I wrote once:

MEASURE EVERYTHING

 

It strikes me as even more obvious now:

How COULD YOU NOT measure everything?!

It’s not “why”, it’s “HOW CAN YOU NOT”?

 

When you go to the shop, do you just eyeball how much you should pay?

Do you look at the CART, and GUESS how VALUABLE all that is?

And then take some COINS out of your pocket, and just INTUIT how many of them you should throw at the checkout?

 

Obviously you don’t do that

Yet in this modern era where we can basically MEASURE so many things, TRACK so many things,

You’re like, “nah I’m good”

I’ll just EYEBALL it

Eyeball my entire life

Then wonder what went wrong

 

I will grant some merit to NOT being so fastidious about performance indicators and models and quantification and unending stream of data,

Obviously if MEASURING doesn’t actually measure what you intended to measure — it’s just NOISE. It’s WORSE than nothing

Just like wrong knowledge is worse than knowing nothing and KNOWING it

Likewise, even if you MEASURED something important — you still don’t necessarily know how it works

And again: all this DATA, all this INFORMATION — may LULL you into delusion of KNOWING

Green lumber fallacy, illusion of explanatory depth, etc

Then there’s the problem of overoptimising for a single variable = Goodhart’s law

“When a measure becomes a target, it ceases to be a good measure”

You’re now overindexing on the single target you managed to grasp — but at the cost of missing the forest for the trees

You should have gotten your head out of your ass instead, and focused on the whole

 

I’ll also add psychological argument:

Is the GAIN from all this fussy measuring greater than the cost?

Is it worth:

-added STRESS?

-sacrifice of flexibility, nimbleness?

-and plain extra management, resources, labour, work, hassle required?

Perhaps it’s simpler to just GET IT DONE, then see if it’s improving, or not.

 

With all that said however,

You can rarely improve something without creating some extra problems

And then you don’t solve those problems by ignoring them, or backpedalling

You EXPLORE for the possibilities of improvement, then work around the hurdles along the way — until you have consolidated your advantage

MEASURING is like that

We don’t solve the problems of measuring by CEASING to measure

You don’t solve a broken clock by ceasing to look at it

We solve the problem of measuring by being BETTER AT MEASURING, and USING THAT DATA to our ADVANTAGE

And we solve the broken clock by fucking fixing it