No half measures!

The problem with the supposed science of measurement is that it's really easy to measure trivialities and really difficult to measure whether or not you're achieving the outcome you're aiming for.

So the "more data must be a good thing" industry migrates towards measuring larger amounts of more and more trivial metrics.

Once they can get you to believe that "more data must be a good thing", there's almost no end of trivialities you can spend £millions tracking, measuring, and running performance management systems with.

Here's the problem, though...

No amount data on of trivialities will help you measure what you need to achieve the outcomes you want. Quite the opposite - often a focus on the trivialities will take you away from the goal you're actually trying to achieve.

So, even as everyone is congratulating themselves on a job well done for achieving whatever metric that is, the organisation's real objective is moving further and further away into the distance.

Often, what gets measured is garbage

You're almost certainly familiar with the management aphorism "what gets measured gets managed".

That's only partly true.

Often what gets measured is garbage. And managing garbage doesn't make it anything other than garbage. It just becomes slightly better-organised garbage.

Think about this for a moment.

What would you rather have: 60% of the information you'd ideally like about the achievement of a significant goal in your life, or 100% of the information you could ever possibly get about a whole host of things that don't matter in the slightest when it comes to achieving that significant goal?

It might not surprise you to learn that, when it comes to designing corporate information systems, the answer is often "the latter, please".

Every day, people are doing the equivalent of tracking how many times in the week you put on your running shoes in the morning, but forgetting that your real goal is to run a marathon, not just to put your shoes on every morning.

OK, I won't milk it

One of my favourite examples of this is in the dairy industry.

Specifically the prevalence of semi-skimmed milk.

I can't say for sure it didn't exist when I was little. But when my granny sent me to the shops to get a pint of milk for her, there was only one type - I never needed to have a debate with the shopkeeper about which variety I wanted. The only question was about how many bottles my granny needed.

As far as I can tell, the original impetus behind introducing semi-skimmed milk was to reduce the amount of fat in the average British diet, at a time when the amount of fat in people's diets was seen as a major health issue.

Now, I'm an accountant, not a doctor, but I'm not sure about this.

Firstly, the French eat a diet with much more fat in it than the average Brit, and they don’t seem to do too badly in the health stakes.

And, just for good measure, whole milk is a fairly nutritious foodstuff - only marginally above the threshold for "low fat". (The definition of "low fat" is anything with less than 3% fat.)

Whole milk has a fat content of about 3.5%, so relative to a plate of fish and chips, for example, or a sausage sandwich, the impact a drop of milk in your cuppa has on your fat intake is tiny.

What's even more crazy is that most of the vitamins in milk are of the fat-soluble type which you find in...take a wild guess...about double the quantity in whole milk compared to semi-skimmed milk.

So, whatever the original well-intentioned purpose might have been, the widespread introduction of semi-skimmed milk has not only been an arguable-at-best solution to a problem we may not even have had originally, but it has also introduced a whole new range of problems we didn't have before.

But, like all the half-assed metrics deployed inside many organisations, we persist with persuading people to drink semi-skimmed milk anyway.

That's the power of a metric that gets managed. Even the bonkers stuff still gets managed, tweaked, enhanced, performance tracked and goodness knows what else.

And it's all pointless

For me, at least, this is all a bit pointless.

When I put milk in things (tea, coffee, scrambled eggs, porridge) I'm doing so to get the "mouth feel" of a certain proportion of fat content in what I'm eating or drinking.

So, when there's only semi-skimmed milk available, I just use twice as much of it to bring the fat content up to the level I like in my cuppa.

I've yet to encounter anyone who measures out exactly 10ml of semi-skimmed milk in place of the 10ml of whole milk they used to put in their tea before, which is what you'd need to do to get the benefit of semi-skimmed milk (assuming there are any actual benefits).

We all put a "glug" into our cuppa and if that cup of tea doesn't have the optimum balance of fat to tea for our personal taste, that glug will be a bit bigger next time.

Even if you don't do what I do, I'm prepared to bet that most people put more semi-skimmed milk into their drinks than the amount of whole milk they would have put in before. Even if it's not twice the amount, it's more than they used before because unless you've had a complete taste bud failure, it's an entirely different taste.

(And don't get me started on the heathens who make flat whites with semi-skimmed milk - a process that ought to land those responsible in jail for a few weeks at least.)

But the idiosyncrasies of my taste buds are the least of our problems here.

What do you think happens with the other "semi" of semi-skimmed milk - the half that's removed from the original whole milk?

Well, amongst other things, that's used to make cream, butter, and cheese. Often consumed by the very same people who gave up putting whole milk in their tea "for health reasons".

And, again as a sample of 1, when I think back to my childhood, all those things were much rarer than they are today, especially cream. That was a real luxury product once upon a time, but now whenever I'm in Costa or Starbucks I see people getting drinks with a big pyramid of cream on top.

I'm no scientist, but I'd venture to suggest that making a drink with semi-skimmed milk and then dolloping several ounces of squirty cream on the top is unlikely to be healthier than just making a standard coffee using whole milk.

So, not only didn’t we solve the problem we had originally with this seemingly worthy-sounding objective ("let's halve the amount of fat in milk")...we've probably added in some new problems we didn't have before.

Now people who would previously have consumed a few millilitres of whole milk in a cuppa are now spooning several ounces of 35% fat squirty cream from the top of their Frappuccino into their mouths instead of glugging down a tiny amount of 3.5% fat whole milk in a standard coffee.

Why this matters

In case you thought I had some weird obsession with milk, let me assure you that's not the point of this article.

(Although my obsession about the way people make flat whites is a very genuine source of concern to me...)

My point is simply this.

It's really easy to come up with an idea like "if we are concerned about the amount of fat in the British diet, let's just halve the amount of fat we have in our milk". It sounds like a simple solution to a problem you think you’ve got.

I'm sure there were some statistics at the time about how many gazillion gallons of milk the country drank at the time and some calculations about how many millions of tonnes of fat that would "save" the country.

But as a stand-alone measure, it's poorly designed.

For people like me who just put twice the amount of milk in their tea, there are no benefits whatsoever.

However it means that there is now twice the amount of "spare" cream which will be made into cream, butter, yoghurt or something else and then sold back to me or to someone else. You can be pretty sure that the dairy industry isn't just pouring the extra cream away. They're doing something with it and selling it to someone.

Whether you're eating a pound of cheese at a single sitting or loading up your Frappuccino with a gravity-defying dollop of squirty cream, that's likely to be a good deal less healthy for you than just using whole milk in your tea in the first place.

The dairy industry has had to invest in new factory equipment to skim, separate and store all the extra cream they produce, and truck it to wherever they can sell it, increasing their costs and thereby increasing the prices consumers pay.

And, unless artificially fortified with extra vitamins, there's only about half the amount of fat-soluble vitamins in semi-skimmed milk as there is in whole milk.

And an udder thing…

It doesn't seem to me that semi-skimmed milk serves any useful purpose and yet we have whole industries dedicated to making it, promoting it and distributing it.

In fact, it may even be (on an end-to-end system basis) positively harmful to the overall objective of helping the average Brit consume less fat in their diet if we take some fat out of our milk but encourage more people to consume heart-stopping quantities of whipped cream instead, made from the 50% of the fat that was skimmed off the milk in the first place.

But apart from giving me the opportunity to squeeze a bunch of dairy-related wordplay into this article, why does any of this matter?

The real story here is that, inside your organisation, you'll have a whole range of policies, procedures, and metrics which have come about in exactly the same way.

They were probably put in place once-upon-a-time by someone meaning well, as I’m sure semi-skimmed milk was.

But if you look at those policies, procedures and metrics today and fully consider all the knock-on effects they create for your organisation - many of which might be positively unhelpful in the context of achieving your organisation's overall objective - then they might have outlived their usefulness.

When I've worked with organisations in real trouble in the past, one of the root causes was often their fixation with a metric or an approach which might have made sense at some point in the past, but no longer did by the time I joined the organisations.

However those organisations, over the years, had become so wedded to that metric that they couldn’t imagine life without it...even though that metric was slowly strangling the organisation and preventing it achieving its overall objectives.

Have a look - you might be surprised how many of those you'll find inside your own organisation if you really try.

After all, this isn’t my first rodeo…

As you can probably tell, no AI was deployed to support my meandering prose. So if you like it, why not share this article with another human...?

Whether or not they have strong views on how to make a flat white, they'll thank you for sharing, as will I.