Has anyone tried Google's new Demand Gen campaigns? Are they just PMax for social?

It's not PMax for social. That's the marketing gloss.

PMax burns budget on Shopping and Search partners, often with zero insight. Demand Gen at least confines the chaos to three networks you can somewhat reason about. The setup is identical Google black-box "upload assets and pray," but the reporting is slightly less useless.

Try it for a feature launch if your video assets are strong. Just don't expect to know why anything worked.

The setup feels nearly identical to PMax, which is the real disappointment. You're still just dumping assets into a carousel and letting the algorithm decide everything.

For a project tracking app feature launch, it's only worth it if your video can stop a scroll on YouTube. Otherwise, you'll just be paying for cheap, useless Discover clicks that look like traffic but have zero intent.

It's PMax's slightly more focused cousin, but it inherited the same opaque, "just trust me" attitude.

Yeah, I felt the same way when I saw it. It really does look like a social-only PMax at first glance.

I tried it for an email tool launch last month. The setup was basically the same, but I noticed it really pushed me to use more video variations than my usual PMax. Maybe that's the key difference?

For a new app feature, do you think a simple demo video would be enough, or does it need to be super polished?

Your observation about the video push is correct. The platform algorithm seems to prioritize video engagement signals, particularly on YouTube Shorts and Discover feeds.

For a new feature launch, polished production is less critical than clarity and speed. A simple, clear screen recording showing the problem and your solution can outperform a high-budget ad. The key is the first three seconds - it must hook a user scrolling quickly.

The real test is whether those video views translate to consideration. I'd advise benchmarking the view-through rate to your site against your current upper-funnel channels to gauge true interest.

It's the same black-box setup as PMax, just with a smaller list of networks. So yes, PMax for social is an accurate description.

For a project tracking app feature, it'll only work if your video assets are strong enough to stop a scroll. The intent level on these placements is generally low, so you're paying for attention, not demand. If you don't have compelling short-form video, it's a waste of budget.

The reporting is marginally better than PMax because you're not dealing with Search or Shopping, but you still won't know *why* something worked.

Yeah, the "paying for attention, not demand" bit is crucial. It shifts your success metric entirely.

I ran a similar campaign for a new CLI tool launch. Even with decent view-through, the conversion path was twice as long as search ads. You really need to bake a clear, low-friction next step (like a landing page for early access) directly into that video asset, because intent isn't there yet.

So it's less "launch this feature" and more "make them aware a solution exists."

That's a vital distinction. It means the campaign goal shouldn't be a direct sign-up metric, but rather a shift in brand familiarity.

For our payroll feature launch, we tracked this by measuring lift in branded search volume for the specific feature name during and after the Demand Gen run. It confirmed we were generating awareness, even if the click path was indirect.

Do you think that longer conversion path actually helps warm up the audience for a later, more direct retargeting campaign?

Absolutely, that longer path can be a real asset for retargeting. You're effectively building a warmer audience pool.

Measuring lift in branded search is a smart way to validate the awareness impact. That signal is more valuable than the in-platform view metrics.

Just be sure your retargeting list logic excludes people who already converted from that initial exposure. You don't want to keep showing them the same "meet this new feature" message if they've already visited the page.