The Google Ads Bargain You Forgot About Ends August 17
Somewhere in your Google Ads account is a number you typed in once, probably during setup, probably on the advice of an agency you no longer use. A Target CPA. A Target ROAS. You set it, the campaign ran, the leads came in, and you stopped looking. For years that was fine. Google quietly let your budget-limited campaigns beat that number, and you pocketed the difference without knowing it existed.
On August 17, 2026, that quiet generosity stops. Google announced on June 15, 2026, through its Accelerate updates and a post from Ads Product Liaison Ginny Marvin, that budget-limited campaigns using Target CPA or Target ROAS will be steered back toward the target you actually set, rather than the better result they have been delivering. The stale number becomes an enforced number. This article proves one thing: the fix is not panic, it is a 20-minute audit between July 6 and August 17 to find out whether your low costs are real efficiency or just Google under-spending a capped budget, then resetting your targets to your true business math.
The number you forgot is about to start governing your spend
Start with what these settings actually do, because most owners have never had them explained in plain English. Target CPA means "cost per acquisition," the price you told Google you are willing to pay for one conversion, usually a lead or a sale. Target ROAS means "return on ad spend," the revenue you want back for every dollar you put in. You set a number. Google's automated bidding tries to hit it.
When a campaign runs out of budget before the day is over, Google labels it "Limited by budget," which is a polite way of saying you ran out of money before you ran out of demand. In that situation, Google has been free to deliver conversions cheaper than your stated target, because it was rationing a small budget across the best available clicks. You set a Target CPA of $50 and got leads at $35. The gap was yours to keep.
Google's own documented example uses a campaign with a Target CPA of $10 that has been delivering conversions at around $5. After August 17, that campaign will begin delivering closer to $10. Google says the goal is more consistent, predictable performance in line with the targets advertisers actually set, especially as budgets change.
That phrase, "based on the targets you set," is a paraphrase of Google's stated rationale from its June 15, 2026 announcement, not a personal quote from anyone. Read it carefully, because it contains the whole problem. It assumes the target you set still reflects what a customer is worth to you. For most owners, it does not. It reflects a decision made years ago, under different margins, by someone who has since left. The setting that protected you because Google ignored it is about to start governing your spend.
The affected list is broad: Search, Shopping, Performance Max, Demand Gen, and Travel campaigns managed in Google Ads or Search Ads 360, plus Demand Gen managed in Display & Video 360. For the multi-channel ones like Performance Max and Demand Gen, Google notes that traffic distribution across channels may shift too, so the campaigns where you have the least visibility are also the ones most likely to move.
You treated a one-time setting like a one-time decision
Margins change. Customer value changes. The target almost never does. That is the gap this change exposes.
Picture a roofer in Tampa, just as an example. He set a $60 cost-per-lead target a few years back, when his jobs were smaller and that number felt about right. Today his average job is worth far more, and he could comfortably pay $120 a lead and still come out ahead. But his account still says $60, and for two years Google was quietly delivering leads at $48 because his budget capped out by mid-afternoon. He has been leaving volume on the table the whole time, and he never knew, because the number worked and working numbers do not get audited.
Now run it the other way. A boutique e-commerce store set a Target ROAS of 400% during a flush quarter, then watched Google deliver 600% because the budget was small and Google cherry-picked the cheapest converters. The owner saw 600% in the dashboard and assumed that was the campaign's natural ceiling. It was not. It was the side effect of a starved budget. After August 17, that campaign drifts toward 400%, costs rise, and the owner panics at a change they could have anticipated in twenty minutes.
The mistake is the same in both directions. People read the strong result in the dashboard as a permanent feature of the campaign, when it was a temporary product of budget rationing. This is the same confusion that lets accounts quietly bleed money in plenty of other ways, and it is worth understanding how a budget-limited account leaks 25 cents on every dollar before you assume your low costs are pure skill. Efficiency that depends on Google holding back is not efficiency you own. It is a loan, and the loan is being called.
The mechanic only bites budget-limited over-performers, and that detail is your map
This is not a blanket cost increase across every account. The change has a specific target, and knowing its shape tells you exactly where to look.
Three conditions have to line up for a campaign to be affected. It has to have been "Limited by budget" at some point in the last 12 months. It has to use Target CPA or Target ROAS. And it has to have been delivering better than that stated target. Miss any one of those and the campaign is untouched. A campaign that hits its target exactly does not move. A campaign that has never been budget-constrained does not move. The squeeze applies only to the gap between your set target and your actual over-delivery.
Some campaign types are exempt entirely. App campaigns, Video reach campaigns, and Video view campaigns are not affected. If your spend lives mostly there, you can close this tab. For everyone running Search, Shopping, or Performance Max on a target-based strategy, assume you are in scope until you check.
Google gives you a way to check and a clock to do it by. A Bid Target Adjustment Tool launches inside Google Ads on July 6, 2026. If a campaign was "Limited by budget" in the last 12 months and uses an affected strategy, you get an in-account notification pointing you to it. That leaves roughly a six-week window: tool opens July 6, change lands August 17. Google will not adjust your targets or budgets for you. Silence is a choice, and the default outcome of doing nothing is your costs drifting up toward whatever number you typed years ago.
One number from Google is worth knowing here, with the caveat that it is Google's own figure and not independently verified. Google reports that its Smart Bidding Exploration produced 18% more unique converting query categories and 19% more conversions in an internal test run March through April 2025. Treat that as a sales argument for trusting automation, not as proof. The reason to act is not the lift number. It is the documented mechanic: a target you forgot is about to govern money you spend.
Run this audit before August 17 and you make the change a non-event
Twenty minutes, no technical skill required. The sequence, whether you do it yourself or hand it to whoever runs your account.
First, after July 6, open Google Ads and look for the notification about the Bid Target Adjustment Tool. No notification across your campaigns points to you being out of scope. Confirm your campaign types before you relax. If you see it, you have a campaign to review.
Second, for each flagged campaign, write down two numbers side by side: the Target CPA or ROAS you set, and the actual CPA or ROAS the campaign has been delivering. The gap between them is exactly what is about to close. A $50 target delivering $35 leads is a $15 gap headed back toward $50.
Third, decide what a customer is genuinely worth to you today, not in 2022. Take your average sale value, your close rate from leads, your margin. If a closed job nets you $4,000 and one in five leads closes, a lead is worth $800 to you in gross terms, and a $50 cost is nowhere near your real ceiling. Knowing this number is the entire game, and if your conversion data is shaky you are flying blind, so it is worth confirming your account is reporting real conversions and not lying to you before you trust any of these figures.
Fourth, when the tool offers its three options, match the option to that math. Option one keeps your stated target as-is and accepts that costs drift up toward it, which is right only if your set target already reflects your true business value. Option two matches the target to your recent actual performance, locking in the efficiency you have been getting, which is right when you are happy with current volume and do not want costs to move. Option three sets a custom target based on your real math, which is right for the Tampa roofer who could pay far more per lead and wants the volume that comes with it.
Fifth, if a campaign is a genuine workhorse and you simply want maximum results without a target ceiling, there is an off-tool path: switch it to Maximize Conversions or Maximize Conversion Value. That removes the target governor entirely. It is an alternative, not a default, and it suits campaigns where you trust the budget to do the limiting rather than the target.
If you delegate this, the exact question to ask your agency is short. "Which of our campaigns got the Bid Target Adjustment notification, what is the gap between our set target and actual performance on each, and which of the three options did you pick and why." An agency that cannot answer that in plain numbers is an agency that has not looked.
Watch the cost per lead and the volume together, or you will read the change wrong
After August 17, the dashboard will tell two stories at once, and the trap is reading only one of them.
Track your actual cost per lead or your ROAS week over week against what you set in the tool. If you chose to lift your target toward your real business value, expect cost per lead to rise and lead volume to rise with it. That is the trade working as intended. Rising cost on its own is not a problem. Rising cost with no extra volume is.
The vanity-metric trap here is ROAS worship. A high ROAS number on a small budget feels like winning, but it often just means Google is skimming the cheapest converters off a campaign that could profitably do much more. If your Performance Max ROAS looks gorgeous, be skeptical, because PMax ROAS routinely takes credit it did not actually earn. The honest scoreboard is total profit, not the percentage that looks best in a screenshot. A 600% ROAS on $2,000 of spend makes less money than a 400% ROAS on $10,000.
Give it two to three weeks before you judge anything. Automated bidding re-learns after a change, and the first few days will be noisy. Set a calendar reminder for early September to compare cost, volume, and total profit against your pre-change baseline. If profit held or grew, you made the right call. If profit fell, revisit the tool and adjust the target again, because nothing you set in August is permanent.
Frequently Asked Questions
Do I have to do anything, or will Google handle it for me?
You have to act yourself. Google has stated it will not auto-adjust your targets or budgets. If you do nothing, affected campaigns will drift toward the target you originally set, which for most owners means costs creeping up toward a number chosen years ago. The Bid Target Adjustment Tool opens July 6, 2026, and the change takes effect August 17, 2026, so you have roughly six weeks to review and decide.
How is this different from the earlier pacing change Google made in June?
They are two separate things. The earlier June change was about how Smart Bidding paces budget through the day and is covered in detail in the piece on why Google made bidding smarter while your budget did not get the memo. This August 17 change is specifically about budget-limited campaigns that have been beating their Target CPA or Target ROAS being steered back toward the target you set. One is about pacing, this one is about target enforcement. Both can affect the same account, but they are not the same mechanic.
Which campaigns are safe from this change?
App campaigns, Video reach campaigns, and Video view campaigns are not affected at all. A campaign is also untouched if it never hit "Limited by budget" in the last 12 months, if it does not use Target CPA or Target ROAS, or if it was already delivering at or above its target rather than beating it. The change only bites the gap between a stated target and the better-than-target performance a budget-limited campaign was delivering. If none of your campaigns got the in-account notification, you are clear.
The uncomfortable lesson is not about August 17. It is about the fact that a setting you forgot has been silently shaping your results the entire time, in your favor, by accident. Google is closing one accidental advantage. There is no reason to assume it is the only one in your account, or that the next correction will land in your favor. The owners who come out ahead are not the ones who react fastest on August 17. They are the ones who decided to know their own numbers before Google decided for them.
