Maximise conversions and target CPA at a glance
Maximise conversions bids in every auction to win the most conversions the budget allows. The Smart Bidding strategies guide covers how the strategies use auction-time signals and what each needs; for this comparison the key point is the one Google documents for the strategy: without a target, it will try to spend the budget. It does not care what each conversion costs, only how many it can buy. That makes the budget the main lever and the cost per conversion an outcome.
Target CPA adds a constraint. The advertiser states the average cost per conversion it wants, and the strategy bids to get as many conversions as it can at or near that average. Google documents that the target is an average rather than a cap, so some conversions will cost more and some less, and that a campaign with a target may spend less than its budget when it cannot find enough auctions to meet it. That makes the target the main lever and the spend an outcome.
The two are not different technologies. In the current interface target CPA is an optional setting on maximise conversions, which is why the decision is less about which strategy to choose and more about whether, and when, to add the constraint. The choice between holding a cost and holding a return is a different one, covered by the comparison of target CPA and target ROAS.
Compare table
Maximise conversions vs Target CPA
| Row label | Maximise conversions | Target CPA |
|---|---|---|
| Constraint honoured | The budget: spend it all for the most conversions | The average cost per conversion: hold it near the target |
| Budget use | Tries to spend the full budget | Spends only what the target allows, which can be less than the budget |
| Data needs | Conversion tracking; works with thin history | Conversion tracking and enough volume to hold an average reliably |
| Cost stability | Varies with competition and budget; no anchor | Anchored near the target over time, with day-to-day variation |
| Volume | As much as the budget can buy | As much as the target allows, often less than an unconstrained campaign |
| Learning behaviour | Learns what the campaign can achieve; establishes a baseline | Learns how to reach the stated average; a new target restarts learning |
| Risk when budget changes | A larger budget buys pricier conversions; a smaller one cuts volume | A larger budget changes nothing unless the campaign was limited by budget |
| Best-fit stage | New campaigns, launches and campaigns with a fixed spend to deploy | Established campaigns with a known, affordable cost per conversion |
The budget-use and risk rows are the ones that surprise people. Under maximise conversions, the budget is a decision about how many conversions to buy at whatever price results. Under target CPA, the budget is a ceiling that only matters if the campaign reaches it; the real decision is the target.
When to choose maximise conversions
Choose maximise conversions when the campaign is new. Without history there is no cost per conversion to hold, and setting a target from guesswork either strangles delivery or sets a cost the business cannot afford. Running without a target for a few weeks shows what the campaign can achieve at the budget, and that observed figure is the starting point for a target later.
Choose it when the budget is what the business wants to control. Some campaigns exist to deploy a fixed sum, such as a launch budget or a seasonal allocation, and the goal is the most conversions from that sum. Maximise conversions does exactly that. The budgets and pacing guide explains how daily budgets and pacing work; under this strategy they are the steering wheel.
Choose it when conversion volume is low. A target needs enough conversions for the strategy to hold an average; with very few, the average swings and the target is either missed or met by withdrawing from auctions. Maximise conversions makes the most of a thin signal by spending where the strategy sees the best chances. When volume stays low even after that, the guide to Google Ads bidding with few conversions covers consolidating campaigns, bidding on a more frequent action and judging over longer windows.
Watch the marginal cost. Because the strategy spends what it is given, raising the budget buys the auctions it had passed on, and each extra conversion usually costs more than the average. Judge budget increases on the cost of the additional conversions, not on the new blended figure.
When to choose target CPA
Choose target CPA when the campaign has a stable cost per conversion and the business knows what it can afford. The target should come from economics, the most the business can pay for a conversion given what that conversion is worth, and the campaign should already be running near it. A target set far below current performance tells the strategy to withdraw from most auctions.
Choose it when cost matters more than volume. Lead generation with a firm ceiling on cost per lead, and sales campaigns where each conversion has a similar value, both benefit from a strategy that holds a cost rather than one that spends whatever it is given. The target also protects the campaign from paying for expensive conversions on days when competition rises.
Choose it when several campaigns need a consistent standard. A shared target across a portfolio of campaigns, which the Smart Bidding guide describes, keeps cost per conversion comparable and lets budget move to where conversions can be bought within the limit. Under a target, a Bid StrategyThe rule an ad platform follows to set the bid in each auction. status reporting "limited by budget" is the signal that more budget would buy more conversions at the target.
Accept the trade. A target caps the average cost by giving up some volume. If the business needs more conversions, the lever is to loosen the target, not to raise a budget the campaign is not spending.
A driving school moves from no target to a target CPA
Mossway Driving School runs a Search campaign for lesson enquiries on maximise conversions with a £1,800 monthly budget. After two months it is spending the full budget for about 90 enquiries a month, £20 each. It works out that it can afford up to £30 per enquiry given how many become paying pupils. It raises the budget to £2,700 and gets 115 enquiries, an average of about £23.50, but the extra 25 enquiries cost £36 each, above its limit. It switches on a target CPA of £24, close to the new average, and leaves the budget at £2,700. After the learning period the campaign spends about £2,350 a month for 100 enquiries, £23.50 each, and no longer uses the full budget. Mossway keeps the target there for a month, then tests a step to £26 to see whether the extra enquiries it buys stay under £30 each.
Decision table
| Situation | Choose | Why |
|---|---|---|
| New campaign with no conversion history | Maximise conversions | No cost per conversion exists yet to hold |
| Fixed launch or seasonal budget to deploy | Maximise conversions | The goal is the most conversions from a set sum |
| Stable cost per conversion and a known affordable ceiling | Target CPA | Holds the cost the business can pay |
| Very few conversions a month | Maximise conversions, or pool campaigns before adding a target | A target needs volume to hold an average |
| Campaign spending its full budget at a cost the business is happy with | Maximise conversions, or a target near the current average | Either works; a target protects against rising costs |
| Target CPA campaign leaving budget unspent | Keep target CPA and loosen it in steps, or accept the volume | The target is the constraint, not the budget |
Moving from one strategy to the other
The usual path is from maximise conversions to target CPA. Run without a target long enough to see a stable cost per conversion, then set the first target close to that observed figure rather than at the economic ideal. Google documents a learning period after significant changes, so allow it to pass before judging the result, and change nothing else during it. If the business needs a lower cost, move the target towards it in steps, waiting for the campaign to settle each time.
The reverse move, removing a target, is sometimes right too. If a campaign under a target has stopped spending and the business wants volume more than it wants the cost limit, removing the target hands control back to the budget. Expect the cost per conversion to rise as the strategy buys auctions it had been declining, and set the budget at the amount the business is willing to spend at that higher cost.
Either way, change one thing at a time and record why. Budget and target changes made on the same day cannot be told apart afterwards. If results fall sharply after a switch, the troubleshooting page for results that dropped after a bid strategy change checks the usual causes, and if a target campaign stops spending, the diagnosis for a budget that is not spending separates a tight target from the other constraints. Good Budget PacingHow fast a campaign spends its budget across the flight compared with the plan. depends on knowing which of the two limits the campaign is under.
Common mistakes
- Setting a target CPA on a brand-new campaign before there is any cost history to base it on.
- Setting the first target far below current performance, which makes the strategy withdraw from most auctions.
- Raising the budget on a target CPA campaign that is not spending, when the target is the constraint.
- Judging a budget increase on maximise conversions by the new average rather than by the cost of the extra conversions.
- Changing target and budget together, so neither effect can be read.
- Reacting to a few days of results during the learning period after a switch.
- Treating the target as a cap. It is an average; individual conversions will cost more and less.
