The world is becoming increasingly process-orientated. There are positive aspects to that – if you’re the pilot of a 747, the Operations Director of a nuclear power plant or a surgeon carrying out microsurgery on a vital organ, I’d feel a lot more comfortable knowing that what you do, and the order you do them in, has been designed in a reliable, dependable way.
Even though most of our daily work activities aren’t anything like as mission-critical as those examples, the business world seems determined to introduce ever-greater numbers of processes and procedures to govern every aspect of what they do.
I’m not sure that’s a good thing for a number of reasons, but in my capacity as a Finance Director and CFO what bothers me most about our process-heavy world is that this approach often turns into a really expensive way to do something simple, and can easily lead to a range of less-than-wise decisions.
There is a better way…which we’ll get there in a moment…
But first, the inspiration
This article was inspired by a tweet from the always-interesting Mark Pollard (@MarkPollard on Twitter) reporting on a conversation with Phil Adams (@Phil_Adams). Here it is:
Their conversation was specifically in the context of producing good creative work, but I’d argue this concept matters for just about everything a business does.
The maths is unarguable.
Phil Adams points out, correctly, that even with just a 5-step process, where each step is done right 90% of the time, the likelihood of getting to a perfect answer each time is less than 60% (59.049% to be exact), assuming we’re dealing with a sequence of independent events (quick statistics primer here if you would like a refresher)..
Yet the holy grail for most large corporations is a detailed process with dozens of precisely crafted steps, each done right at least 95% of the time.
For a business process with even just 12 steps, however, each done right 95% f the time, the likelihood of everything happening exactly as planned is even less than Phil Adam’s example above. In fact, that process will only work as intended 54.04% of the time on average for a 12-step process, or not much over half.
Think about this. With a 12-step process where each step is done right 95% of the time, nearly half the time this process will go wrong somewhere along the line.
With a more complicated process, say one with twice that number of steps, you’re down to things going right less than 30% of of the time (29.2%). More than two-thirds of the time, your customers will be disappointed or something will be going wrong in your business. This will inevitably, one way or another, increase your costs.
I’ve worked for some large multinationals in my time and it was never much of a challenge to find processes 24 steps (or more) long. But even in relatively small businesses activities such as manufacturing products, responding to customer enquiries and tendering for new contracts can quite easily have two dozen steps or more to follow before reaching the right answer.
So this is a real problem for many businesses. Especially in large businesses which don’t often realise how often the end-to-end process goes wrong because each department involved in the process trumpets their 95% success rates on the individual elements.
That goes double when the process involves several different departments within the business and everyone tries to claim their share of the credit for doing their bit well, while doing their best to deflect the blame for customer dissatisfaction, or manufacturing inefficiencies, onto some other part of the process for which they are not responsible.
But it gets worse…
After a while, let’s say someone notices that the customer service department has ballooned to 20 people because of the volume of customer service incidents generated. Each of these customer service people costs £30,000 a year with on-costs.
Someone who feels they have a point to prove about how good their thrusting entrepreneurialism is for the business turns up at a management meeting and says something like “over half a million quid a year on the call centre – we need to save some of that cost”.
Strategies what what happens next vary, but common solutions are one or more of the following…
First, starting salaries get pared back. In a high labour-turnover environment like a call centre, there are always people leaving. So, thinks some aspiring company superstar, when one of our £30,000 a year people leave, we’ll replace them with someone on minimum wage so that, over time, we’ll only spend half as much. The Finance Director and the CEO will be delighted, they imagine.
I’m all for saving money and for giving entry-level people an opportunity to embark on a meaningful career. The part of the equation that’s often overlooked is that junior, inexperienced people are very unlikely to carry out the assigned tasks as well as the more experienced person who has just left.
Our 95% success rate for each step in the process goes down to 90%, let’s say. So our 12-step process with each step done right 90% of the time means only 28.24% of transactions will go through properly. And for a 24-step process at 90%, only 7.97% of transactions will go through right first time.
What looks superficially like a smart cost-saving move is often one of the least smart decisions a business can take, for reasons we’ll cover in a moment.
Another popular option is to look at a £600,000 cost and try to outsource the work. Although the outsourced customer service industry doesn’t enjoy the best of press, there are good businesses in there (I know, I used to run one of them).
But the first thing any self-respecting customer service outsourcer will do is ask you to be really explicit on what steps there are in the process, how you want their agent to respond in a number of different scenarios, how the interface to the client’s IT systems will work and so on.
It’s almost nailed on that your 12-step process will have just increased to, let’s say, an 18-step process. However your 95% success rate for each step is unlikely to change. After all, you’ve told the outsourcer exactly what to do, haven’t you, which you’ve naturally based on what your current staff do now.
An 18-step process performed as intended 95% of the time will have a 39.72% rate of successful completion. A similar 50% uplift on a 24 stage process means it’s right just 15.8% of the time.
I won’t even calculate the 90% options – let’s just say they’re almost never right.
These scenarios are not the fault of either the new, junior member of staff or the outsourcer. They’re only working with the material you’ve given them, and frankly they were never likely to result in a positive outcome, although both are common “budget saving” strategies.
The final common approach is to decide that everything is going to be self-service through your website which means you don’t need a customer service department at all.
There are two major problems with this.
Firstly, it is literally impossible to do everything through a self-service website. Even the kings of web self-service, Amazon, have telephone based support for particularly thorny issues…which is, perhaps surprisingly, very good, in my own experience.
If Amazon can’t make an entirely web-based service work, let’s just operate under the assumption that your business is unlikely to achieve a goal Amazon hasn’t any time soon.
The second issue is that developing web self-service options to cover every eventuality is expensive. An external firm will charge a high six-figure sum to develop one for you, or you can do it yourself but you’ll need to beef up your IT department to do it. At least over time, you’re likely to spend as much in IT resources, internal and external, as you’ve been spending in customer service costs.
Oh no, it’s even worse than that…
Sorry to say, we haven’t finished with things getting worse.
All the common solutions lead to an increase in costs.
Hire much less-experienced labour who get things wrong more often and you’ll have to beef up your management resources to deal with queries, handle dissatisfied customers and authorise refunds, special deliveries and whatever it takes to try and put things right again for the customer.
You have fewer £30,000 a year call centre agents, but a lot more £50,000 a year managers to look after the new minimum wage staff. (Again, this is not the staff’s fault, the deck has been unwittingly stacked against them.)
Outsourcing can be a good idea, but if you’ve got a £600,000 a year outsourcing contract which is your main interface with all your customers, unless you’re an extreme risk-taker (which I don’t recommend if you still want to have a business to run in a couple of years) you’ll need a relatively senior manager to make sure the outsourcer keeps on their toes and delivers what they say they will.
So you can probably add £70-80,000 back into whatever savings you make…perhaps even more than that if you deal with complex processes or work in an industry where there’s regulatory oversight to contend with.
And as for the web self-service option, there’s a real danger that you’ll just swap 20 customer service agents on £30,000 a year for 10 software engineers in the IT department, each on £60,000 a year.
The simple, low-cost solution
The biggest, fastest, lowest risk way to reduce your costs (outside the flight deck of a 747 or the control room of a nuclear power station) is to look at the process and take some of the steps out.
If you reduce the steps by only 25%, what was a 12-step process becomes a 9-step process. At a 95% success rate for each step. a 9-step process goes right first time 63% of the time.
That’s getting on for a 20% improvement over the 12-step process, even though the average success rate for each individual step has not improved. In my experience, they often do, just because there’s less going on in a customer service agent’s mind and they’re likely to make fewer errors. But let’s not even factor that very pleasant surprise on the upside into the equation.
If we consider the 24-step process with the same success rates as above, a similar 25% reduction in the number of steps makes it into an 18-step process. End-to-end, that’s likely to work as intended 39.7% of the time. That might not sound like much to write home about, but it’s about 30% better than the old 24-step process.
The changes are even more dramatic when the percentage going right is 90% instead of 95%.
The old 12-step process, now a 9-step process, is now right 38.7% of the time, and the old 24-step process, now 18 steps, is right 15% of the time. That’s a 30% improvement and a near-doubling, respectively, compared to the original success rates.
So next time you want to save money, don’t just work with the numbers, think about the underlying processes. Simplify those and you’re well on the way to a low-risk way to save a lot of your budget.
And maybe ease back on the need to have detailed processes at all.
Years ago, I heard the CEO or Ritz-Carlton speak at an event and they had absolutely minimal procedures for their staff. Instead they told their staff to focus on the customers needs and deliver whatever they, in their best judgement, thought best-served the customer’s needs.
My memory is a little hazy on the detail, but I seem to remember that any staff member could, on their own authority spend a significant amount of money on the spot to satisfy a customer need. It might even have been as much as $5.000, but don’t quote me on that. It was certainly a number in the thousands of dollars.
I remember most of the audience wincing at this approach, but it’s actually one of the smartest ideas I ever heard.
How many complaints did Ritz-Carlton get into their call centre? Almost none as all their customers were satisfied at the point the problem arose. They spent a bit more on the front end, but they saved a fortune in their call centre.
How much did Ritz-Carlton need to spend in IT resources to develop a self-service model? Nothing at all. The staff member they first spoke to sorted out whatever the problem was and they never had to go onto the website to try to find a way to resolve their issues.
And how much did this really cost Ritz-Carlton? The CEO was a little coy on that point, but he did say that firstly they worked hard to make sure very few customers were dissatisfied, so complaints were relatively few anyway. Secondly, he hinted that the average charge was a lot less than the $5,000 (or whatever the number was).
Of course, some did cost the full allowance, and some went over that limit at which point a manager did need to get involved. The impetus was still centred around making the customer happy, so there were still no calls to the call centre and so on, but a manager had to authorise the budget in light of the amounts involved.
More often, I’m sure the guest’s problems would be rectified with some express dry cleaning or the cost of an extra cab to send on the glasses someone had left in the hotel. Minimal costs against a top-dollar five-star hotel room.
And that, for me, is the secret to cost saving.
To people who hadn’t thought this through, giving every member of staff $5,000 they could spend if they had to sounded like a needless extravagance.
It was actually the cheapest way to run their business. There was just a single step in the process “do whatever it takes to satisfy the customer, up to a limit of $5,000”.
In practice only tiny amounts were spent by staff members and Ritz-Carlton saved a fortune on call centre and IT resources.
I’m not suggesting this is the right approach for every business in every set of circumstances, but there’s definitely something worth thinking about in there.
So next time you want to do things differently in your business, or you’re under pressure to save costs, why not experiment with having fewer steps in your processes, or even just fewer processes, full stop.
Perhaps try giving your people a little more discretion. Even allowing for the fact that they’ll get it wrong some of the time, this is likely to be a much cheaper way of running your business than adding an extra half-dozen steps to your current processes in an ultimately futile attempt to “engineer out” things that go wrong.
It might seem a little counter-intuitive for a Finance Director or CFO to recommend a lighter touch on the process front, but even after allowing for the fact that things will still go wrong from time to time, that may very well reduce the overall costs in your business. And isn’t that what a good Finance Director or CFO is supposed to be concentrating on?
(Photo by Campaign Creators on Unsplash)