DEV Community

Cover image for But This Is The Way We’ve Always Done It
Alan Hylands
Alan Hylands

Posted on • Updated on • Originally published at

But This Is The Way We’ve Always Done It

Something Just Ain't Right Here

You’ve spent days investigating something. There was this constant niggle every time you ran a report and something about it just didn’t sit right with you.

So, in you jumped, two footed. Brave of heart, steadfast of mind and foolhardy enough to ignore the nagging doubts that you were wasting your time on a fool’s errand.

You wanted to make a difference. Right a wrong. Reset the true course for future generations of report-running BI analysts.

With your sword of Truth and your shield of Virtue, onward you plough, digging through the legacy spaghetti code and mountains of flat file inputs towards your goal.


And then you find it. The smoking gun. The data WMD that you knew was hidden in there and now your hands are on it.

Some muppet years ago had inner joined when they should have left joined and it’s blown the whole report into the realms of random number generator for years.

A Harlem gospel choir has struck up in your head, singing Hallelujahs to the heavens.

James Brown is dressed like a priest, dancing and jumping on a trampoline.

A shaft of light comes through a window and illuminates you in a clear message from the data gods.

You are the data Neo, the chosen one, bringing balance to the Force and whatever other pop culture iconography you want to throw in here.

Strutting to the boss's office

You can’t wait to take it to your boss and show what a smart, intelligent, motivated badass of the analytics world you are. You are walking on air.

Floating like a butterfly and getting ready to sting like a bee but what’s this? He doesn’t look too impressed.

Quite the opposite. His top lip is curling up. Now both lips are puckering and his eyes have narrowed.

Forehead creased, he’s steepled his fingers and is looking at your comparison report on the screen and nary a breath has passed his lips in nearly a minute.

Oh fuck.

It’s not looking good. Your palms are sweaty. Knees weak. Your arms are heavy. The music in your head has switched from hosts of angels to O Fortuna and it’s about to get loud.

The sense of foreboding sits in your stomach like a lead bowling ball and, as your boss turns to look directly at you, it starts a slow but steady movement up into your throat.

“Oh fiddly fuck, what have I done?” plays on repeat. You think it’s inside your head but it might be on the office tannoy system.

Oh shitting hell.

Time for the judgement. He pauses and says something that shocks you.

“Yeah, you’re right”.

Say what now?

Your heart starts to beat again but there is a kicker.

“We still can’t change it though. Better to be consistently wrong than tell anyone we made a mistake for the past five years and it’s going to be right from now on”.

Excuse me? Beg your pardon? Did I hear that right?

So it’s better to leave a report consistently incorrect rather than correct it as that will knock out all of the historical trends and actually make the department look worse by improving something?

Or even worse, knowingly report incorrect figures as long as the general trend is right as that’s all that “really matters”?

Truth is often stranger than fiction

The stories I have heard from far and wide would make the average analyst’s hair curl and their butthole pucker.

If you haven’t experienced this particular variation on the old “but that’s the way we’ve always done it around here”, you just haven’t tried hard enough to fix something you should have left well alone. It's often accompanied by the wide eyed piercing stare of death from the longtimer who always runs the report.

In the manager’s shoes I’d never accept that attitude of course but then I’m a proper diamond geezer that way. Always kicking at the future's door. Continually pushing for improvement and squeezing that last ounce of progress out of seemingly bone-dry sources.

Or so I’d like to think anyway.

Corporate culture in this regard speaks volumes for the things you may be asked to cover up in future. If ethics are your thing (and they damn well should be), take the warning signs seriously.

The Analytics Manager’s Pledge

 “I shall wear no crowns and win no glory. I shall live and die at my post. I am the sword in the darkness. I am the fire that burns against cold, the light that brings the dawn, the horn that wakes the sleepers, the shield that guards the realms of men.”


And the best I can do is to not allow any of these god-awful frustrations to ever enter the lives of my analyst team.

This is the analytics manager’s pledge.

We don’t choose this life (anyone that does is a proper masochist) but we take on the burden anyway.

It’s what we’ve always done around here (oh goddammit…)

Top comments (3)

helenanders26 profile image
Helen Anderson

Great post! It had me nodding in agreement too. I have had the same experiences, especially when it comes to any reporting revolving around finance. Better to do it the wrong way consistently than have to rewrite and retract any past reports.

On the flip side, sales reporting and metrics change so much that it's hard to keep up. Commission rates change, account managers change territories and back again, but they want to keep reports consistent. Total nightmare.

alanhylands profile image
Alan Hylands

Total nightmare is right. I approach the whole job as an iterative process of continuous improvement but that doesn't always gel with non-data savvy stakeholders. The idea that something is only as "right" as current data and knowledge allow it to be is a hard pill for some people to swallow.

If we frame it like "Smoking was encouraged in the 1950s but medical advances show it's dangerous now. Should we all still be smoking filterless cancer sticks now because that's what people did for years?", they might start to get the concept!

helenanders26 profile image
Helen Anderson

Total nightmare, because on the face of it it's like gold. I found something that was wrong and now I can fix it. The ultimate in job satisfaction.

It's not just non-data savvy stakeholders it doesn't gel with, but stakeholders who are either incredibly risk adverse 'if we change that metric then everything will break/the big boss will ask questions', or the stakeholders that can see the work it will take to change every report that the wrong metric flows into and just can't be bothered.