Google stopped letting your campaigns beat their target

What actually changed

Until 17 August, a campaign carrying a Limited by budget status could beat its stated target. Google's documentation describes this as campaigns that may be overperforming on bidding targets, with performance fluctuating when budgets are adjusted. After that date those campaigns perform more consistently toward the target instead.

The mechanism is not mysterious. A budget cap forces the system to be selective: it can only buy so many auctions, so it buys the cheapest ones that convert. That selectivity produced a cost per acquisition below the stated target as a side effect, not as something anyone asked for.

Google's stated reason is predictability. The old behaviour made budget changes hard to reason about, because raising a budget released the constraint and performance moved in ways nobody had predicted. That is a fair description of a real problem.

It is also a change that takes something away, and the documentation does not put a number on what.

The arithmetic nobody is running

Take Google's own example and finish it. A campaign with a $10 target delivering a $5 actual cost per acquisition moves toward $10. That is not a small drift. It is the cost of a lead doubling, in an account where nobody changed a setting.

Now hold the budget still, because a budget-limited campaign by definition has no room. Spend is capped at the same number. If each acquisition costs twice as much, the same spend buys roughly half as many.

So the shape of the change in a fixed-budget account is: spend flat, cost per lead up toward the target, lead volume down. Not because performance got worse, but because the system is now doing what the target told it to do.

This is our reading of the mechanism rather than a figure Google publishes. What Google does publish is the direction and the example, and the volume consequence follows from holding budget constant. It is worth checking against your own numbers rather than taking ours.

Who is exposed, and it is not who you would guess

Exposure is not about spend level. It is the gap between the target you set and what the campaign has actually been delivering. A big account with carefully maintained targets has almost no exposure. A modest account with a round-number target set eighteen months ago has all of it.

Mid-market lead generation sits badly here for two reasons that compound. Budgets are small enough that Limited by budget is a chronic state rather than an occasional one. And targets in these accounts are very often inherited: a round number someone picked at setup, or a figure carried over from an agency transition, which then quietly beat itself for a year while everyone read the reported CPA and was pleased.

That is the uncomfortable part. A target that has been beaten by half for a year was never functioning as a target. It was a ceiling nobody was near, and the account was being run by the budget cap. This change converts that ceiling into an instruction.

The tell is simple: open the campaign, compare the target against the last ninety days of actual cost per acquisition, and the size of that gap is the size of your problem.

What is not affected

Campaigns that are not budget-constrained do not change behaviour. Google is explicit that these already scale in line with the stated target and will continue to. The change also does not touch manual CPC or Target Impression Share, and it is a bidding change rather than an auction change.

It applies to Target CPA and Target ROAS across most campaign types, and to Target CPC for Demand Gen. App campaigns, Video reach and Video view campaigns keep the previous behaviour. Hotel and Display campaigns already worked the new way.

If your campaigns are managed through Search Ads 360 the adjustment tool does not appear in Google Ads at all, which is the kind of detail that turns a fifteen-minute task into a confused afternoon.

What to do, in order

Find the gap first, then decide what the target should actually be, then decide whether you are willing to pay it. Google has shipped a Bid Target Adjustment Tool for this, reachable from the Review your campaign targets notification or from the campaign's bidding settings.

Start by listing every campaign with a Limited by budget status on a target-based strategy, with its target beside its trailing actual. That list is the whole exposure, and in most accounts it is shorter than people fear.

Then answer the question the old behaviour let you avoid: what is a lead actually worth? A target set to what the campaign has been delivering preserves current economics. A target set higher is a decision to buy more expensive leads, which is sometimes right, and should be a decision rather than a default.

If the budget genuinely cannot move, Google's own recommendation is worth reading twice: switch to Maximize conversions or Maximize conversion value and accept fluctuation when budgets change. That is the platform suggesting you stop using a target at all, which is a reasonable answer to a fixed budget and an odd thing to arrive at from a change sold as better target control.

And do the thing this whole episode is really about. A target that has not been revisited since it was set is not a target, and the same is true of the conversion it is optimising toward. If that conversion is a raw form fill, the target was never denominated in anything the business cares about, and resetting the number will not fix that.

What this means for an operator

List every Limited by budget campaign on a target-based strategy with its target beside its trailing ninety-day actual. That gap is your exposure. Reset the target to what the campaign has been delivering unless you have a reason to pay more.