Your Business Case Was Never the Problem
When foundational BI work doesn't get funded, the postmortem always lands in the same place. The business case wasn't strong enough. Leadership didn't see the value. Somebody should have told the story better.
I've watched that diagnosis get applied for decades. I've applied it to myself, sitting with a rejected request for a data warehouse redesign, wondering which slide lost it.
The case was usually fine. What killed it was the budget column it arrived in.
The BI and analytics function sits inside the enterprise as a cost center. That was true of nearly every one I've worked in, run, or been brought in to fix. And that structural position licenses exactly one argument: this will reduce overhead. So semantic layer work, metric definition ownership, warehouse redesign, validation discipline, all of it gets dressed as efficiency before it ever reaches a finance committee, because efficiency is the only dialect a cost center is permitted to speak.
Then it competes for money against initiatives argued on growth. It loses on category. Nobody in that room was evaluating whether the work was necessary. They were comparing a savings number against an expansion number, and savings numbers do not win that comparison.
Watch it happen right now and it's almost too clean. In most enterprises the AI budget is the one line being argued on growth terms. Same company, same fiscal year. One request framed as expansion. One framed as overhead reduction. The AI request wins, and it wins carrying a hard dependency on the exact foundation the other request was asking permission to build.
The force underneath this is structural, and it predates every technology cycle I've lived through. The analytics function measures the return on every other function in the building and has never been positioned to argue its own.
My doctoral research found that executive sponsorship and governance as trust predicted return on data initiatives more reliably than technology stack selection. I read that finding for a long time as a statement about enthusiasm, as though sponsors mattered because they cared. Standing is the better word for it. A sponsor is the mechanism that moves a request out of the overhead column into one where it can be compared against growth. Sponsorship predicts return because a sponsor changes the category of the ask.
Three moves for anyone running a BI or analytics function and tired of losing an argument they're right about.
Stop pricing foundation work as savings. Attach it to revenue initiatives that already have money. This isn't a rhetorical trick. If the AI program, the customer platform, or the regulatory reporting build genuinely depends on governed definitions and a warehouse that holds up, then the foundation work is part of the cost of that initiative and belongs inside its budget. Describe the dependency accurately and the column changes on its own.
Get the sponsor before you build the case. Most analytics leaders I've watched do this in the wrong order, spending six weeks on a business case and then shopping it around for support. The sponsor determines which column you're arguing in. Building the case first means writing it in a language you've already been assigned.
Concede the requests that genuinely are overhead. Some of the backlog is housekeeping and everyone knows it. Saying so out loud is what makes the rest of the list credible. I have gotten more foundation work funded by withdrawing three items than by defending all ten.
For twenty years the standard advice to BI leaders has been to get better at telling the story. Sharper slides. Tighter narrative. Speak the language of the business. It's decent advice and it has never been sufficient, because a better argument delivered from the wrong column is still an argument from the wrong column.
Know where you're standing before you open your mouth. Then go change where you're standing.
If you found this briefing valuable, share it with a colleague in data and analytics.
