DoubleVerify just surveyed 2,020 marketers across 25 countries about how they measure what works. Incrementality testing, the method everyone keeps calling the future, wasn’t even one of the listed options. Platform reports and last click attribution still won the most votes, and that gap matters more than the hype.
- DoubleVerify’s November 2026 survey of 2,020 marketers found platform native reporting (55%) and marketing mix modeling (48%) lead the way, ahead of last click attribution (35%). Incrementality testing wasn’t even on the list.
- Winterberry Group’s survey of 121 enterprise marketers found 83% call unified measurement a top priority, but only 9% say they can actually pull it off.
- MarTech Cube says the industry is moving from claimed attribution to proven causation. The surveys above suggest that move is still mostly talk.
- The real gap isn’t a missing tool. It’s that most teams run several measurement methods side by side without reconciling what they disagree about.
Why Did DoubleVerify’s Own Survey Skip Incrementality Testing?
DoubleVerify’s list of measurement methods, published November 18, 2026, named five choices. Incrementality testing wasn’t one of them, and that tells you more about where marketer attention sits than any keynote does.
The survey covered 22,000 consumers in 21 countries and 2,020 marketers in 25 countries, fielded by Sapio Research. Here’s the breakdown:
| Measurement method | Share of marketers using it |
|---|---|
| Platform native reporting | 55% |
| Marketing mix modeling | 48% |
| View through attribution | 40% |
| Multi touch attribution | 38% |
| Last click attribution | 35% |
I wouldn’t call that a surprise. Platform native reporting wins because it’s already built into the ad account and costs nothing extra to read. Marketing mix modeling comes second because it doesn’t need user level tracking to work, which matters now that so much of that tracking is gone. Testing for incrementality takes real setup: a holdout group, a test window, someone who can read the result without panicking the moment a number dips. Most teams haven’t built that muscle yet, survey or no survey.
Is Last Click Attribution Actually Dying?
No, not yet. DoubleVerify’s survey put it at 35%, close behind the newer methods rather than replaced by them, even though almost nobody will defend it out loud. I don’t blame them.
MarTech Cube’s November 20 analysis frames this as the end of what it calls attribution theater: marketers defending whichever channel claimed the sale last, instead of proving which channel actually caused it. That’s a fair read of the mood in the industry. It’s a less fair read of the budget meeting, where last click attribution still shows up on the dashboard every Monday morning.
Why would a team rip out a method that’s free, already wired into the ad platform, and easy to explain to a CFO, even if everyone agrees it overstates the channel that happened to close the deal? It survives for the same reason bad habits usually do. Replacing it takes work, and the current number, however wrong, is at least familiar.
What Does Winterberry Group’s Measurement Gap Actually Show?
It shows marketers want unified measurement far more than they can currently deliver it. Winterberry Group surveyed 121 enterprise marketers at the VP level and above, then published the results on September 30, 2026.
Eighty three percent called unified measurement a leading priority. Only nine percent said they could practice it extremely well. That’s a 74 point gap between ambition and ability, and it lines up with what we found when we looked at how many measurement methods marketing teams run at once: most teams aren’t missing a method, they’re juggling four or five of them without a way to make the numbers agree.
Here’s what I’d check first if I managed a measurement budget this quarter: not which new method to add, but whether anyone on the team can explain why two of the methods you already run disagree with each other. If nobody can answer that, one more test just gives you a sixth opinion instead of a clearer one.
What Should Marketing Teams Do With This Measurement Gap?
Start by reconciling what you’ve already got before you buy anything new, especially if your team is already living the problem we described in channels your mix model can’t see.
- Audit every measurement method you’re running and write down where two of them disagree on the same channel.
- Run one small holdout or geo incrementality test on your highest spend channel before you claim the method works everywhere.
- Label every number you report upward with the method that produced it, not just a final figure.
- Treat platform native reporting as a directional signal, never as the sole justification for a budget decision.
None of this needs new software. It’s not complicated. It needs someone willing to say a number’s wrong even when it’s the easy number.
FAQ: Incrementality Testing and Marketing Measurement
What is incrementality testing in marketing?
Incrementality testing measures how much of a result, like sales or sign ups, would have happened anyway without a specific ad or channel. It works by comparing a holdout group against a group that saw the campaign, so it isolates real causal lift instead of crediting a conversion to whichever touchpoint happened to show up last. DoubleVerify’s survey didn’t include it among its listed methods.
Why do marketers still use last click attribution?
It’s free, already built into most ad platforms, and simple to explain to a budget committee. DoubleVerify’s 2026 survey found 35% of marketers still rely on it despite wide agreement that it overstates the channel that happened to close the sale. Convenience, not accuracy, keeps it in use.
What is marketing mix modeling and how common is it?
Marketing mix modeling uses statistical models on historical spend and results to estimate each channel’s contribution without tracking individual users. DoubleVerify’s survey found 48% of marketers use it, the second most common method behind platform native reporting at 55%.
What is the gap between measurement priority and measurement ability?
Winterberry Group’s September 2026 survey of 121 enterprise marketers found 83% call unified measurement a leading priority, but only 9% say they can practice it extremely well. That 74 point gap shows most teams know what they want from measurement and haven’t built the process to get there yet.
Sources: DoubleVerify’s path to purchase survey (November 18, 2026), covered by PPC Land, with more at doubleverify.com. Winterberry Group’s unified measurement survey (September 30, 2026), see winterberrygroup.com. MarTech Cube’s analysis The End of Attribution Theater (November 20, 2026).