The Google ceiling you can't optimize your way out of

Every paid search lead has sat with this account. Performance Max and Brand Search are running clean. ROAS is respectable. The team has pulled every lever available. New asset groups. Tighter audience signals. Adjusted bidding. Still, the number won’t move quarter over quarter. The natural conclusion is that the account has hit its ceiling. That conclusion is usually wrong. It’s an expensive place to stop looking.

Google built PMax and Brand Search to capture demand. Both win auctions using signals that already exist. Think a branded search, a remarketing list, or someone who has already half decided your brand is worth considering. In 2023, Google folded its Discovery ad format into what it now calls Demand Gen campaigns. This extended automated buying across YouTube, Gmail, and Discover. In doing so, Google effectively split its own advertising ecosystem into two separate jobs. One set of channels builds awareness before anyone searches. The other converts the search once it happens. Most advertiser budgets only staff the second job. Then they wonder why the first one never grows.

Why the blind spot is structural

Part of the reason so many accounts stop at PMax and Search isn’t neglect. It’s visibility. Search marketers have criticized PMax since its 2021 rollout for collapsing several campaign types into a single automated system with limited channel-level reporting. You can see that the campaign converted. You often can’t see what warmed the customer up three touchpoints earlier. A channel you can’t see clearly is hard to defend in a budget meeting. So upper-funnel spend gets treated as optional exactly when it’s most load-bearing.

Demand creation and demand capture aren’t competing for the same job

The reasonable-sounding fear is that adding another channel just spreads the same budget thinner. Two mechanics explain why that’s usually not what happens. A broader mix gives Google’s own systems more signal and more touchpoints to learn from across a customer’s path to purchase. Upper and mid funnel activity builds familiarity with a brand before someone ever opens a search bar. By the time that person reaches a PMax or Brand Search auction, they convert at a lower cost. They arrived already warm.

That’s a different mechanism to the one most reporting is built to measure. It’s also a large part of why long-running research on marketing effectiveness, including the IPA’s work by Les Binet and Peter Field, has argued for years that brand-building and short-term activation aren’t really rivals for the same pound. They compound each other on a lag that most dashboards aren’t set up to show.

What the data from one advertiser cohort adds

Fospha’s Full-Funnel Google Report tracked Q4 2025 performance across a cohort of retail ecommerce advertisers. It found a pattern consistent with that theory. Accounts that grew Demand Gen and YouTube spend alongside their existing PMax and Brand Search budgets, rather than instead of it, saw the ROAS on those existing channels move up, not down, as the new spend matured. A comparable group that pulled back on upper-funnel spend saw the opposite. Read against the mechanics above, that’s a reasonably intuitive result. It’s harder to capture demand that was never created.

Where this goes wrong in practice

The most common mistake is treating a new channel as something to fund by cutting an existing one. If Demand Gen gets its budget by taking a slice out of PMax, the account isn’t testing diversification. It’s just moving the same demand-capture dollars around and calling it something new. The brands seeing a real lift grew every channel together. That’s a harder budget conversation to have internally than “we’re reallocating.” But it’s the conversation the mechanics actually require.

The question worth asking before the next optimization sprint

If an account has been optimized hard and the number still isn’t moving, a better bid strategy inside that same channel probably isn’t the fix. That channel is already converting everyone it can reach. The real question is whether anything upstream is creating new people for it to convert. That’s a less comfortable question than tweaking a campaign. It usually means asking finance for budget on a channel that won’t show a clean ROAS of its own for a few months. But an account that’s stopped growing on its most efficient channel was never actually stuck. It was starving.