I am only semi kidding with calling this “vibes attribution.” Let’s downplay ROAS and trumpet incrementality as some new and better attribution philosophy…
See this tweet and embedded video from Google Ads from late last week (you can click on embedded video and it will play here or click Show more and get to the full tweet over on Twitter or read full text at end of this post for those not on Twitter):
Get ready to start hearing about incrementality a lot more. And I want to be clear, I don’t think incrementality as a concept or something to test is at all a bad thing. But it is at least kind of a “vibes” based metric. And because of that, where I am having some difficulty is with the notion that is should replace or supersede our more traditional methods of attribution or determining if paid digital ads are “working” or not.
Because one of the biggest selling points of doing digital ads, since they began, has been that you can track with a lot more detail and certainty which dollars spent with which messaging shown to which audiences on which platforms actually resulted in sales or leads being generated. That has always been a very powerful reason for businesses to spend on digital marketing. It was a key advantage over what are now considered more legacy advertising networks (print, TV, outdoor, etc.).
And let’s be totally real, attribution as we have known it and sold it has been degrading for the past 5 years. More and more data becomes hidden and we are provided a lot less direct data and a lot more modeled data for conversions. It has not been great, but it has still held reasonably close to the model we and businesses who advertise have known and accepted for the past 20+ years.
But now, it feels like Google Ads is getting ready to really push this idea of incrementality on us and by extension the brands we work with. And I’ll be honest, I don’t like it. It is a whole lot of “trust us” from a platform that has not done a ton to earn trust from advertisers lately. And yes, the decision to advertise in the digital space utilizes very different calculus than it was even 15 years ago when you had a distinct advantage if you did it when many of your competitors were not. Today, most of your competitors are also doing some type of digital advertising, so the landscape has certainly changed.
And I know that we can’t really expect this to change if Google decides this is the way forward. Removal of controls and available data has been moving in this direction for at least the last 5 years. But we had better figure out how we, as PPC professionals, want to talk about this and how we can explain it to clients and stakeholders. Because this does represent a big shift in digital advertising. Even if it isn’t presented quite that way.
I have a hard time envisioning businesses jumping on board the incrementality vibes train and leaving their beloved ROAS or CPA behind. And I don’t think they should leave those stronger metrics behind. But Google might keep telling us and them that they should…
I will definitely be including incrementality on my 2026 Google Marketing Live bingo card…
———————————————————————————–
If you are not on Twitter, here is the full text of the Google Ads tweet too:
“If you aren’t testing for incrementality, you’re likely paying for customers who were going to buy from you anyway. ROAS is often a “real-time gauge” that reflects correlation – it tells you what happened, but it doesn’t prove your ads caused it. To find that truth, you need to shift your marketing culture from assumed credit to causal proof. As Gaurav Bhaya (VP of Ads & Analytics at Google) and Patrick Gilbert (CEO at
@adventureppcinc) emphasize, you need to upgrade your technology to move beyond defensive reporting and start reporting offensively. Testing turns marketing costs into growth investments. Calibrate your data to invest with data-backed confidence. Here’s how:
Move beyond the real-time gauges: Attribution tells you what’s happening now, but causal gauges like incrementality help you understand the why behind your performance.
Calibrate for reality: Use incrementality as your ground truth to calibrate all your other gauges, ensuring your budget reflects actual causality rather than just data points.
Avoid emotional budgeting: Don’t cut budgets based on short-term anxieties; let a proper time horizon of data prove the effectiveness of your brand investments.
Lead with a true hypothesis: Define exactly what you expect to happen – like a lift in brand search – before you launch.
Run ongoing experiments: Running a high volume of experiments per year is how you prove marketing is a growth driver. By calibrating your models with incrementality, you stop guessing at ROI and start investing with mathematical certainty. How often is your team running incrementality tests?”
Leave a Reply