Marketing ROI Is a Trust Problem, Not a Math Problem

August 22, 2026 • 6 min read
Marketing ROI Is a Trust Problem, Not a Math Problem

Almost every marketing ROI argument I've ever sat through eventually gets framed as a math dispute. It rarely is. The formula for ROI is not complicated. What's complicated, and what actually derails these conversations, is whether the person across the table trusts the inputs going into that formula.

The formula was never the sticking point

Revenue attributed to marketing, divided by marketing spend, minus one. Any finance team can do that math in their head. When a CFO pushes back on a marketing ROI claim, they're almost never disputing the arithmetic. They're disputing whether the attributed revenue number is real, or whether it's been generously assigned to marketing to make the number look better than the underlying reality.

Where trust breaks down

Trust erodes in a few predictable places: attribution models that conveniently credit marketing for deals sales clearly drove on their own, revenue numbers that shift depending on who's asking, and a track record of rosy projections that didn't hold up. Once trust breaks on any of these, no formula, however accurate, will be believed without a fight.

Trust versus formula in marketing ROI
The formula was never the hard part. Trust in the inputs was.

How I rebuild it

I use a conservative attribution model on purpose, one that under-credits marketing rather than over-credits it, and I say so explicitly when I present it. That single choice has done more to win skeptical finance partners over than any amount of methodology explanation, because it signals I'm not trying to win the argument, I'm trying to get the number right.

I also show my attribution logic openly instead of presenting a finished number as a black box. When a CFO can trace exactly how a dollar of pipeline got credited to marketing, they stop debating the model and start debating the strategy, which is a far more useful conversation.

Consistency over time matters more than precision in any one quarter

A model that's directionally consistent quarter over quarter, even if imperfect, builds more trust than a model that's technically more precise but changes its assumptions every time the numbers move in an inconvenient direction. I'd rather be slightly conservative and consistent than technically correct and constantly adjusting.

The real work is earned, not calculated

Marketing ROI, as a number, is easy to produce. Marketing ROI as a number people actually believe and act on is earned over multiple quarters of honest reporting, conservative attribution, and owning the misses as openly as the wins. That's not a math problem. It's a trust-building exercise that happens to end in a spreadsheet.