The Report No One Read

There's a specific kind of frustration every analyst knows. You spend a week on something. The SQL is clean, the numbers check out, the chart actually tells a story. You put it in a deck, or a PDF, or a dashboard link you paste into Slack. And then — nothing. The meeting moves on. The decision gets made anyway, on instinct, or on whatever someone's boss said. Your work lands in an inbox somewhere and is never opened again.

This happened to me more times than I can count. Working in analytics at a football club, I produced reports on fan behaviour, ticketing patterns, commercial performance. Some of them were good. Most of them disappeared.

For a long time I assumed this was a me problem. The analysis wasn't sharp enough. The framing was off. I needed to get better at presenting. And some of that was true — communication is a real skill and I was still learning it.

But the longer I've spent in this industry, the more convinced I am that most of it wasn't a me problem at all. It was a structural one.

---

The pipeline everyone ignores

The data industry has invested extraordinary effort in the earlier stages of the analytics pipeline. Data warehouses got fast and cheap. Transformation tooling got principled and version-controlled. Exploration tools got powerful enough that a reasonably technical analyst can answer almost any question about their data in an afternoon.

What got almost no investment is what happens after the analysis is done.

The "last mile" — getting finished, polished, actionable outputs in front of the people who need to act on them — is still mostly solved with email attachments, Slack links, and decks that were built in PowerPoint at 11pm. The same way it was done fifteen years ago.

This is strange when you think about it. We have incredibly sophisticated tooling for moving data around, transforming it, and querying it. The thing analysts actually hand to decision-makers is still mostly a PDF or a screenshot.

---

What actually happens to analysis

Let me be more specific about the failure modes, because they're distinct.

The inbox problem. Analysis gets sent to someone who asked for it. They open it once, skim it, close it. Two weeks later they have a question that the analysis answered. They can't find it. The link expired. The attachment is buried under 400 emails. The insight might as well not exist.

The static problem. A report gets built once, reflects the state of the data at a point in time, and then immediately starts going stale. Nobody updates it. By the time it's being referenced in a meeting, the numbers are three months old. Nobody quite trusts it, but nobody wants to say so.

The context problem. Analysis that made complete sense to the person who built it arrives in front of a decision-maker with no surrounding context. What was the question this answered? What assumptions were made? What does this number actually mean? The analyst isn't in the room. The work has to speak for itself and often can't.

The format problem. The medium the analysis arrives in doesn't match the medium the decision-maker thinks in. Executives want a narrative. They get a pivot table. Analysts build for comprehensiveness. Stakeholders need a single number and a recommendation.

---

The wrong diagnosis

The standard response to these problems, in the industry, has been to invest in self-serve. Give decision-makers direct access to the data. Let them pull their own numbers. Cut out the analyst as intermediary.

This doesn't work, for the simple reason that most decision-makers don't want to query their own data. They want answers. The self-serve dream consistently runs into the same wall: the people who need insights most are also the least inclined to go find them.

Another response has been better dashboards. More interactive. More real-time. The theory is that if the dashboard is good enough, people will check it habitually and the insights will flow naturally.

This also largely doesn't work. Dashboards get built and then not looked at. The number of dashboards in your organisation that someone spent weeks on and are now visited by nobody is almost certainly higher than you'd want to admit.

---

The actual problem

Here's what I think is actually going on. The analytics industry has confused access to data with delivery of insight.

Access is solved. Anyone with a data warehouse and a BI tool has access to their data. The problem was never that the data didn't exist. The problem is that turning data into something that gets acted on — that changes a decision, that shifts a priority, that creates genuine understanding in someone else's mind — is a different problem entirely. It's closer to publishing than it is to analysis.

Publishing is a discipline. It's about format, presentation, timing, context, audience. It's about meeting people where they are rather than expecting them to come to you. It's about making something that can stand alone without the analyst in the room to explain it.

The analytics industry is exceptional at the analysis part. It's still mostly winging the publishing part.

---

Why this matters

I've been thinking about this for a while, partly because it shaped the problem I'm now trying to solve with Infigured. But the reason I wanted to write about it here is that I think it's a framing shift worth making independent of any particular tool.

If you think the problem is data quality, you invest in better data pipelines. If you think the problem is tooling, you buy another platform. If you think the problem is analyst skill, you hire more senior people or send them on training courses.

If you think the problem is publishing — the last mile from analysis done to insight acted on — you start asking different questions. What does a finished piece of analysis actually need to communicate? Who's reading it, and what do they need from it? How does it stay live and trustworthy rather than going stale? How does it arrive in a form that invites action rather than archiving?

Those are harder questions than "is the data correct." But they're the ones that determine whether any of the work upstream actually mattered.

---

Max is building Infigured, a tool for analysts to publish analysis — not just share it. If this resonated, sign up for updates or reply if you found this via email. I'd genuinely like to hear your version of the report no one read.