Maximise conversion value with or without target ROAS: which is better?

The choice depends on which number in your plan is fixed

The usual advice ties the choice to data volume and account maturity. Those are prerequisites, not a decision rule. I think a different question is the real one: what is allowed to move in your plan, the advertising budget or the return?

According to Google, Maximize conversion value is meant for the case where you want as much value as possible within the daily budget. Target ROAS adds an efficiency limit on top. A simple mapping follows:

The third row is a typical planning error. A budget that has to be spent and a return that must not be undercut only fit together by coincidence. The strategy enforces one of them, the other is an outcome.

Without a target, the campaign buys down to the last euro

Without a target, the strategy tries to spend the full average daily budget, according to Google's help page. If you were clearly below it before, you may spend noticeably more after the switch. ROAS is not a requirement here but an outcome, and it fluctuates with every budget change.

The problem is the average. The report shows ROAS across all spend, and the cost of the last euros is not visible in it. A worked example with round, invented numbers:

Suppose a shop has a break-even ROAS of 3. The campaign spends €10,000 a month and generates €50,000 in revenue, so ROAS 5. The first €8,000 bring €46,000 (ROAS 5.75), the last €2,000 only €4,000. The marginal ROAS of those last €2,000 is 2, below break-even.

The report shows ROAS 5 and looks good. In commercial terms the shop loses money on the last €2,000. Without a target this is not a flaw of the strategy but its job: it is meant to spend the budget. If you work without a target, you therefore steer return through the budget level yourself. How to determine the lower limit for that is covered in Calculating break-even ROAS.

With a target, the budget is only a cap

With target ROAS, the target decides which auctions are worth it, and the budget only limits from above. For the target to steer, the budget needs room. For budget-limited campaigns with a target, Google recommends a daily budget comfortably above average spend.

For budget planning this means that with target ROAS the budget in the account is no longer a forecast. Spend results from the demand that meets the target. In a weak week the campaign spends less, in a strong one more. If you have to report a fixed monthly budget internally, clarify that beforehand, otherwise the under-delivery looks like a problem although it is intended.

You get more volume with target ROAS through a lower target, more efficiency through a higher one. After every change, Google advises waiting one to two conversion cycles. How to derive the value is described in Setting a target ROAS.

The loose target with a tight budget no longer works

A common hybrid was a deliberately low target as a safety net, plus a tight budget that capped spend. The campaign then regularly beat the target, because the budget confined it to the best auctions.

Since the update of August 2026, budget-limited target ROAS campaigns, according to Google, steer more consistently towards the set target, including when budgets change. If you used to beat the target, you will land closer to it. Google does not adjust targets or budgets automatically in the process. Optmyzr concludes that deliberately low targets used as a buffer lose their effect. Details of the changeover are in Budget-limited campaigns with target ROAS since August 2026.

For the choice between the two strategies, that is the real point. The hybrid spared you the decision, and now you have to make it. For inflexible budgets where you accept swings in return, Google explicitly names switching to Maximize conversion value without a target as an option. If you want to secure a return instead, the target should hold the real lower limit and not a value with a haircut.

When to start without a target

I would start without a target if at least one of these conditions applies:

• The budget is fixed and has to be spent, for example a test budget or an amount approved for a promotion.
• The campaign has too little data for a target. For Search and Shopping, Google cites at least 15 conversions in 30 days and conversion values over one to two conversion cycles. More in How many conversions target ROAS needs.
• The budget is so tight that the campaign only gets the strongest auctions anyway and ROAS sits well above your lower limit.

Without a target does not mean without control. You regularly check whether ROAS is above the lower limit, and take budget away as soon as it approaches it. Keep the budget rules in mind: according to Google, a single day may cost up to twice the daily budget, and a month at most the daily budget times 30.4.

When to set a target

I would set a target if the campaign has enough data and a return should not be undercut. This typically concerns campaigns that are meant to scale: there the budget is not the scarce thing, the question is how much additional revenue is still worth it.

One limit to this rule: target ROAS protects return only as well as the values it bids on. If the conversion values are wrong, for example because returns or margin differences are missing, the target secures a wrong number. That applies to both strategies, but it is less noticeable with a target because it looks like control. What Smart Bidding optimises here and what it does not is covered in Smart Bidding for online shops.

What the choice means for budget allocation between campaigns

When several campaigns run side by side, the strategy changes who does the allocation. With Maximize conversion value without a target, you allocate yourself: each campaign spends what it has, and ROAS is a consequence of your budget split. If one campaign runs at ROAS 8 and another at ROAS 4, that says little about where the next euro brings more. Only a budget shift and the comparison afterwards show that.

With target ROAS and sufficient budget, demand does the allocation at its core: each campaign takes what its target allows. Your lever is then the target per campaign, not the budget. An account in which some campaigns run without a target and others with one is possible, but needs a clear reason for each campaign. Otherwise additional budget ends up where there is no return limit.

Where the naming stands

From June 2026, Google is gradually renaming the former strategy "Maximize conversion value with target ROAS" to Target ROAS. According to Google this is a pure renaming, bidding behaviour stays the same and nothing needs doing in the account. The variant without a target keeps its name and behaviour. During the changeover, old and new names can appear mixed, so reports may split by strategy name. If you filter analyses or scripts by these names, catch both labels.