Do you see the crescent…or do you see the whole of the moon?

Actually, that’s a trick question.

You can’t see the whole of the moon from Earth, not least because half of the moon is around the other side, only visible from somewhere deep in outer space.

And if you fly off to outer space to have a look at the other side of the moon, now you won’t be able to see the side we can see from Earth, because now that’s around the other side.

Maybe if you’re Elon Musk you can build a giant array of mirrors in space so you can see both sides of the moon at the same time from your back garden, but I wouldn’t bet on that happening any time soon.

You might wonder why this matters, but bear with me. This isn’t just a tortured metaphor…although by the end of this newsletter it will almost certainly feel like one. There’s a solid business point in here too.

Credit where credit’s due

Firstly, let me give credit to the Waterboys, whose mega-hit “The Whole of the Moon” popped up on my Spotify feed the other day. It’s their metaphor I’m about to torture here.

If you’re not familiar with the song, a 1991 UK number three record, the lyrics contrast the singer’s perspective on a situation with someone else’s perspective on the same situation. At least that’s what it’s about on the surface – I’m sure there are deeper meanings in there as well.

But because I’m an accountant, not a songwriter, it made me think about monthly reporting systems. (And yes, I know I ought to get out more…)

Sellers of financial software regularly tell me that their magic box of tricks provides complete traceability of everything a business could possibly need to know.

Of course, that’s nonsense. No software can do that.

Can some software do a better job than others of tracking and reporting data? Well, yes, of course.

But any reporting system has three big blind spots…at least:

1. When you’re reporting data, by definition you’re reporting exclusively on the past because, until an event takes place, there is no data.
2. Any system, however brilliant, can only track things someone tells it to track. If you don’t think it’s important, you won’t build a system to track it. Even if it really is important.
3. It can only report on data which is inside your organisation or accessible via publicly-available data feeds. By definition a fair chunk of the data you’d ideally like to know is in neither of those categories.

Flying blind – at least some of the time

Where data exists you should absolutely use it. Be careful about how you interpret the data, though – that’s a subject for another day – but use it, for goodness’ sake. Don’t just take wild guesses.

But there comes a point where you have to get comfortable making decisions where no data exists, and where there is no reasonable likelihood of that data ever becoming available.

Some of the information you would ideally like may well be “on the far side of the moon”. Invisible. Inaccessible. Imperceptible.

In some organisations, without data, they refuse to change anything.

But if that data is never going to be available, they’re on the fast track to irrelevance, because someone else will take the risk.

And if the idea works, it’s probably curtains for any business still waiting for more data to turn up before they make a decision.

As good a rule as any

I’ve always rather liked Colin Powell’s 40/70 rule for making tough decisions.

In his view, if you have less than 40% of the information you need then you shouldn’t make a decision as you’re just shooting from the hip.

But if you wait until you have more than 70% of the information you need, then odds are you’ve left it too late and someone else has beaten you to the draw.

Somewhere between those two points, in Powell’s view, is the sweet spot for decision-making.

Making a decision at 70% is not a guaranteed success. But it’s very likely to take you a big step in the right direction, even if you need to tweak a few things after go-live.

At any rate, making a decision before others do gets you out in front of the pack, and means you’re setting the agenda for others to follow. Not the other way round.

So what about the numbers?

This isn’t an article about decision-making, but about reporting. So I don’t want to labour the point.

But the relevance to reporting is this.

It costs time, money, and effort to develop effective reporting systems – despite what “magic beans” software providers say.

And it becomes significantly more expensive the closer you get to 100%. In all likelihood, it costs as much to get from 98% to 100% as it did to get from 0% to 98%.

More often than not, it’s impossible anyway, because the data you’d ideally like to have isn’t publicly available. In reality, the best you can hope for is probably 90% or less.

So, whatever personal threshold you set, collecting more data beyond that point becomes exponentially more expensive.

Why not try this?

As a sweeping generalisation, just a small handful of mission-critical factors account for almost all of that 40% – 70% zone.

It’s rarely more than 3-5 absolutely key metrics, whether that’s at company level, department level, or project level.

So why not structure reporting around those factors, and only those. Nothing else is likely to move the needle enough to make a difference anyway (or if it does, then you’ve misdiagnosed what the key metrics were in the first place).

By doing that, you can focus everyone’s attention on those 3-5 mission-critical factors, instead of diverting their attention with dozens of only tangentially relevant pieces of information.

And guess what? With only a small number of things to obsess over, people spend more time on them, so they are likely to be more robust and more reliable than if people spread their attention across 40 or 50 different KPIs.

(I’m not joking – in one organisation I was responsible for 43 different KPIs.)

What…just 3-5…?

Some people freak out a little at this idea, but that’s only because it seems to have seeped into our collective consciousness that all data is equal.

It most certainly isn’t.

The information about the Waterboys’ most famous song at the start of this piece is pretty much the only hard data in it. And yet if all you got from this article is the data that their most famous song reached number three in the UK charts in 1991, you’ve probably missed the point.

The learning is in the sections without hard data, things you couldn’t make into KPIs even if you wanted to. That’s where the value is.

And that’s the problem in organisations which only move in response to data.

They’ll still be completing balanced scorecards full of irrelevancies while you’re getting out in front of them and setting the agenda in your industry.

Sure, you need to manage the risk and protect the downside when implementing change of any sort. But all you’re really doing is getting very expensive people in departments like Finance, HR, and Marketing to do lots of analysis into issues which – win, lose or draw – won’t make much of a difference anyway.

Beyond a sensible threshold, “more data” has a very low – and often negative – RoI.

You don’t need the giant space mirrors…

The route to success, I’d argue, is for more decision-makers to get comfortable that they will never be able to see “the whole of the moon”, and to stop trying.

Even if you can afford a giant array of space mirrors, you’ll never see everything anyway.

Instead, how about businesses spending as much time, money, and effort delivering the 3-5 mission-critical metrics as they would spend chasing an improvement in data reporting from 98% to 100%. Odds are the business would run much better.

Reporting at great length on metrics which aren’t significant enough to move the needle materially isn’t “making data-driven decisions”. It’s “wasting company funds by tracking data that doesn’t matter”.

Knowing the difference between those two concepts is key.

As the Waterboys put it: “I spoke about wings. You just flew.”

This post originally appeared on LinkedIn.