How the ask arrives
The quarterly review goes well: pipeline is up, the blended return-on-ad-spend figure clears the internal hurdle, and the agency's deck ends with a slide titled "opportunity to scale". The CMO, eighteen months into the role and needing a visible win, proposes doubling media spend for the next fiscal year. The platforms' own dashboards agree enthusiastically, as platform dashboards reliably do. Finance is handed a growth case built on marketing's measurement, and the CFO must decide whether scepticism is prudence or obstruction.
Why this is genuinely hard to adjudicate
The information asymmetry here is structural, not personal. Attribution models are largely scored by the channels being evaluated; the agency recommending more spend is typically remunerated in proportion to it; and the platforms grading their own homework are the same platforms selling the media. Meanwhile the CMO is not wrong to push. Under-investment in a working channel is a real and expensive error, just a less visible one than overspend. The CFO who simply refuses learns nothing; the CFO who simply approves converts a measurement problem into a seven-figure one.
- Attribution credits touchpoints; it does not prove the sale needed them.
- Returns measured at current spend say little about returns at double spend: saturation is precisely what the dashboard cannot show.
- Brand and pipeline effects arrive on different clocks, letting advocates cite whichever window flatters.
- Nobody in the approval chain is incentivised to fund the one thing that would settle it: proper holdout testing.
What the doubled budget reveals a year later
The characteristic discovery is diminishing returns arriving early: the first tranche of spend was harvesting demand that partly existed anyway, and the incremental euros buy progressively less convinced customers at progressively higher auction prices. Acquisition cost drifts up while the blended average still looks respectable, hiding the deterioration at the margin. Channel mix drifts towards whatever the attribution model over-credits. And because the increase was granted in one step rather than earned in stages, there is no clean read on where the money stopped working, only a debate about whose fault the miss was.
What to establish before the money moves
- Incrementality evidence: has any geographic or audience holdout test shown what sales look like without the spend?
- Marginal, not average, returns: what did the last increment of budget return, as distinct from the blended figure?
- Saturation logic: at what spend level does the model itself predict returns fall below the hurdle, and does anyone believe that number?
- Independent measurement: who, unpaid by media volume, verifies the figures the case rests on?
- A staged path: what would releasing the increase in tranches, each gated on incremental evidence, actually cost in lost opportunity?
Where to take the question beyond the marketing department
The agency and the platforms have already voted. Internally, finance can interrogate the arithmetic but rarely the marketing science, which leaves the CFO challenging confidence with suspicion, a losing exchange. The input that changes this conversation comes from operators who have scaled media budgets at other companies and carry the scars: they know which measurement claims held up, where saturation actually bit, and what they wish finance had asked them. A confidential brief to selected senior operators from the Global Board puts that lived pattern-recognition alongside the internal case before the commitment is made, without turning the CMO's request into a political trial.
