How Do I Set a Target CPA for My Google Ads?

Google Ads is asking for a target CPA, and the number you type will steer every bid the campaign places from tomorrow. Finance has a figure in mind, and last month's real cost per conversion sits above it. Choose wrong and you either choke your volume or let spend drift past what a conversion is worth.
Set your first target CPA at the campaign's actual average cost per conversion over the last 30 days, which is the figure Google itself recommends. Then lower it in small steps while conversion volume holds. Your break-even cost per conversion is the ceiling the target stays under, not the place to start.
What should your first target CPA be?
Google defines the setting as "the average amount you'd like to pay for a conversion." Its Target CPA help page says where to start. If the campaign has conversion history, Google Ads recommends a target equal to your average CPA from the last 30 days, adjusted for conversion delays. Without history, the recommendation falls back on whatever business goal you give it.
The delay adjustment matters more than it looks. Conversions often land days after the click, so the most recent days of any 30-day window are under-reported. A raw average pulled from the dashboard therefore overstates what you are really paying. If you work the number out yourself, allow for that lag.
The bigger mistake is starting below the average. A target is not a price cap. It tells Smart Bidding what level of efficiency to hold, and the system bids up or down in each auction to average out at that figure. If you ask for an efficiency the campaign has never delivered, it stops competing in auctions it doesn't expect to win at that price.
Google's own warning is that a target set too low "may cause you to forgo clicks that could result in conversions, resulting in fewer total conversions." In practice, that looks like a campaign that hits its target on paper while spending a fraction of its budget and delivering a fraction of its conversions.
Starting at the average asks the bidding system to repeat what it already does on day one. Every move after that is made from evidence, not hope.
Where does break-even fit in?
Break-even is the most you can pay for a conversion before it loses money. It is revenue per conversion, minus the cost of delivering what you sold, minus the profit you need to keep. For lead generation, revenue per conversion is deal value multiplied by the share of leads that close. The full working is in our guide to working out your break-even cost per lead. The wider question of how much you can afford to pay for a lead sits above both.
In Target CPA, break-even sets the ceiling, and your starting target should sit at or below it. If your 30-day average is already above break-even, the target isn't the problem. The campaign is, and the fix lives in keywords, offer or landing page. Dropping the target straight to break-even only shrinks volume while the economics stay broken.
Two adjustments keep the ceiling honest.
Compare like with like. Google Ads can only bid on the conversions it can see. Phone orders, offline sales and conversions that return later through organic or direct visits all make the reported CPA look worse than the true one. Suppose tracking captures only half of what the campaign drives. In that case, a reported CPA sitting at break-even means you are actually earning margin.
Count repeat value. For businesses where customers buy again, a thin margin on the first order can support a higher ceiling once you count what the customer is worth over time. That only holds if you also count what it costs to keep serving them.
What has to be in place first?
Conversion tracking. This is a hard requirement, because Google won't let you set up the strategy without it. Working tracking is not the same as tracking the right thing, though. Target CPA optimizes toward whatever you count as a conversion. Duplicate counts, page-view goals, or a newsletter signup counted alongside a demo request all teach the system to buy the cheapest action on the list. Check which actions are set as primary before you pick a number.
Volume. Google says you can start Target CPA with no conversion history. You can, but a 30-day average built on a handful of conversions is noise. If the campaign converts only a few times a month, run Maximize Conversions until volume builds and the CPA stops swinging week to week. Then switch with a real average in hand.
Budget headroom. Google recommends being comfortable spending up to twice your average daily budget on a given day, within your monthly charging limit. If your daily budget only buys one or two conversions at the target, the system has little room to find the auctions worth winning.
Equal conversion value. A single CPA target treats a large order and a small one as identical. If order values or lead quality vary widely, Target ROAS with conversion values will usually serve you better than any CPA you could choose.
How do you lower the target over time?
You set the target in the campaign's bid strategy settings. Where economics differ, you can override it for individual ad groups. Once it's live, tighten it gradually. Lower it in small steps and let each change run through enough conversions to show its effect before you make the next one.
Don't judge a step on CPA alone, because a tighter target almost always produces a lower one. Watch whether conversions and spend fall with it. If CPA improves while conversions drop and budget goes unspent, you have pushed past what the market will give you at that volume. The previous step was the right one, so go back to it.
Actual CPA also moves for reasons the target doesn't control. Changes to your site, new ads and heavier competition in the auction all shift it, and none of them mean the target is wrong.
Catching the turn early means looking at the right numbers daily, not at month end. SproutMe Companion sends a daily brief on what changed in spend and conversions overnight. A target that has started starving the campaign then shows up as a falling conversion count against live spend, rather than as a missed number in next month's report.
Timing matters this year too. Google says that from August 17, 2026, it is updating its bidding systems for campaigns limited by budget. Target CPA campaigns may see temporary performance and traffic fluctuations as a result. If yours is budget-limited, don't read a wobble in that window as proof your target is wrong. Hold off on tightening until performance settles.
Conclusion
Your first target CPA should be the cost per conversion the campaign already achieves, with conversion delays allowed for. Break-even sets the ceiling, history sets the start, and conversion volume tells you when to stop lowering. Get tracking and budget right before you switch. Then move in small steps and judge each one on what happened to conversions, not just on the CPA it produced.
See how SproutMe Execute adjusts bids and budgets continuously from performance data, within spend limits a human approves.
Frequently Asked Questions
Target CPA is an average across conversions, not a cap on each one, and you still pay per click. Google states that actual CPA depends on factors outside its control, such as changes to your website or ads and increased competition in the auction. Short gaps are normal. A persistent gap usually means the target sits below what the campaign can realistically deliver.
Yes. In Google Ads, open Ad groups under the Campaigns menu and find an ad group using a Target CPA strategy. Select the figure in its Target CPA column, enter the ad group's own target and save. The strategy must already exist. Targets can differ by ad group, but the bid strategy itself is controlled at portfolio level.
No. Starting in June 2026, Google Ads is relabeling "Maximize conversions with a Target CPA" as simply "Target CPA." Google states that the underlying bidding behavior remains exactly the same. If the new name appears in your account, nothing changes in how you should set, tighten or judge your target. It is a label change, not a strategy change.
Get a complimentary audit to uncover AI opportunities hidden in your data.
Put these strategies to work


