Google Ads' August 17 Bidding Change: What PPC Managers Must Do

Google Ads is changing how Smart Bidding works for budget-limited campaigns on 17 August 2026, and if you run Target CPA or Target ROAS you need to know about it. If you've had the email, seen the LinkedIn meltdown, or just noticed everyone in PPC getting twitchy, this is why. Campaigns that have been quietly beating their targets will stop. Google won't fix it for you. The fix takes ten minutes. You just have to actually do it.

Right now, if a campaign is "Limited by budget" and running Target CPA or Target ROAS, Smart Bidding has been cherry-picking the best auctions and quietly overdelivering. Your £35 tCPA has been hitting £22. Your 4x tROAS has been landing at 5.5x.

Lovely. Google now calls that unintended behaviour.

From 17 August, the system optimises more literally toward the target you actually typed in. That £22 drifts back toward £35. The 5.5x comes back down to 4x.

Same budget. Same campaigns. Higher cost per customer.

Google's own documentation uses a $10 target that's been delivering $5, and states plainly it will deliver closer to $10 from 17 August. So this isn't speculation. It's written down.

If you'd rather hear it from the horse's mouth first, Google Ads has published its own explainer walking through the change and how to keep performance predictable: Google Ads Bidding Update: How to Maintain Predictable Performance (2026 Guide). Worth ten minutes before you touch anything. Then read on for what they don't spell out quite so cheerfully.

Is This Even You? Who's Actually Affected

In scope (Target CPA, Target ROAS, plus Target CPC for Demand Gen): Search, Shopping, Performance Max, Demand Gen, Travel. Across Google Ads, Search Ads 360, DV360, Editor and the API.

Not affected: App campaigns, Video reach, Video view. Carry on. Same goes for any campaign not limited by budget, for the reasons covered up top. And the auction itself isn't changing. Google has been explicit, this is a bidding change only.

You'll see coverage claiming Display and Hotel are affected. Google's docs say those already run the new behaviour, so there's nothing new to do. We're going with Google on that one.

But Why Only Budget-Limited Campaigns?

Fair question. Because your other campaigns are already doing exactly this.

An unconstrained campaign already delivers in line with the target you set. That's the normal behaviour, and nothing changes for it on 17 August. Google says as much in its FAQ, "campaigns not limited by budget will carry on exactly as they are".

The over delivery only ever happened on the budget-limited ones, and here's the mechanism nobody's spelling out. When a limited budget can't fund every conversion, the system is forced to spend it on the very cheapest ones. That's not Smart Bidding being clever, it's Smart Bidding rationing. Strip the budget cap away and there's no cheap-only cherry-picking to "correct."

So this change isn't Google reaching into all your campaigns. It's Google taking the one group that got a quiet discount off the back of being starved of budget, and making them behave like everything else. Which, annoyingly, is a perfectly logical thing for them to do.

Yes, We Know How This Looks (and What Google Says Back)

Google's Ads Product Liaison Ginny Marvin has been out defending it, and credit where it's due, she hasn't dodged the main point. She's confirmed that budget-capped campaigns currently over-performing their targets will see performance align with those targets.

She's also been firm that this won't increase your spend. Budget caps still apply. She's pushed back hard on the idea that Google's advice is to "let the system spend more money."

Fine. Technically true. Not really the complaint though, is it?

Nobody thinks Google will blow past their daily budget. The worry is that the same budget now buys fewer, pricier conversions. That isn't extra spend, it's worse efficiency on the same spend. Different argument, still unanswered.

Kirk Williams asked the question most of us were thinking: is Google "building the system to literally choose to be dumber when limited by budget?"

Google's response is basically "no, the system still finds as many conversions as possible at the target you set." Which is a denial rather than an explanation. Make of that what you will.

Her clearest line, and the best one-sentence summary of the whole change, is this: on budget-constrained campaigns, "the target will now be your efficiency lever." That's the honest framing. The knob you set now actually controls the outcome, instead of being a loose ceiling the system undercut for you.

The honest reading, and the best one I've seen came from Mike Ryan, is probably this: that efficiency was never yours. If you asked for 5x and got 10x, Google doesn't owe you the 10x, and those profitable auctions go to advertisers who explicitly bid for them. Harsh, but coherent.

The real gripe isn't the motive anyway. It's that a setting you could safely ignore has just become a decision you have to make. A target that was quietly overdelivering didn't need babysitting, because the system was beating it for you. Now every one of those numbers is load-bearing.

A management burden, handed to you, labelled "predictability."

Cool. Thanks Google.

What To Do Before 17 August, Reviewing Your Target CPA & ROAS Campaigns

Enough grumbling. Here's the work.

1. Pull every budget-limited campaign.

Filter for "Limited by budget" status running Target CPA or Target ROAS. That's your exposure list.

Google will flag them too. The Bid Target Adjustment Tool went live on 6 July, and notifications go to anyone with a budget-limited target-based campaign in the last 12 months. Broad net, so don't be shocked when it surfaces something you'd forgotten was running.

You'll spot it when you check your bid strategies: a little amber warning against the campaign, a line reading something like "Recent performance as of [date]: £0.47," and an "Apply target" button sitting right there ready to update you to that figure in one click. That's the exact prompt Google serves when it wants you to change your CPA target, shown in the screenshot above. Handy. Also exactly the button you should not press on autopilot, which we'll come to.

2. Compare stated target against actual performance.

That's the whole job. What target did you set, and what is it actually achieving? Look at 30 to 60 days, not last week.

And if the two numbers are already close, breathe out. The impact on that campaign is likely to be minimal. This change bites hardest where the gap is widest, so a campaign sitting near its target has very little to give back.

3. Where there's a consistent gap, close it.

Beating your target? Lower the target to match reality. Achieving £22 on a £35 target, set it to £22. You're locking the efficiency in as the stated goal rather than letting the algorithm treat £35 as the bar.

Target still right for the business? Leave it. Just know the campaign will drift toward it.

Want something in between? Set a custom number. If £28 is what actually works commercially, use £28.

Budget genuinely fixed and you'd rather chase volume? Switch to Maximise Conversions or Conversion Value. Fair warning those spend the full budget with no target, so your CPA and ROAS will bounce around when budgets change. Pick your poison.

4. Don't blindly hammer "Apply".

The Bid Target Adjustment Tool is useful. It is guidance, not gospel. Same as everything else in the Recommendations tab.

It tells you what the campaign did. It doesn't know your margins, your close rate, your seasonality, or that one client who takes six months to pay. You know your account better than Google does. The right target comes from your business, not from last month's average, so don't ruin a good account by applying a suggestion blindly.

5. Extra care with PMax and Demand Gen.

These don't just shift CPA. Google's docs confirm you may also see changes in how traffic is distributed across channels. That's a structural reallocation of spend, not just an efficiency wobble. Review them properly rather than skimming.

6. Then leave it alone for a bit.

Two things will tempt you into meddling, and both are traps. Google has said the rollout happens over a few weeks with a brief calibration period, so don't panic-fix a campaign on 18 August because it looks odd for three days. And Performance Planner forecasts are explicitly unreliable between 17 and 31 August while the tools catch up, so no big bets on those numbers.

The upside of sitting on your hands, scaling should actually get more predictable after this. Bumping the budget on an over-performing constrained campaign used to send efficiency all over the place. Now budget increases should deliver closer to your stated target. If you are scaling, keep daily budget comfortably above average daily spend, and wait 1 to 2 conversion cycles before judging the result.

The Bigger Question - Should That Campaign Be on tCPA At All?

Here's the thing most of the coverage skips. This update is a good prompt to ask whether a target-based strategy is the right fit in the first place.

The two strategies do fundamentally different jobs. Maximise Conversions is built to spend the full available budget and get you as much as it can for it. Target CPA constrains spend in order to hit a cost goal. Those are not interchangeable, even though they often get treated that way.

So if an account is genuinely, persistently budget-limited, and the priority is squeezing maximum volume out of the budget you actually have, tCPA may not be the right tool at all. That's arguably been true for a while. This update just makes it more obvious, because the quiet efficiency that made tCPA look good on constrained campaigns is exactly what's going away.

Worth a proper look at any budget-limited campaign you've recently moved onto tCPA, or are about to. The honest question isn't "what target should I set," it's "what am I actually asking this campaign to do?" If the answer is volume within a fixed budget, you may be reaching for the wrong strategy.

What If My Budget-Limited Campaign Isn't Hitting Target?

Good question. And it's the one Google has been slowest to answer clearly.

Every worked example in the Help Centre is about campaigns overperforming. $10 hitting $5. £35 hitting £22. There's still no example spelling out what happens to a budget-limited campaign that's currently missing its target. So if you're sitting on stretch targets that are spending hard and falling short, and plenty of us are, there's no neat before-and-after diagram to point you at.

But it's not a total black hole anymore. Ginny Marvin has since explained the underlying mechanism, and it fills the gap. On budget-constrained campaigns, she says the target now becomes "your efficiency lever," and with no action the system "will aim to find conversions... at the average target set," which "may mean entering different auctions than you previously competed in."

Read that carefully, because it answers the fear. This isn't the system being made deliberately dumber. It's the system changing which auctions it competes in to hold your stated target, in either direction. A campaign beating its target gets pulled back toward it. A campaign set to a genuinely tighter target should, on the same logic, be steered toward that too, which likely means fewer, more selective auctions rather than more spend.

The honest caveat, that's the mechanism, not a guarantee about your specific missing-target campaign, because Google still hasn't published that exact scenario. Google's own page is worth reading in full before you act: Changes to target based bid strategies.

What we'd do:

• Leave genuine under performers alone for now. Google's own FAQ warns against reactive changes like data exclusions or bid limits purely because of this update, because it causes fluctuations.

• Put 17th August in your calendar and watch these campaigns properly, but don't judge them on a bad day or two. Give them time to gather enough conversions to show a real pattern first. If your customers typically take a week from click to converting, that's your window. That window is what's called a conversion cycle, so wait one or two of those before you draw any conclusions. Ginny's own advice is the same "judge against a full conversion cycle, not a bad Tuesday".

• Don't rebuild a campaign around a scenario Google hasn't actually worked through yet.

The 10-Minute Version - Your Pre-17 August Action Plan

Working straight down your account:

• Filter to "Limited by budget" plus Target CPA or Target ROAS

• Add an "actual CPA/ROAS, last 60 days" column next to your targets

• Flag every campaign beating its target by a wide margin

• For each flag, decide on purpose. Lower it to actual, set a custom number, or leave it

• Park anything underperforming or volume-focused for a proper strategy review, don't touch it yet

• Put "17 August, check rollout" in the calendar, and hold off judging anything for a full conversion cycle or two

🔹 Final Thoughts

Is this a good change? Not if you were benefiting from a system that beat your targets. Google is taking that back and calling it a feature, and the scepticism is earned. But being right doesn't lower your CPA.

The whole remedy is this, look at your targets, and make them mean what you want them to mean. Ten minutes of filtering and a few honest decisions about what the business actually needs. Ignore it, and the algorithm will treat whatever number you last set as gospel, whether or not it still reflects reality. Which, now you say it out loud, was always a bit mad anyway.

You've got until 17 August. Plenty of time. But only if you actually look. And as ever, if you're not sure whether a campaign's affected, ask before you click. Prevention beats cleanup.

Don't Panic. Just get your targets right first.

Sources: Google Ads Help Centre (Changes to target-based bid strategies, plus the FAQ), Ginny Marvin's public clarifications, and industry commentary from Kirk Williams and Mike Ryan.

---

4M Digital is a paid media consultancy specialising in Google Ads, Microsoft Ads, and Paid Social. With over 15 years of expertise, we help businesses unlock the full potential of their digital advertising strategies through tailored management, audits and training.