Last-click is killing your creator budget.

The brand called the meeting to discuss whether to renew. Their dashboard showed $90,000 in creator spend across the quarter and $14,000 in attributed revenue. Negative ROAS on paper. The performance director had already drafted a memo recommending they reallocate the budget to paid social.

The dashboard was telling the truth about last-click. It was telling almost nothing about what the campaign did.

Creator content sits at the top of the funnel for most consumer brands. The viewer who decides to buy after watching a 90-second integration almost never finishes the journey inside the click-through window. They watch on Tuesday, they think about it for a week, they google the brand on a Sunday morning, they click a branded paid search ad, they convert. Last-click writes that conversion home to paid search. The creator that triggered the entire chain shows up as $0.

What we pulled instead

We took 48 hours to rebuild the campaign measurement before the renewal meeting. Four signals. None of them required new instrumentation the brand did not already have access to.

1. Brand search lift

The simplest signal and the one almost nobody on the brand side pulls. We compared Google Trends data for the brand name in the eight weeks before the campaign and the eight weeks during and after. Search volume for the brand grew 67% during the campaign window and held a 22% lift for six weeks after. That growth is not last-click revenue. It is people deciding to look up the brand because they heard about it. Branded search converts at three to five times the rate of cold paid search. The volume increase, multiplied by the brand's own branded search conversion rate, was roughly $130K in incremental revenue.

2. Post-window assisted conversions

The brand's analytics platform recorded multi-touch attribution data the dashboard did not surface. We pulled the full path data for the 90 days following each integration. Roughly 41% of conversions from creator-audience identifiers had a touchpoint outside the click-through window. Some converted through email five weeks later. Some through retargeting six weeks later. Some through direct type-in. Conservatively credited, this was an additional $85K in revenue the dashboard treated as belonging to email and retargeting.

3. Watch-time-anchored direct sales

Two of the creators had unique promo codes. Most viewers who buy do not use the code: they just buy. We took the redemption rate for the codes that did get used and modeled the silent conversions among viewers who recalled the brand but never punched in the discount. The math here is unfashionable because it estimates rather than measures. It is also more honest than zero. The estimated floor was $70K.

4. Geographic incrementality

The brand had run the campaign in five regional markets. We compared category sales in those markets to three matched control markets where the brand had not advertised at all during the window. Sales in the exposed markets ran 9% above the control after seasonality controls. Applied to the brand's regional revenue base, that was $55K of incremental volume directly attributable to the campaign window.

Two numbers, same campaign

Last-click view

• $90K spent
• $14K direct attribution
• 0.16 ROAS
• Recommendation: cut the line
• Creator gets blamed in the QBR

Full-funnel view

• $90K spent
• $340K combined attribution
• 3.8 ROAS
• Recommendation: scale
• Creator gets renewed at higher rate

The steelman of last-click

It is not nothing. Last-click is auditable, defensible, and consistent across channels. A finance team has no business signing off on a marketing budget that runs on vibes. The reason last-click survives is that the alternative usually arrives as a pitch deck from an agency that has every incentive to inflate the number.

Fair. The fix is not to throw out the last-click number. The fix is to pair it with three or four independent signals, each one with a method the brand can audit on its own. The point of brand lift is that it can be checked on Google Trends in five minutes. The point of geographic incrementality is that it runs in the brand's own BI tool with control regions the brand chose. The numbers we hand back are reproducible. That is what makes them credible.

What this changes operationally

If you are running creator campaigns and the only number on your dashboard is direct revenue inside a 7-day click window, you are not measuring creator marketing. You are measuring the slice of creator marketing that overlaps with direct response. That slice is real but it is a fraction of what the campaign produced.

Before the next campaign ships, ask the analytics owner to baseline brand search volume, set up matched-market geographic controls, and pull multi-touch path data for at least a 60-day post-window. None of that requires new tooling. It requires asking before the campaign ships, not at the renewal meeting.

The cheapest mistake in creator measurement is to use the same attribution model you use for paid search. The model is right for paid search because paid search is bottom-funnel. It is wrong for creator content because creator content is the part of the funnel where the buyer is still deciding to care.