Picture the meeting. Someone pulls up a dashboard. The numbers are there — sessions, clicks, conversions, a trendline pointing somewhere.
In the buzz of the conversation, something starts dawning on people.
Ultimately, everyone in the room realizes that the report being presented is not measuring the metric that matters most. That the truth they needed is still missing. That somewhere between the first brief and this conference table, months of work — a data system built carefully, by real people, at real cost — missed the mark.
Before any dataset gets built, before a single tag fires or a report gets designed, someone has to answer a question that sounds simple and almost never is: What are we actually measuring, and why does that specific thing matter?
That answer belongs to one or two people. Not a committee. Not the tech team working from a brief handed down from above. One or two people who have sat with the actual end users of the data, asked real questions, and listened — carefully enough to understand which aspects of a metric carry real weight, and which ones are noise dressed up as signal.
Here’s what that looks like in practice.
Say you’re building conversion tracking for a nonprofit donation site. The instinct is to define “a conversion” and move on. But move on to what, exactly?
Are you most interested in the moment someone gives money? Or are you trying to grow a remarketing list — to capture an email address that enters your database as an asset you can reach again?
Those are two different answers. And they lead to two completely different tracking setups.
If what you care about is the lead — the person who has raised their hand and said yes, I want to hear from you — then your tag needs to fire on the thank-you page. Not the donate button. The thank-you page, which only loads after someone has successfully submitted their email into the system.
That distinction is small in code. It’s enormous in data quality.
This conversation happens too rarely. The reason is almost always the same: defining what to measure gets delegated to the technical team. The people who know how to tag things. Who often sit in a completely different part of the building — or a completely different part of the org chart — from the leaders who will eventually need to act on what the data says.
The gap between those two groups is where most measurement goes quietly wrong. Not because anyone is careless. Because no one has been given the job of standing in the middle, making sure both sides understand each other — before the months begin, before the system gets built, before everyone ends up in that room.
A data storyteller enters at the beginning. At the moment when the right question, asked of the right people, determines whether your numbers will ever actually tell you something worth knowing.
That moment is not glamorous. It is mostly conversation, and listening, and a willingness to slow down before the build starts. But it is the only moment that prevents the other one.
allegramira.com 