‘It’s the worst execution I’ve seen’: Confessions of a marketer on pulling every client off Google’s PMax

In a response to a request for comment, a Google spokesperson claimed that PMax was used by over 1 million advertisers across Google’s channels. “In 2024, we launched more than 90 quality improvements in Performance Max that increased conversions and business value for advertisers,” they said in a statement.

In this edition of our Confessions series, where we trade anonymity for candor, a 20-year growth marketing veteran explains why they’ve advised clients walk away from PMax, and will continue to do so — having already shifted aggregate budgets in the million double digits away from the platform.

This conversation has been edited and condensed for clarity.

What have you thought of Pmax since it launched to all advertisers in November 2021?

Initially I loved the idea of Google’s Performance Max, because they have the data and information available to make the best decisions. The problem is in its execution. For me, it’s the worst execution I’ve seen from a big player, ever.

Why do you think that?

I’ve been in this industry since 2000. I’ve worked in-house at brands, as a consultant, and as part of a number of agencies. So I’ve seen what typically works, what doesn’t and can generally connect the dots.

When it first launched, PMax was a total black box. My team saw great results on the platform, but on our end, nowhere near those results. To give you an example, I’ve worked with multiple big brands which were also fortunate enough to have budget to also invest in incrementality providers. And we test all of our channels for incrementality. 

To give you some context, non-branded search incremental revenue is typically in the 80% to 90% range. Google Shopping is typically between 70% and 80%. But the first test on PMax for one of the larger brands I’ve worked with, incrementality came back less than 10%. In other words, it didn’t produce incremental revenue. It took the budget, but clients actually got a lot less bang for their buck.

Some fashion brands I’ve worked with over the years, for example, which had invested 80% in Google Shopping then shifted to PMax — because they were told PMax uses the same products they had, but PMax on the whole is better — barely produced 50% of less incremental revenue than they were making before.

And it’s not just me seeing these types of results. I had a meeting two weeks ago with an exec from an incrementality provider I work with and they said that out of 100 plus tests they had conducted, 75% of them came back with awful results.

So this isn’t just a recent problem?

Not at all. From what I’ve seen, it’s been a long-standing problem a number of my clients have experienced since PMax launched.

If this has been such an ongoing issue, what’s been Google’s response to it?

Every single time, Google has told us we are wrong, that we have no idea what we’re talking about. To them, PMax is the best thing since sliced bread.

So have they not offered you any support at all to rectify the issue?

Quite the opposite. What actually started was almost warfare from Google. Each time we told our reps that we were switching back to Google Shopping and coming out of PMax completely with each client, they challenged us. Which makes sense. They wanted to know what tests we did, did we even set up the test correctly, how do we approach data — it was like an inquisition. Because in their eyes, we were wrong, and that was it.

When we didn’t agree with what they were claiming about the product, reps have gone behind our backs, talking to our clients teams, right up to the CEO, claiming we’d made the wrong decision on their behalf and they’ll regret leaving money on the table. 

And this happened every single time?

Yes, unfortunately. This has happened multiple times.

What if they brought in an expert in the areas where you’ve had issues?

There was one instance in early 2024, where Google flat out didn’t agree with what we were choosing to do. So they brought someone from their measurement team over because they were so sure we were wrong. In this instance, the guy who joined the meeting had just joined Google two months prior from an incrementality provider. And the Google reps indirectly implied that he’s an expert and will find out what we’re doing wrong, which is the reason PMax isn’t working for us. 

This guy asked me and my team every possible question he could ask about every type of test we could do, which would (in Google’s eyes) “prove” we’d made a mistake. But I was able to challenge him on everything with facts and figures. In the end, he just said there’s nothing else he can advise or do to help because we’d done everything we could.

It was clear that the Google rep was so disappointed that they couldn’t even support their message that PMax was the better way of working. That in itself was really telling.

So with not much help, what have you done to try and rectify the issue?

I moved all my clients from PMax back to Google Shopping, and have done since — we’re talking at least a dozen. And as soon as I did, their results immediately got better.

Define better.

ROAS went up by 20% while the cost per new customer reduced by an equivalent amount, in almost every situation.

Has this ultimately changed what you think about Google?

Yes and no. I like some of Google’s products, I think Google Shopping is a great product — which is why I’ve switched my clients back to it. But since about 2020, it has felt like Google’s attitude has switched to a more arrogant and condescending attitude, where Google is right and we are wrong. And I’m not the only one who feels this way.

The problem is, their supposed help just hasn’t felt like support at any point. It feels like there is an agenda behind it. 

Agenda in what way?

I got confirmation in June when I met with a Google rep that they don’t push PMax in the same way they used to anymore, it’s more about the wider product portfolio. But he did admit that for a long time the team was incentivized to push the product out. There were bonuses for the staff who could get their clients on PMax. So that explains the push.

If you have any insights to share about Google, or any other platform, or if you’re a current or former employee of any of the platforms, contact Krystal Scanlon securely using a non-work device on Signal: @krystalscanlon.27, email: krystal@digiday.com or via LinkedIn.