Are we going back to 2020, Google? In June the bidding menu showed four strategies again, the same four I chose from before 2021. Every campaign that ran Maximize Conversion Value with a target became Target ROAS, and nobody in my accounts clicked anything to make that happen.
Two months later, on August 17, Google changed how Target ROAS spends when a campaign is limited by budget. In one of the e-commerce stores I manage, Performance Max dropped to a ROAS of 7 in a single week and kept sliding to 5. On September 12 I moved every Performance Max campaign to Maximize Conversion Value with no target. Two weeks later the same budget was bringing back 12 to 20.
Four strategies, then two, then four again
Before 2021 you picked one of four: Target CPA, Target ROAS, Maximize Conversions or Maximize Conversion Value. In 2021 Google folded the two targets into the Maximize strategies as an optional field, and for five years a target was something you added on top.
In June 2026 Google started showing Target CPA and Target ROAS as separate strategies again. Its help page calls it "purely a visual change to the labels that won't change how your bid strategies work." Source: Changes to how Smart Bidding strategies are organized. The label change also put every campaign with Maximize Conversion Value and a target under the name Target ROAS. That mattered two months later.
How I used Target ROAS before August
I launched a campaign on Maximize Conversion Value with no target and let it collect data. Once it had a few weeks of sales, I added a target as a limit, so the campaign wouldn't spend on anything just to spend. A target of 300% often came back as 900% or more, because Google used a small budget on its best auctions.
A capped budget was the plan, and I set it on purpose. The businesses I work with don't have unlimited money for ads, because the same revenue pays for production and for the specialists who run the store and the campaigns. So we chose a daily budget the business could carry and put a low Target ROAS under it as a floor. Google spent that budget on its best auctions, and the campaign came back far above the target.
What Google changed on August 17
Google's help page says: "On August 17, 2026, changes to target-based bid strategies rolled out globally." The rollout finished on August 27. The change covers Target CPA and Target ROAS in Search, Shopping, Performance Max and Demand Gen, and only in campaigns limited by budget. Those campaigns now bid to reach the target more consistently. Source: Changes to target-based bid strategies.
Since August 17, Google reads my old setup as room to grow. It spends the budget on cheaper conversions until ROAS drops toward the number you typed in. Your 300% target stopped being a floor and became the number Google aims for.
Google says the change only affects campaigns limited by budget. In practice, that's every campaign that was beating its target: if it brought back 900% on a 300% target, it was spending its whole budget, whatever the status said.
Raising the target doesn't fix it
Google's own advice is to set the target at your recent average. A higher Target ROAS means Google bids less per click, because it expects more revenue from each one. Your competitors with a 300% target bid more and take the auctions you used to win. I tested it in the same store in late August: I raised the target in two campaigns to 3,500% and 2,845%, and within two days they almost stopped buying traffic. I set the target back after that test.
A low target drags ROAS down, and a high one switches the campaign off. That's why I removed the target altogether.
What happened in the store I manage
The store sells doors in the US, and most of its Performance Max campaigns ran on Target ROAS. In the week of August 17 they spent $2,223 at a ROAS of 7.4. The next week it was 6.1 on $2,657, and the week of August 31 it was 5.1 on $3,129.
On September 12 I switched every Performance Max campaign to Maximize Conversion Value with no target and left the budgets where they were. The week of September 14 brought 12.3 on $2,568, then 14.3 on $2,328, then 20.5 on $2,408.
I compare with late August and not with July for one reason. On August 26 we switched off a duplicate purchase action that had counted some orders twice, so the July numbers in this account are higher than the store's real sales. Late August against late September is the fair comparison, and it shows ROAS two to four times higher on the same money. Other PPC specialists posted the same pattern after August 17, so the drop wasn't specific to this store.
How to check your own campaigns
Open your Campaigns table and look at two columns: the bid strategy and the status. If a campaign shows Target ROAS or Target CPA and the status says limited by budget, the August change applies to it. Then compare its ROAS or CPA for the four weeks before August 17 with the weeks after. A campaign that used to beat its target by a wide margin and now sits close to it is the pattern I described.
You have two ways out. Google suggests setting the target at your recent average, and you'll trade some traffic for it, as above. The other way is Maximize Conversion Value with no target, and you control spend through the budget. Whichever you choose, stay on it. You can still change the budget, because the campaign keeps spending the way it did, but jumping from one strategy to another resets what it has learned. I give a new strategy three to four days before I judge it.
My tip: I don't use Target ROAS as a safety net anymore. Since August 17 a low target pulls ROAS down and a high one cuts traffic, so my Performance Max campaigns run on Maximize Conversion Value with no target, and I control spend with the budget.
Maximize Conversion Value bids on whatever your account counts as a primary conversion, so check that first: What Your Performance Max Campaign Is Really Optimizing For. If you'd like me to look at your bid strategies and budgets with you, book a consultation.
