The Google Ads Bidding Change on 17 August: What It Actually Does to Your Targets

On 17 August, Google changes how Smart Bidding behaves in campaigns that are limited by budget. If you run Target CPA or Target ROAS and one of your campaigns has quietly been beating its target, this Google Ads bidding change takes that advantage back. Not as a penalty — as a deliberate design decision. Here is what actually changes, who it affects, and what to do with the days you have left.

What is actually changing

Today, a campaign sitting in "Limited by budget" with a target-based bid strategy can overperform its stated target. The system has room to be more efficient than you asked it to be, and plenty of accounts have been living off that gap for years without realising it.

From 17 August 2026, budget-limited campaigns using target-based bid strategies will optimise more consistently toward the target you set, including when you adjust budgets. Google's own worked example is blunt: a campaign with a Target CPA of $10 that has been delivering an actual CPA of $5 will start delivering closer to $10.

Read that twice if you have ever set a target loosely. A "safe" number you typed in eighteen months ago and never revisited is about to become the figure the system genuinely aims at.

Who this hits

The change applies to Target CPA and Target ROAS across Search, Shopping, Performance Max, Demand Gen and Travel campaigns in Google Ads and Search Ads 360, plus Target CPC for Demand Gen and Demand Gen campaigns in Display & Video 360.

It does not touch manual CPC, Target impression share or Target CPM. App campaigns, Video reach and Video view campaigns keep the old behaviour. Display and Hotel campaigns already work this way.

The critical qualifier: campaigns that are not constrained by budget are unaffected. If your campaigns have never worn the "Limited by budget" label, you can stop reading and get on with your day. If they have — and Google is notifying any advertiser whose campaigns were budget-limited at any point in the last twelve months — keep going.

The honest reading

For a slice of advertisers, doing nothing means a real efficiency loss. Your CPA rises toward the target you set. Your volume rises with it. Whether that is a good trade depends entirely on whether the extra conversions are worth the extra cost, which depends on your margin, and only you know that number.

The old behaviour was not free either. It is the reason so many advertisers have watched a campaign fall apart the moment they increased its budget — the system was never truly holding to the target, so scaling produced results nobody could forecast. After 17 August, the target becomes the control and the budget becomes the throttle. That is a better model to manage against, even if the first fortnight of it stings.

What to do before 17 August

Filter by status first. Pull the campaigns that are both "Limited by budget" and running a target-based strategy. Everything else can be ignored.

Compare actual to target over a real sample. Not seven days. Google recommends evaluating Smart Bidding over a period containing at least 30 conversions, or 50 for Target ROAS. If a campaign cannot produce that in a month, its targets were guesswork already and consolidation is the bigger conversation.

Then take one of four positions:

• Keep the target, and accept that performance drifts toward it — more volume, less efficiency.
• Match the target to recent actual performance, which holds your current numbers roughly where they are.
• Set a custom target between the two, based on what you can genuinely afford to pay for a customer.
• Switch to Maximise conversions or Maximise conversion value, which spends the budget and lets efficiency float.

Google's Bid Target Adjustment Tool has been live in accounts since 6 July and shows historical performance alongside a recommended figure. It will not generate a recommendation for campaigns with fewer than seven conversions, which is itself a fair warning about those campaigns.

What we would do in most accounts

Match the target to recent performance, then raise the budget. It protects the efficiency you have already proved you can hit, and it puts you in the position this change was designed to reward: a campaign that can scale at a known target rather than one that looks efficient only because it is starved.

Switching everything to Maximise conversions is the move we would push back on hardest. It feels safe because your spend stays fixed, but you are surrendering the one control that matters when you have a real cost-per-acquisition ceiling. If you have a margin floor, keep a target.

One more thing worth flagging. Google has said Performance Planner forecasts may be inaccurate between 17 and 31 August while the system settles. If you were planning to build a Q4 budget case off those numbers, wait a fortnight and pull them again.

The part nobody else will mention

If your only lever is the bid target, you are negotiating with an auction you do not control. The other lever sits entirely on your side of the click. A landing page converting at 3% rather than 2% lets you afford a target roughly half again as high, and that arithmetic owes nothing to any Google announcement. If you have been treating conversion rate work as a nice-to-have, this is the month it stops being one.

Account structure matters for the same reason. Consolidating thin, budget-starved campaigns under shared budgets or portfolio bidding can lift them out of the constrained state altogether, and this change only bites campaigns that are constrained. Fewer, better-funded campaigns were the right answer before 17 August. They are more obviously the right answer after it.

If you want a second pair of eyes on your targets before the deadline, or you would like to see how we approach Google Ads management and what that has looked like across our client work, get in touch. It is a short conversation, and the date is not moving.