On paper, the paid account was fine. Last-click return on ad spend sat in a range the finance team could live with. The problem was that the number was flattering a strategy that wasn’t building the business — a classic case of ROAS being read in isolation.
The number was hiding the strategy
Run purely on last-click ROAS, the account had drifted toward whatever converted most cheaply — which meant branded search and bottom-of-funnel retargeting were taking credit for demand that other activity had created. Meanwhile, the campaigns that actually generated new interest looked “inefficient” and were being starved of budget. The account was optimising itself toward a slow stall.
Rebuild around profit and incrementality
We restructured the whole programme by intent rather than platform. Spend that merely harvested existing demand was cut back; budget moved toward the paid search and social activity that testing showed was genuinely incremental — the campaigns that, turned off, actually reduced total pipeline. We paired that with attribution modelling that spread credit fairly across the journey, so the reporting stopped rewarding the last click and started reflecting what drove new customers.
None of it works without clean measurement, so we rebuilt conversion tracking and the landing-page experience in parallel with our analytics and CRO team — because a 3.9x return means nothing if the tracking underneath it is counting the wrong things.
As with our other case studies, the figures here are illustrative of a representative engagement and the client is not named. The transferable lesson is the mindset: judge paid media on blended profit and incrementality, not the most convenient ratio on the dashboard.
If your paid reporting stops at last-click ROAS, there’s almost certainly a better decision hiding underneath it. Let’s find it.