Google’s change to target-based bid strategies took effect on August 17, 2026. Budget-limited Search, Shopping, Performance Max, Demand Gen, and Travel campaigns on Target CPA or Target ROAS now deliver toward the target you typed into the account — not the better number Smart Bidding had been quietly finding on its own.
For a lot of Michigan accounts, that target was set once, years ago, and never touched. If yours has been beating it, you now choose between two levers that pull opposite directions: lower the target, or raise the budget.
What actually changed in Google Ads on August 17, 2026?
Budget-constrained campaigns using Target CPA or Target ROAS now optimize toward your stated target instead of drifting to a more efficient number. Google’s own documentation gives the example plainly: a campaign with a $10 Target CPA that has recently been achieving a $5 actual CPA “will deliver more closely to a $10 actual CPA starting August 17, 2026.”
Scope matters. Display and Hotel campaigns already ran on the new behavior; App, Video reach, and Video view campaigns keep the old one for now. Everything else is in, across Google Ads, Search Ads 360, Display & Video 360, Google Ads Editor, and the API — there is no interface where the old logic survives.
Google shipped a Bid Target Adjustment Tool on July 6, 2026 to let advertisers reset targets ahead of the switch. Find it in the “Review your campaign targets” notification banner, or on the Campaigns page under the settings icon → Bidding → Review campaigns. Search Ads 360 users won’t see it in Google Ads at all; it appears in SA360 under “Review bid strategies.”
Why did a campaign ever beat its own target in the first place?
Because a budget-limited campaign was solving two constraints at once, and the budget was the one that bound. Understanding this is what lets you predict which of your campaigns will move and which won’t.
Smart Bidding sets a bid for every auction based on predicted conversion probability. When daily spend runs into a hard budget cap, the system can’t enter every auction it wants, so it prioritizes: it takes the cheapest, highest-probability conversions and skips the expensive tail. The result is a realized CPA well below the stated target — not because the algorithm was being generous, but because the budget ceiling forced it to cherry-pick. Your $120 target was never really operating. Your $150 daily cap was.
After August 17, the target becomes the binding constraint. The system bids up into auctions it previously skipped — more competitive queries, more expensive audiences, in Performance Max a different channel mix entirely — spending the same budget on fewer, costlier conversions that were previously unreachable.
That mechanism tells you which campaigns are safe. If a campaign is not limited by budget, the target was already binding and nothing changes. If it is limited by budget but realized CPA already sits at target, there’s no gap to close and nothing changes. Only budget-capped campaigns with a meaningful spread between stated target and actual performance will move. Sort by that spread and you have your work list.
Does this update actually cost advertisers money?
For the specific set of accounts that were overdelivering, yes — and the framing of “predictability” is doing a lot of work. Search Engine Roundtable’s read was blunt: the update looks likely to cost advertisers more. That’s a fair reading of the direction, though not the whole picture.
The predictability benefit is real. Under the old behavior, raising a budget on an overperforming campaign produced erratic efficiency swings, which is why so many advertisers learned to scale in nervous 10% increments. That coupling is gone.
But notice the asymmetry. The drift is one-directional: campaigns performing better than target move toward target, while nothing moves underperforming campaigns down toward theirs. Google is standardizing on the number that produces more revenue per impression and calling it consistency. Both things are true at once — the feature is useful, and it is not neutral.
The deeper problem is Google’s “no action needed” path, which assumes your stated target is a considered business number. In practice, most small in-house accounts set a Target CPA once during setup and left it alone precisely because it was overdelivering. Those accounts have the largest gap and the least monitoring — the people most exposed are the ones least likely to open the notification.
Which Michigan campaigns are most exposed heading into fall?
Seasonal service businesses with hard budget caps and stale targets — which describes most of metro Detroit’s home services market. The timing is unkind: the change landed six weeks before Michigan furnace season starts pulling emergency HVAC demand.
Take a plausible case. An HVAC contractor in Livonia runs Search at a $150 daily budget — roughly $4,500 a month — with a $120 Target CPA set in 2024, converting at about $71 for roughly 63 booked calls a month. That same $4,500 delivering closer to $120 produces about 37. Same invoice from Google, forty percent fewer calls, arriving as the phones should be ringing hardest.
Two other Michigan patterns carry the same risk. Multi-location dental and legal practices across Rochester Hills, Southfield, and Sterling Heights cap per-location budgets tightly so no one office eats the group spend — nearly all of those campaigns are permanently budget-limited. And Tier 2 suppliers in Warren and Grand Rapids often set a Target CPA against a close rate measured years ago, then let long B2B cycles hide the drift.
Our Detroit PPC management team has been sorting accounts by target-to-actual spread since the tool went live in July, because the fix takes ten minutes per campaign and the alternative is finding the shortfall in a September pipeline review.
How should you decide between lowering the target and raising the budget?
Lower the target to protect volume at flat spend; raise the budget to buy more volume at a worse unit cost. The deciding variable is whether you have capacity to service more jobs, not which number looks better in the interface.
- Pull a 90-day view of every campaign on Target CPA or Target ROAS, add the “Budget” status column, and filter to campaigns showing “Limited by budget.”
- Calculate each spread: stated target versus 90-day actual. Within 10% needs no action. At 25% or more, it’s on your priority list.
- Write down your real break-even cost per lead — close rate times average job value times target margin. Not the number in the account. The number from your books.
- If actual CPA is close to break-even, lower the target to match recent performance in the Bid Target Adjustment Tool. Volume holds, efficiency holds, spend holds.
- If actual CPA is comfortably under break-even and your team can absorb more work, raise the budget and leave the target alone. You trade a higher CPA for more total conversions — the trade this update was designed to make safe.
- Under roughly 30 conversions a month, the target was statistically thin anyway. Consider Maximize Conversions, accepting that CPA will float with budget.
- Annotate every change with the date and reason. When September performance looks strange, you need to separate your edits from the platform’s.
- Re-check after one to two full conversion cycles, not one week — for B2B accounts a cycle can run 30 to 60 days.
What not to do is split the difference — nudging the target down 10% and the budget up 10% at once changes both sides of the equation and leaves you unable to attribute the result to either.
What dates should be on your calendar right now?
The enforcement date has passed, so your calendar is now about measurement windows, not deadlines.
- July 6, 2026 — Bid Target Adjustment Tool went live. Still available; it did not disappear at enforcement.
- August 17, 2026 — New bidding behavior in effect for Search, Shopping, Performance Max, Demand Gen, and Travel.
- Week of August 31, 2026 — First clean two-week read on post-change CPA. Compare to the two weeks before August 17, not to last year.
- Mid-September 2026 — Michigan heating demand begins climbing. Any target correction should be stabilized before this, not made during it.
- October 2026 — Holiday retail auction prices start rising. Shopping and Performance Max targets set in a summer auction environment need a second look regardless of this change.
Frequently asked questions
Does this affect my Performance Max campaigns differently?
Yes. Google notes that multi-channel campaigns like Performance Max and Demand Gen may also see shifts in how traffic is distributed across channels, not just a change in cost per conversion. A budget-limited PMax campaign might quietly move spend from Search inventory toward Display or YouTube placements as it bids up toward your target. Watch channel-level reporting, not only the headline CPA.
My campaign has never shown “Limited by budget.” Am I safe?
Largely, yes. If budget was never the binding constraint, your target was already governing bids and there is no gap for the system to close. You may still receive the account notification, since Google triggers it for anything budget-limited in the past 12 months. Verify against your actual budget status column before making changes you don’t need.
Is switching to Maximize Conversions a safe escape hatch?
It’s an escape from target enforcement, not from cost pressure. Maximize Conversions and Maximize Conversion Value spend the full budget with no target, so your CPA or ROAS will fluctuate whenever you adjust budget. That’s a reasonable trade for low-volume campaigns where the target was never statistically meaningful. It’s a poor trade if you have real margin thresholds to defend.
How long before I can tell whether the change hurt me?
One to two full conversion cycles, per Google’s guidance. For a Livonia HVAC account with same-day conversions, that’s about two weeks. For a Warren industrial supplier with a six-week research cycle, it’s closer to three months. Judging a bidding change on seven days of data produces confident, wrong conclusions in both directions.
Start this week: filter to budget-limited campaigns on target-based bidding, rank by the gap between stated target and 90-day actual, and fix the widest gaps using a number from your margins rather than the tool’s suggestion. If you’d rather have that audit run for you, get in touch with ROSH Media.
Comments are closed