted what was a 2-minute job, for both me and the business, into a 10, 15, 20 minute job entirely for me where I am required, for free, to give up my valuable time to do something your business should be doing for me if it wants to retain my custom.

Here, whatever the tech people said the RoI to the business might have been on a spreadsheet, I’ve gone from being a loyal customer of many years’ standing to someone who is unlikely ever to buy from you again, thanks to an interaction with your tech.

Ultimately, your customers make buying decisions by factoring in the RoI on their time too. Any business which becomes too demanding on their customers’ time, with too little upside for their customers in the process, just encourages their existing customer base to buy elsewhere instead.

The RoI you really need to work on with tech projects is the RoI to your customer, not the RoI for your firm’s operating costs.

Making your business simple and easy to deal with is a sales growth superpower – because many businesses in many sectors are neither of those things.

It becomes easy to make a sale when your service is both first-rate and reliable, and doesn’t require me to take 20 minutes of my time, without remuneration, to navigate my way around the third-rate tech that some tech evangelist has convinced you will deliver a great RoI for your business.

It’s not just tech folk

As I said earlier, these problems are not exclusively caused by tech folk – I find engineers and accountants also particularly susceptible to most of them.

But a mindset of “all tech is good” or “tech is always a guaranteed money-saver” is almost certainly taking your business in the wrong direction, in the company of grifters who make big promises but don’t deliver.

Not that they don’t deliver to the narrow tech spec which you signed off. They nearly always do that.

But because they don’t deliver where it really matters – with your customers.

Because the RoI on tech projects tend not to consider the points listed above on anything other than a superficial level, it’s becoming rare to see genuinely value-adding tech in a business setting.

I’m not suggesting all tech is bad – computerised accounting was a definite improvement (albeit not without its downsides), as is email as a form of communication. The banking app I use is genuinely excellent. And Google Pay/Apple Pay have made paying for things in shops much easier.

But increasingly Big Tech is running up against the problem that only 5% of the world’s problems can be solved exclusively with logic under laboratory conditions. Those things already have a range of perfectly serviceable tech solutions in place – like Xero and Sage for small business accounting, for example.

Everything else in the world is varying degrees of not logical and/or not operating under laboratory conditions.

While ignoring non-logical and non-laboratory factors can help you show a positive RoI on a spreadsheet relatively easily, the problem is that business case doesn’t reflect reality – either for your business or your customers.

Price in the non-logical and non-laboratory factors, and increasingly I suspect tech solutions deliver a negative real-world RoI.

If you want to avoid doing the same, make sure you check off the seven points above next time you consider making an investment into new tech.

Sometimes the best solution is not to invest in new tech at all. But to do something else instead.