If your Target CPA or Target ROAS campaigns have looked worse since last August, the instinct is to blame creative fatigue, a new competitor, or your targeting going stale. For a lot of accounts, none of that is what happened. Google changed how budget-constrained bid strategies behave between August 17 and August 27, 2026, and the accounts feeling it aren’t broken. They’re finally showing you the number you actually told Google to hit.
Quick Summary
Google’s bidding target optimization change rolled out between August 17 and August 27, 2026, changing how Target CPA, Target ROAS, and Demand Gen Target CPC behave once a campaign is budget constrained.
Campaigns that quietly beat their target while marked “Limited by budget” now drift toward the actual target you set, so a higher CPA isn’t automatically a quality problem.
To confirm this is the cause, check whether the affected campaigns carried “Limited by budget” status before August 17, and whether the shift lines up with that window rather than a seasonal or creative shift.
Google isn’t auto-adjusting anything. A target you set six months ago stays wrong until you change it yourself.
Some accounts get more predictable scaling out of this, since budgets no longer need padding against a campaign quietly overachieving past what the business can support.
Confirm the Target CPA Timing Lines Up
Start with the calendar, not the dashboard. Google’s change touched Target CPA and Target ROAS on most campaign types, plus Target CPC on Demand Gen, rolling out globally between August 17 and August 27. Before this, a budget-constrained Target CPA campaign could spend only on its cheapest, best-converting opportunities, which let it beat its stated target. A $10 Target CPA campaign running at $5 actual CPA looked efficient, but that was partly a side effect of scarcity, not proof the target was well calibrated. After the update, that same campaign optimizes toward the $10 target it was actually given.
So the first diagnostic step is simple: pull your CPA or ROAS trend line and look for a step change that starts inside that ten-day window, not a slow drift that began in July or a jump that only shows up in September. If the shift predates August 17, this update isn’t your explanation.
Check Which Campaigns Were Actually Budget-Constrained
This change only touches campaigns that were hitting a budget ceiling. Pull up your campaigns from late July or early August and check which carried a “Limited by budget” status. Those are your candidates. A campaign with budget headroom the whole time should show almost no change, since old and new behavior converge once budget isn’t the binding constraint.
Manual CPC, Target Impression Share, and Target CPM are untouched entirely, so if performance moved there too, you’re looking at a separate issue. Target CPC only changed for Demand Gen. If the campaign showing the shift wasn’t budget constrained and wasn’t running one of these specific strategies, this update is very likely not what you’re dealing with.
Separate This From Seasonality and Creative Fatigue
Even when timing and budget status line up, don’t stop there. Run the same period against last year to rule out a normal seasonal cost curve; late August into September carries its own auction pressure in plenty of verticals. Check frequency and CTR separately for a creative fatigue signal, since a tiring ad set shows its own decline that this bidding change didn’t cause and won’t fix. Look at auction insights for impression share lost to rank, which points at competitors bidding harder rather than your target behaving differently.
None of these rule each other out. Plenty of accounts will find two things stacking in the same three weeks. Treat the bidding change as one input to confirm, not the automatic answer the moment the dates match up.
Decide What the Target Should Actually Be
Once you’ve confirmed the bidding change is a real contributor, the fix isn’t reflexively lowering the target back toward your old actuals. Google’s default suggestion in the Target Adjustment tool is to match recent actual performance, a reasonable floor, but it just recreates the old artifact under a new number. A better move prices the target from your actual unit economics: what CPA you can afford at your current margin and average order value, independent of what the campaign happened to spend before.
If budget was the real limiter and the business can support more volume at that true target, this is also the moment to raise the budget rather than the target. A higher budget on an aggressive target used to risk runaway spend past what you could justify; now that campaigns hold closer to the number you set, added budget behaves more predictably.
Watch for the Accounts This Actually Helps
It’s worth naming the upside, since most coverage of this change treated it as a problem to defend against. If your reported CPA was quietly subsidized by budget scarcity for months, you were scaling decisions off a number that wasn’t real. Finding that out now, heading into Q4 planning, beats finding it out after committing next quarter’s spend to a target that was never achievable at volume.
This mostly matters if you’re running Target CPA or Target ROAS with a genuine budget constraint and enough weekly conversion volume [Link: How the Conversions API Recovers Signal Loss], roughly 30 to 50 conversions a week or more, for the target to carry statistical meaning. Below that, this shift likely gets lost inside your normal week-to-week noise. If you’re already on Maximize Conversions without a set target, or on Manual CPC, none of this applies to your account directly.
Build a Habit of Catching This Kind of Change Early
Google and Meta both ship changes like this several times a year, usually as a support-page update, not a headline. A team reviewing bidding performance monthly can burn two or three weeks of spend reacting to a stale target before tracing it to a specific platform change. That lag is exactly what continuous account monitoring closes, catching a step change within days instead of a full billing cycle later. Maino.ai has reduced Customer Acquisition Cost (CAC) by an average of 46% across its client portfolio, and a meaningful share of that comes from catching platform-side shifts like this one before they compound into a quarter of wasted spend.
Our take: this update is Google correcting a rounding error that a lot of advertisers had started reading as skill. A budget-constrained campaign beating its target was never proof the target was right; it was proof the campaign couldn’t spend enough to test it. Now it can. The right response is a short diagnostic before you touch a single number: confirm the timing, confirm the budget status, rule out the obvious confounders, and only then decide whether the target itself needs to change. If you skip straight to panic-lowering your target, you’re just rebuilding the same artifact one dollar cheaper.
Frequently Asked Questions
What is the Google Ads bidding target optimization change?
It’s an update Google rolled out globally between August 17 and August 27, 2026, that changes how budget-constrained Target CPA, Target ROAS, and Demand Gen Target CPC campaigns behave. Instead of letting a tight budget cause a campaign to beat its stated target, the campaign now optimizes more consistently toward the actual target you set.
Which bid strategies does the August 2026 change affect?
Target CPA and Target ROAS across most campaign types, and Target CPC within Demand Gen. Manual CPC, Target Impression Share, Target CPM, and campaigns without a real budget constraint are unaffected.
Why did my Target CPA go up right after August 17?
If your campaign was marked “Limited by budget” before the change, it was likely spending only on its cheapest converting opportunities and beating its target as a side effect of scarcity. Once the update landed, that same campaign began optimizing toward the full target instead, which shows up as a higher reported CPA even though the target itself didn’t move.
Does Google automatically adjust my bidding targets after this change?
No. If the number you set no longer matches your current margin or business goal, it stays exactly as you left it until you update it yourself in the Target Adjustment tool or campaign settings.
How do I tell if this change caused my performance shift instead of seasonality?
Check whether the affected campaign carried a “Limited by budget” status before August 17, and whether the shift began inside the August 17 to August 27 rollout window. If both line-up and frequency, CTR and auction insights show no separate creative or competitive shift, the bidding change is very likely the cause.
Should I switch off Target CPA or Target ROAS because of this update?
Not by default. Switching strategies just trades one set of assumptions for another. Review actual performance across at least one full conversion cycle first, then decide whether the target needs to be repriced based on your unit economics or whether the strategy itself no longer fits your budget shape.