If your paid media team hasn't touched Target CPA or Target ROAS settings since July, you've probably already seen the damage in your August close. Campaigns that quietly delivered $35 conversions against a $50 target are now delivering $48 conversions, and nobody changed a single setting. The algorithm did exactly what it was told. The problem is that nobody updated what it was told to do.
Google's documentation frames the change as a predictability improvement: budget-limited campaigns using target-based bid strategies will now "more consistently perform toward your bid target." That's accurate, but it buries the operational consequence. If your stated target was always looser than your actual performance, the system will now spend up to that looser number. Your CFO will notice before your media buyer does.
The Mechanic That Changed
For years, Smart Bidding treated budget constraints as an implicit efficiency signal. A campaign capped at $100 per day with a $50 Target CPA didn't just limit spend; it forced the algorithm to cherry-pick only the cheapest, highest-intent conversions available within that dollar limit. The result was a misleading baseline. A campaign "beating" its target by 30% wasn't necessarily well-optimized. It was simply buying a narrow slice of inventory because the budget cap prevented it from bidding on anything else.
PPC Hero's analysis captures the downstream problem: when advertisers raised budgets on these campaigns, performance destabilized. The algorithm suddenly had to re-evaluate broader auction pools it had never learned, causing CPA spikes and extended re-learning periods. Budget and efficiency were entangled in ways that made scaling unpredictable.
After , Google decoupled those levers. Budget controls spend. Target controls efficiency. The algorithm now optimizes strictly toward the number in the box, regardless of whether the campaign is constrained or fully funded. That's cleaner, but it means every target you set is now a commitment, not a ceiling the system might quietly beat.
Who Got Hit
The scope is narrower than the panic suggests, but it covers the campaigns that matter most to B2B pipeline. Optmyzr's breakdown confirms the affected campaign types: Search, Shopping, Performance Max, Demand Gen, and Travel, running Target CPA, Target ROAS, or (in Demand Gen only) Target CPC. Display and Hotel campaigns already operated this way. App, Video reach, and Video view campaigns are explicitly exempt.
Performance Max deserves special attention. Because PMax distributes budget across multiple channels internally, the change doesn't just affect cost per conversion. It can shift where conversions come from. A campaign that was quietly over-indexing on Search because the algorithm found cheap inventory there may now redistribute toward YouTube or Display to hit the stated target across a broader mix. If your attribution model weights channels differently, your reported pipeline quality may shift even if total conversion volume holds steady.
The Audit You Should Have Run in July
Google shipped a Bid Target Adjustment Tool on , six weeks before the deadline. Digital Applied's playbook walks through what it surfaces: recent actual performance, the currently entered target, and a recommended target calculated from that recent performance. One button applies the recommendation.
If you didn't use it, you can still run the audit manually. Pull the last 90 days of conversion data for every budget-limited campaign using a target-based strategy. Compare actual CPA or ROAS to the stated target. Any campaign where actual performance beat the target by more than 15% is a candidate for immediate adjustment.
The decision isn't automatic. Some of that headroom was intentional. A $50 Target CPA on a campaign delivering $35 might have been set that way to give the algorithm room to chase incremental volume during peak periods. If that was the strategy, document it and leave the target alone. But if the gap exists because nobody updated the target after the campaign matured, you're now paying for that oversight.
The Recovery Playbook
Three moves, in order of urgency.
First, reset targets to match recent performance on any campaign where the gap was unintentional. Use the Bid Target Adjustment Tool if it's still surfacing recommendations, or set targets manually based on your 90-day actuals. This stops the bleeding.
Second, re-examine budget constraints. The old behavior rewarded keeping campaigns artificially capped because it forced efficiency. That incentive is gone. If a campaign can profitably absorb more spend at its true efficiency level, raise the budget. If it can't, the constraint was masking a targeting or creative problem, not solving it.
Third, stress-test your forecasts. Miraflow's post-change analysis notes that campaigns which looked efficient purely because of unused headroom will now report efficiency that matches their actual creative and targeting quality. If your Q4 pipeline forecast assumed August's inflated efficiency would continue, you need to rebase. Pull the last three weeks of post-change data, recalculate your blended CAC, and update your models before the next board review.
The Upside Nobody's Talking About
Most coverage of this change focuses on lost efficiency. That's the wrong frame. What Google actually fixed is the scaling problem that made budget increases feel like rolling dice.
Before , raising budget on a constrained campaign meant watching the algorithm lose its footing while it re-learned broader auction pools. After , the algorithm already knows what efficiency level you're targeting. Budget changes affect volume, not efficiency. That's a cleaner operating model for anyone trying to scale paid acquisition predictably.
PPC Hero frames it well: Google is uncoupling the budget lever from the efficiency lever. For operators who actually set targets based on unit economics rather than hoping the algorithm would beat them, this is a net positive. The campaigns that suffer are the ones where the target was always a fiction.
The Board Conversation
If you're presenting paid media performance this month, lead with the mechanism change, not the results. A 20% CPA increase in August looks like a failure until you explain that the prior baseline was an artifact of how Google's bidding system handled budget constraints, not a reflection of true campaign efficiency.
The right framing: "Our stated targets were looser than our actual performance. Google's system now enforces those targets. We've reset targets to match our true efficiency, and forward performance will reflect that reset." That's a one-time adjustment, not a trend. If your CFO asks why the targets were wrong in the first place, the honest answer is that the old system rewarded leaving them loose. The new system doesn't.
Set your targets like you mean them. The algorithm will take you at your word.
