Paid revenue climbs while organic quietly falls by the same amount

You can’t fabricate a sale. So when a new agency’s reported revenue climbs month after month, it feels like the one number that can’t be gamed. It isn’t – because you don’t have to invent revenue to make paid look like it produced it. You just have to move credit for revenue that was already going to happen.

The single most diagnostic check costs you nothing and takes one minute: pull the whole-site view and put paid next to organic and direct. If paid is rising and organic is falling at roughly the same rate over the same period, you have your answer. That is not incremental growth (new revenue that wouldn’t have existed without the ads). That is cannibalisation – the same demand, re-labelled as paid.

There are two common mechanics. The first is changing the attribution model – the rules that decide which touchpoint gets credit for a conversion – so that organic and direct lose their share and paid absorbs it. The second lives in the Google Ads conversion window – how long after a click Google will still credit a conversion to that ad. Stretch that window toward its 90-day maximum and paid quietly lays claim to conversions it barely touched.

Why agencies do it

Re-credited conversions renew retainers. A chart that climbs from the month the agency took over is the cleanest possible story to tell in a review meeting, and it requires no actual account improvement – just a settings change in the first week that reshapes how every future conversion is counted.

Real incrementality work points the other way. A proper geo-holdout test (running ads in some regions and deliberately not in others, then comparing) or a controlled pause routinely shows that a chunk of “paid” revenue would have arrived anyway through organic, direct, or brand search. That result is true and useful to you – and directly threatening to the agency’s growth story. So it doesn’t get run.

What it looks like in your report or account

• Reported paid revenue is up and to the right, but whole-site revenue is roughly flat – and organic and direct are down by a similar amount.
• The conversion window is set at or near the 90-day maximum, with no orchestrated nurture journey that would justify it.
• The attribution model was changed around the time the agency took over – often from last-click or data-driven to something more generous to paid – and the change wasn’t flagged to you.
• There is no incrementality testing, no geo-holdout, no controlled pause. “Incrementality” isn’t a word that appears anywhere in the reporting.
• Brand-search spend is folded into the headline number without being broken out – people searching your name were already coming.

What to ask your agency

“Show me the whole-site channel view next to your paid report, and tell me what the attribution model and conversion window are set to – and whether either changed when you took over.”

What it means if you get the bad answer

It means the growth line you’ve been shown may be mostly accounting, not marketing. The danger isn’t just a misleading report – it’s that budget decisions get made on it. You pour more spend into “paid” because it looks like it’s working, organic gets de-prioritised because it looks like it’s declining, and a year later you’re spending far more to stand in roughly the same place.

The fix is a whole-site view as the default reporting frame, an attribution model and conversion window you’ve actually signed off on, and at least one incrementality test – a geo-holdout or a controlled pause – to calibrate how much of the paid number is real. An agency confident in its work will welcome that test. An agency that resists it is telling you what the test would find.