By Lee Moran, Founder of Veritas Digital
From 17 August 2026, limited by budget campaigns on Target CPA and Target ROAS will stop quietly beating their targets and start delivering on target only. For some accounts, that potentially means that CPA doubles. There is roughly four weeks to decide whether that happens on your terms or on Google’s.
If you run, manage or oversee Google Ads and you have ever set a Target CPA of £50 and quietly enjoyed a £25 one, this article is about you.
On 15 June, Google announced a change to how its bidding systems handle campaigns that are limited by budget. It was framed as an improvement to predictability, and it was technical enough that a lot of advertisers skimmed past it. Then on 2 July, notification emails started landing in accounts, and the PPC industry worked out what it actually means.
Here is the short version: campaigns that have been overperforming their targets will stop overperforming. Not because Google is changing your targets, and not because Google is raising your budgets. It’s because Google is going to start taking the number you typed in very seriously.
For a lot of accounts, that number is out of date. And that’s where budgets will soon be wasted.
What is it that’s changing on 17th August?
Today, when a campaign has “Limited by budget” status and uses a target based bid strategy, Smart Bidding behaves conservatively. It cannot spend everything it wants to, so it becomes selective. It enters only the auctions where a conversion looks cheaper and near certain, and it leaves the more expensive, less certain ones alone. The budget cap is doing the filtering.
The side effect is a campaign that beats its target. You asked for a £50 CPA, the system gave you £25, and nobody complained. Google’s position is that this was never the intended behaviour.
From 17 August, that changes significantly. Smart Bidding will optimise more consistently toward the target you set, regardless of whether the campaign is limited by budget or not. If your target is £50 and your actual is £25, your actual moves toward £50.
Google’s own example in the help documentation is a Target CPA of £10 against a recent actual of £5. Their guidance is that the campaign will deliver closer to £10 after the change. Read that carefully, because it is Google telling you the cost per conversion could double, using their own numbers. This isn’t hyperbole – Google is issuing a direct and clear warning on this update.
What is not changing
It’s worth being very precise here, because the panic has outrun the facts in places:
- Google will not automatically adjust your bid targets.
- Google will not automatically increase your daily budgets.
- Target Impression Share, Target CPC, Target CPM and manual CPC are untouched.
- App campaigns, Video reach and Video view campaigns keep the original behaviour. Hotel and Display campaigns are already on the new smart bidding.
There is also a separate, cosmetic change that landed in June: “Maximize conversions with a Target CPA” is now just “Target CPA” in the interface. That is a rename with no behavioural impact. It has arrived in the same window as a genuine functional change, which is unhelpful, and I have already seen people confuse the two.
Who is affected, and how bad will it be?
The update applies to Search, Shopping, Performance Max and Demand Gen, across Google Ads, Search Ads 360 and Display & Video 360, plus the Google Ads Editor and the API.
The scale of your exposure comes down to one number: the gap between your stated target and your actual performance.
| Your situation | What happens on 17 August | Urgency | Waste |
| Not budget-limited, hitting target | Little to nothing. Carry on. | Low | Low |
| Budget-limited, hitting target | Broadly stable. Scaling should become more predictable. | Low | Low |
| Budget-limited, beating target by 10-20% | Efficiency drifts toward target. Noticeable, manageable. | Medium | Medium |
| Budget-limited, beating target by 50%+ | Same spend buys fewer, more expensive conversions. | High | High |
Performance Max and Demand Gen advertisers have a second problem. Because these are multi-channel campaigns, Google has said you may also see shifts in how traffic is distributed across channels. That is not just a CPA movement, it is a structural reallocation of where your money goes. When you look at the channel performance of Performance Max, you may see increased allocation into Display and Video (and possibly Discover as well).
The part Google is not saying out loud
I will say it once and then move on to the fix, because the fix is what matters.
Your budget cap has been working as a free efficiency lever. By starving a campaign, you forced the system to be picky, and picky meant cheaper. That lever is now being switched off and Google’s recommended response, in its own documentation, is to give your budget room to scale and to check the Recommendations page to forecast additional conversions at your target. Translated: remove the cap and let the system spend more.
More spend at higher CPAs is good for Google’s revenue. That is not a conspiracy, it is arithmetic, and Google would not pretend otherwise. However, the change has a real logic to it too: predictable behaviour when you adjust budgets is genuinely useful, and a system that quietly ignores the instruction you gave it is not a system you control.
But here is the honest read: this only costs you money if your targets are wrong. Google is holding you to a number you chosen yourself. The uncomfortable question is why so many accounts have a number in there that nobody has revisited in two years.
That is not really Google’s fault. That is account management and needs to be reviewed closely.
What to do before 17 August
A Bid Target Adjustment Tool has been live in Google Ads since the 6th July. It shows you affected campaigns, their recent performance, and offers a one-click “Apply” to reset the target to recent actuals. Consider it, use it, but do not let it think for you. Here is the sequence I personally would run:
1. Build your exposure list
Pull every campaign that has held “Limited by budget” status in the last 12 months and uses a Target CPA or Target ROAS bid strategy. Put actual CPA or ROAS next to stated target. Sort by the size of the gap. The larger the gap, the great the risk and for most accounts, those gaps may be larger than people realise.
2. Decide whether each gap is strategy or drift
This is the judgement call and depends on the strategy of each campaign.
If the gap is deliberate headroom, you set a loose target on purpose so the system had room to find volume, then you may want to keep it. If the gap is drift, a target set eighteen months ago against a different margin structure that nobody has touched since, then it is an out-of-date target. Lower the CPA target towards your actual target.
The lazy move is to hit “Apply” on every campaign and lock in recent actuals across the board. That freezes your account at whatever efficiency the budget cap happened to produce, which was never a good decision in the first place. It was an accident of the algorithm being constrained.
3. Set targets from your unit economics, not your history
The right target is the one your margins can carry, not the one your budget cap accidentally delivered. If a £25 CPA is what you have been getting but a £35 CPA is still profitable, then setting £25 leaves volume on the table. If £25 is already at the edge of profitability, protect it.
This is where the client side homework matters. If you do not know your close rate and your customer lifetime value, you cannot answer this question, and no tool in the Google interface will answer it for you.
4. Check what you are optimising towards
Smart Bidding scales towards whatever conversions you are feeding it. If your target is built on raw form fills rather than qualified enquiries, a stricter enforcement of that target simply means Google buys you more of the wrong thing, faster and more reliably.
For lead generation accounts, this is the deeper risk and almost nobody is covering it. The bidding change is a forcing function on conversion tracking quality. If your tracking is loose, the 17th August will not be the day you find out, but it will be the day it starts costing more.
5. Leave time to re-stabilise
Smart Bidding reacts to target changes within minutes, but takes one to two conversion cycles to settle, often around seven days and considerably longer in niche B2B where conversion cycles run to weeks.
Make your changes with a fortnight to spare, not on the 16th August. Changing targets on the eve of a system-wide bidding shift means you will never untangle which effect caused what.
A note on switching bid strategy
Google offers Maximize conversions and Maximize conversion value as an escape route. If you want maximum volume for a fixed budget and you are relaxed about efficiency, that is a good alternative.
But be clear about the trade though. Without a target, your actual CPA or ROAS will fluctuate every time you touch the budget. You are not regaining control; you are swapping one kind of unpredictability for another. For most accounts I would rather fix the target than abandon it, as targets themselves create efficiency whilst still protecting most of your volumes.
The Bottom Line
On the 17th August, Google stops letting limited by budget campaigns quietly outperform and starts holding them to the targets you set. If your targets are current and within range of actuals, very little happens. If they are out of date, your CPA moves toward a number you set a long time ago, and in Google’s own worked example that means it can double.
The work is not complicated or time heavy. Find the limited by budget campaigns with target based bidding, measure the gap between stated and actual, and decide deliberately which targets are out of date and which are ok to leave. Then set targets from your unit economics rather than from whatever the algorithm happened to deliver while it was constrained.
What this change really exposes is how many accounts are running on automation no human operator has checked in a while. Google is not taking your efficiency: it’s tapping into accounts that are left to automation and haven’t been reviewed for some time.
At Veritas we review accounts frequently, including CPA and ROAS targets, with a human operator reviewing every change before it goes live. You can read more about how we approach paid search on our Paid Search services page, or skip straight to getting in touch to hear more from us directly.
Get in touch with us
If you are looking at the 17th August and are not sure which of your campaigns are exposed, or whether the gap in them is deliberate or just out of date, that is exactly the kind of work we do. Get in touch with us and we will review your limited by budget campaigns, pressure test your targets against your unit economics and give you a clear plan before the change lands rather waiting until spend is quietly wasted.
References
- Google Ads Help – Changes to target based bid strategies
- Google Ads Help – Frequently asked questions about changes to Target-based bid strategies
- Search Engine Journal – Google Clarifies Smart Bidding Update After Advertiser Concerns
- Search Engine Roundtable – Google Ads Promotion Mode Beta, Smart Bidding Exploration Expands & Bidding Target Optimization Changes
- PPC Land – Google Ads gets promotion mode and a major bidding overhaul this August
- PPC Land – Google Ads forces some CPAs to double starting August 17
- PPC Land – Google Ads bidding overhaul forces CPAs to double, sparking backlash
- PPC Land – Google gives Demand Gen advertisers until August 17 to fix bid targets
