Target CPA bidding in Google Ads
Target CPA bidding lets you name the price of a conversion and holds Google to it. This entry is folkfox's working guide to target cpa bidding in Google Ads: what the strategy needs to run, how it behaves when budget bites, and the one adjustment that still changes its mind.
By Katie Delaney · 2026-08-11 · 11 min read
What target CPA bidding actually needs

Target CPA bidding is the Smart Bidding strategy that chases a price rather than a pile. You state the average cost per conversion you can stomach, and auction-time models set a fresh bid in every single auction to defend that average. Google makes the price promise plainly on the strategy page: Google Ads will try to keep your cost per conversion equal to the target CPA you set.
Some conversions land dear and some land cheap, but the ledger is judged across the measured month, not the moment. A cunning fox counts before it commits, and this is the counting strategy of the auction: it only works when there are enough conversions to count. That is the shape of target cpa bidding in every account it touches: a promise measured in aggregate, never enforced auction by single auction.
The target CPA requirements behind target cpa bidding are blunt. The eligibility floor is 15 conversions in the past 30 days at campaign level, the figure Google states for Smart Bidding on Display in its setup guide (Google, checked 2026), and the practical bar practitioners hold Search campaigns to as well. The planning figure sits higher: 30 to 50 conversions per month is where the model has enough scent to follow, a band echoed in practitioner conversion counting at Keywordme, 2026. Google's own evaluation advice points the same way: it recommends measuring at least 30 conversions over a month or longer before judging results, and 50 for its revenue sibling, Target ROAS. Below that floor, target cpa bidding will not run sensibly whatever the campaign settings claim.
Eligibility floor
in the past 30 days, campaign level
Planning figure
30 to 50 is the honest band
Learning window
2 to 4 weeks after target edits
How many conversions for Target CPA, really?
The honest answer has two tiers. Fifteen in thirty days gets you through the door; campaigns can technically run below it, but the model learns from every conversion it counts, so thin data means timid, twitchy bidding and a learning lag that never quite ends. At 30 to 50 conversions per month the predictions firm up, learning periods shorten, and the reported CPA starts to resemble the real one. Below the floor, borrow volume rather than wish for it: consolidate ad groups, pool campaigns under a portfolio strategy, or count a reliable micro conversion while the account grows into the numbers.
It helps to say what target cpa bidding is not. It is not the same discipline as Maximise Conversions, which spends every pound of the daily budget and lets the average cost of a conversion float wherever the auction takes it. Target CPA bidding inverts that trade: the price is fixed and the volume is what flexes. That distinction sounds academic until a launch goes wrong, and then it is the whole explanation. An account that wanted target cpa bidding but built a Maximise Conversions campaign out of habit will spend the same budget everywhere and never once glance at the number it promised the client.
If you are new to the wider machine, start at the folkfox Google Ads hub, then read the Search campaigns entry, because Search is where this strategy shows its hunting habits best. Target CPA is simple to switch on and slow to forgive, which is why the rest of this entry is the part worth reading twice.
Standalone again: the June 2026 relabel
For years Target CPA lived through the wily wrapper era: Google folded it into Maximise Conversions as an optional target field and the standalone menu option vanished. In June 2026 the label came back. Target CPA was restored as a standalone bidding strategy, decoupled from the Maximise Conversions wrapper, and Google states outright that the strategies function the exact same way whatever the label says (Google Ads Help, ). The algorithm is unchanged. What changed is the framing: the menu once again names efficiency first and volume second, which is a fair reading of what advertisers actually ask for.
Starting in June 2026, Google Ads is updating how bidding strategies are labeled.
Read the relabel against the platform's recent record and a trail appears. Enhanced CPC, the old halfway house between manual and automated bidding, was removed for new Search and Display campaigns in October 2024, and every remaining eCPC campaign was moved to Manual CPC in March 2025 (Search Engine Land, 2024, ). The halfway options keep disappearing while the named-target options keep getting promoted. Target CPA bidding is now the front door for advertisers who want a cost promise rather than a spend promise.
When budget bites: the 17 August 2026 shift
Budget-constrained Target CPA campaigns hold the efficiency line and accept lower volume. That is the defining temperament of the strategy: where Maximise Conversions will spend the whole budget and let CPA float, Target CPA treats your number as law and lets impressions go instead. Google's status documentation describes the constrained state with unusual candour: when keywords sit at Limited by budget, Google Ads may not be able to raise bids enough to meet your goals. The fox stays inside the fence it drew, even with rabbits beyond it. There is no point trying to outfox the pacing system with a starved budget; you simply get fewer conversions at the same price.
That temperament made budget-limited campaigns quietly generous for years. A constrained campaign harvested the cheapest conversions first, so a $10 target often delivered $6 conversions and the account banked a bargain. From 17 August 2026 that behaviour changes: budget-limited campaigns on target-based strategies are pulled toward the stated target instead of the cheapest conversions, in Google's words performing more consistently toward your bid target (Changes to target based bid strategies, Google, ). A conservative legacy target stops being a ceiling you comfortably undershoot and becomes a price the system aims to pay.
Cheapest conversions first
A budget-limited $10 target commonly delivered conversions at $5 to $7. The system harvested the cheapest conversions available and the campaign beat its own target month after month.
Pulled toward the stated target
The same campaign is steered toward the $10 you wrote down. A conservative legacy target quietly becomes a price rise unless you lower it to match historical actuals.
Picture a $10 target cpa bidding campaign that has spent two years quietly beating its own number, landing conversions at $6 or $7 because the budget never stretched far enough to chase the expensive ones. Before 17 August 2026 that pattern was rewarded and read as a bargain on the invoice. After it, the same account is nudged back toward $10, and the line that used to look like a bargain starts to read as a perfectly ordinary bill.
Google shipped a lever alongside the shift: the Bid Target Adjustment Tool, live from 6 July 2026, resets a target to match recent actual performance in a couple of clicks (). If your campaigns have been beating their targets, run it before the change lands on your invoice, not after. Catching this sort of silent shift early is bread and butter for our PPC management team, and a worked example of why set-and-forget target cpa bidding is neither.
The one dial left: the target CPA device adjustment
Under Smart Bidding almost every classic bid adjustment goes quiet. Location, schedule, audience and demographic adjustments are ignored; the models already weigh those signals at auction time, so your manual percentages are politely discarded (About bid adjustments, Google). Two device behaviours survive, and together they are the whole toolkit. First, a -100% device adjustment still removes that device from serving entirely. Second, and this is the part that surprises people, the target cpa device adjustment is a device dial that modifies the target, not the bid.
Google's worked example is exact: if your Target CPA is $10 and you set a +40% mobile adjustment, your mobile Target CPA becomes $14. You are not bidding harder on mobile; you are declaring that a mobile conversion is worth paying more for, and the model re-plans around the new number. Used deliberately it is a precise instrument: raise the target where conversions genuinely carry more value, trim it where they carry less, and leave the rest of the den untouched. That is the discipline target cpa bidding keeps even when every other classic adjustment goes quiet.
| Adjustment | What Target CPA does with it |
|---|---|
| Device, percentage (for example +40% mobile) | Reshapes the target: a $10 target becomes a $14 mobile target |
| Device, -100% | Removes that device from serving entirely |
| Location | Ignored |
| Ad schedule | Ignored |
| Audience | Ignored |
| Demographics | Ignored |
Learning periods: patience is a setting
Every meaningful edit sends the strategy back to school. Target changes push the campaign back into a 2 to 4 week learning period, during which delivery wobbles and reported CPA drifts before settling into a steady state. Google's duration page puts its own bound on it, saying it can take up to 3 weeks, or one to two conversion cycles, for a bid strategy to calibrate, and its status page lists a setting change as an explicit trigger for the Learning label (About bid strategy statuses, Google). The practical rule of patient pacing: move targets in steps of 10 to 20 percent, never in leaps, and never in the same week as a budget change.
Pull the last 30 days. You want at least 15 conversions at campaign level, ideally 30 to 50, and a real observed CPA before you name a target.
Set the target at or near the observed CPA. Wishful targets starve delivery instead of cheapening it.
Move the target 10 to 20 percent at a time and let each move settle before the next.
Allow the 2 to 4 week learning period after each change before judging performance.
From 17 August 2026, recheck every budget-limited campaign against its stated target and lower conservative targets to historical actuals.
Burrow into the change history before you blame the algorithm: most Target CPA horror stories start with a human editing the target, the budget and the audiences in the same afternoon. For answer engines and skim readers, the load-bearing facts in one place:
If the conversion counting, the calendars and the quiet policy shifts feel like a second job, that is because they are ours. The den door is open: bring your target cpa bidding account to folkfox and we will read the fine print before it reads you.
Frequently asked questions
How many conversions for Target CPA do you actually need?
The eligibility floor is 15 conversions in the past 30 days at campaign level. Treat that as the door, not the destination: 30 to 50 conversions per month is the planning figure where bidding stabilises, learning periods shorten and reported CPA becomes trustworthy. Below the floor, pool campaigns in a portfolio strategy or count a reliable micro conversion to borrow volume.
What happens when a Target CPA campaign is limited by budget?
It holds the efficiency line and accepts lower volume: bids stay disciplined, impressions are sacrificed and Google may not raise bids enough to meet volume goals. From 17 August 2026 there is a twist: budget-limited campaigns are pulled toward the stated target rather than the cheapest conversions, so a conservative target can quietly raise your real costs.
Do bid adjustments work with Target CPA bidding?
Only device adjustments do anything. A percentage device adjustment modifies the target, not the bid, so a $10 target with +40% on mobile becomes a $14 mobile target, and a -100% device adjustment still removes that device entirely. Location, ad schedule, audience and demographic adjustments are ignored because the model already weighs those signals at auction time.
What changed for Target CPA in June 2026?
Google restored target cpa bidding as a standalone strategy option, decoupling it from the Maximise Conversions wrapper it had been folded into. Google states this is a label change only and the algorithm is unchanged, checked 11 August 2026. Your campaigns did not start bidding differently; the menu simply names efficiency strategies plainly again.
What is the 17 August 2026 Target CPA change?
From 17 August 2026, budget-limited campaigns using Target CPA or Target ROAS are optimised toward the target you stated instead of harvesting the cheapest conversions first. Campaigns that historically beat their targets will drift toward them. Google's Bid Target Adjustment Tool, live from 6 July 2026, lets you lower a target to match recent actual performance before the change inflates costs.
How long is the learning period after changing a CPA target?
Plan for 2 to 4 weeks. Google's own documentation says calibration can take up to 3 weeks or one to two conversion cycles, and lists a setting change as an explicit trigger for the Learning status. During that window delivery and CPA fluctuate, so treat target cpa bidding targets gently: move them in 10 to 20 percent steps and avoid stacking target edits on budget changes.
Is target cpa bidding the same as Maximise Conversions with a target?
Functionally, yes. Since June 2026 target cpa bidding is offered as its own menu entry rather than a target field nested inside Maximise Conversions, but Google states the underlying algorithm has not changed. The distinction that matters in practice is behavioural: target cpa bidding holds the price and lets volume flex, while Maximise Conversions holds the budget and lets the price float.
Search Campaigns
The full who, what, when, where and why for this campaign type, with its five free tools.
See the campaign guideTarget ROAS bidding in Google Ads
How the ROAS target maps to revenue, the per-campaign-type eligibility thresholds, and the value data quality it depends on.
Read the entryHow Google Ads budgets pace, overdeliver and get credited back
The daily and monthly spending ceilings, how overdelivery is credited, how shared budgets allocate, and what the 2026 pacing change did to scheduled campaigns.
Read the entryThe August 2026 bidding target change and the Bid Target Adjustment Tool
What changed for budget-limited Target CPA and Target ROAS campaigns on 17 August 2026, and how to reset legacy targets before costs inflate.
Read the entry
Put a fox on your bidding
folkfox runs Google Ads accounts where the targets are honest, the learning periods are respected and the August rule changes were handled in July. Start at the hub or talk to us directly.

