Google Just Ended Target Overperformance. Here's What It Costs B2B Teams That Were Quietly Beating Their CPA.

If your Google Ads account has been quietly beating its Target CPA or Target ROAS for the last few months, you weren't necessarily doing anything differently. You may have just been budget-limited - and Google changed that on August 17, 2026.

What "Target Overperformance" actually was

On a budget-limited campaign using Target CPA or Target ROAS bidding, the daily budget cap - not the target itself - was often the real constraint on spend. That let campaigns deliver results better than their stated target: a campaign with a $10 Target CPA could quietly land a $5 actual CPA, because the budget ran out before the algorithm needed to spend up to the less-efficient clicks it would otherwise have bid on to hit exactly $10.

That looked like good performance in a dashboard. It was also, in Google's own framing, the budget cap doing a job the target was supposed to do. Starting August 17, 2026, Google's bidding system began automatically adjusting budget-limited campaigns so their results track more closely to the original stated target instead of continuing to beat it. (Search Engine Journal, August 2026)

In plain terms: if your Target CPA is $10 and your recent actual CPA has been $5, expect your actual CPA to drift back up toward $10 as the system stops constraining spend the way it used to.

Which campaigns are affected

The change applies specifically to budget-limited campaigns running target-based bid strategies - Target CPA or Target ROAS - across Search, Shopping, Performance Max, Demand Gen, Travel, Search Ads 360, and Demand Gen in Display & Video 360. (Google Ads Help) If a campaign isn't budget-limited, or isn't on a target-based strategy, this change doesn't touch it directly.

Google gave advertisers some runway: a Bid Target Adjustment Tool was introduced on July 6, 2026, giving teams a window to review and adjust affected campaigns' targets before the August 17 enforcement date. If your team didn't use that window, the adjustment is now happening automatically, on Google's terms rather than yours.

Why this matters more for lean B2B teams than it sounds

A lean team running a handful of always-on demand gen campaigns is more likely to be quietly exposed to this than a large team running dozens of tightly managed campaigns, for two reasons. First, lean teams are more likely to set a conservative Target CPA once and leave it, rather than actively re-tuning targets every month - which is exactly the condition under which budget-limited overperformance builds up unnoticed. Second, a CPA that drifts back toward target looks, at a glance, like a sudden drop in campaign quality. It isn't. It's the removal of a subsidy your account didn't know it had.

That distinction matters for how you report the change internally. If a stakeholder sees blended CPA rise in September and assumes the campaigns got worse, the conversation goes sideways fast. If you can point to this specific, dated Google mechanic, it's a five-minute explanation instead of a defensive scramble.

A checklist for this week

Audit your budget-limited target campaigns now. In Google Ads, filter for campaigns on Target CPA or Target ROAS that have been budget-limited (not bid-limited) over the last 90 days. These are your exposure list.

Check whether recent actual performance was beating target. For each campaign on that list, compare actual CPA/ROAS to the stated target over the last 30-60 days. A meaningful gap - actual materially better than target - is a strong signal that campaign is about to see its numbers move.

Decide deliberately: loosen budget, tighten target, or accept the shift. You have three real options. Raise the daily budget so the campaign can keep spending at the efficient rate it found. Lower the Target CPA (or raise Target ROAS) to formally lock in the efficiency you were getting informally. Or do nothing and accept a worse blended number - which may be the right call if that campaign was never going to scale further anyway.

Update your forecast before the numbers move, not after. If any of your exposed campaigns feed into a monthly pipeline or CAC forecast, flag the expected CPA drift to whoever owns that number now. A known, explained variance is a footnote. An unexplained one is a credibility problem in your next pipeline review.

The broader point

Google Ads bidding mechanics change often enough that most teams stop reading the fine print - but this one has a specific, dated trigger and a specific, measurable effect on blended CPA. A 20-minute audit this week is cheaper than a confusing conversation about "why did paid performance suddenly get worse" in next month's report.

Nukipa runs paid, organic, and AI-search demand as one connected system for lean B2B teams, so changes like this get caught and explained before they show up as a surprise in a board deck. If your paid media forecasting needs a system behind it instead of a spreadsheet, test Nukipa.