Target ROAS vs Target CPA: What Each Needs and How Long to Wait

Target CPA and Target ROAS look similar but optimise different things and need different data. Here's what each requires, where Google's own pages give different numbers, and how to choose.

Ömer S. Ekinci
CEO, Oxford Impulse
Published
7 min read

You’ve got conversion tracking running, a campaign that’s been live for a few weeks, and a bid strategy dropdown with two options that sound almost the same. Target CPA or Target ROAS? Then you open Google’s help pages to settle it and find numbers that don’t quite match each other. This post sets the two strategies side by side, shows you where the pages disagree, and doesn’t pretend one of them is simply better.

The short answer

Target CPA aims to get you as many conversions as possible at an average cost per conversion that matches your target, and it needs conversion tracking set up first[2]. Target ROAS aims for a return on ad spend you set, based on predicted conversion value, so it needs conversion tracking that reports values too[1]. One is about the cost of each conversion. The other is about the value those conversions bring back.

What each strategy optimises

Target CPA (cost per acquisition, meaning what you pay on average for one conversion) is about cost. You give Google a number, say what you’re happy to pay per sign-up or per order, and the system tries to win as many conversions as it can while the average cost lands near that figure[2]. Note the word average. Some conversions will cost more than your target and some less.

Target ROAS (return on ad spend, meaning conversion value divided by ad cost) is about value. You set the return you want, and Google uses predicted conversion value to decide how much to bid[1]. A small order and a large order stop being the same thing.

That difference matters most when your conversions vary a lot in size. If every lead is worth about the same to you, counting conversions is a reasonable proxy for value. If a basket can range from small to large, a strategy that treats them as equal is working with less information than you actually have.

Tip

Write down, in one sentence, what a conversion is worth to your business before you open Google Ads. If the answer is \”about the same each time\”, that points one way. If it’s \”it depends hugely\”, that points the other.

What each one needs

Target CPA

The requirement Google states is conversion tracking, set up before you switch the strategy on[2]. Google also says you can start Target CPA without any conversion history[2]. For judging how it’s doing, the Target CPA page gives a benchmark: look at the last 30 days and aim for at least 30 conversions[2].

Target ROAS

Here the bar is higher. You need conversion tracking that records values, not just counts. For Search and Shopping campaigns, Google’s Target ROAS page asks for 15 or more conversions in the last 30 days[1].

Watch out

The Target CPA page asks only for conversion tracking[2]. It doesn’t mention conversion values, so don’t assume you need value tracking set up just to use it. The value requirement belongs to Target ROAS.

Value-based data: three weeks or four?

Google’s page on value-based bidding for Search and Shopping says you need about 3 weeks, or 1 to 2 conversion cycles, of value data before you activate it[3]. The Target ROAS page itself says 4 weeks[1]. Both are official Google pages and they give different figures.

We won’t tell you which one is right, because neither is marked as the authority. What you can do is treat the gap as a range. If your account has less than 3 weeks of value data, wait. If it has between 3 and 4 weeks, you’re in the grey zone where Google’s own pages disagree, so check that your values are accurate and stable before you commit. A conversion cycle, by the way, is the typical time between someone clicking your ad and converting. For a quick, low-priced purchase it can be short. For a considered purchase it can be much longer, which is why the cycle-based wording exists.

If you’re also weighing campaign types, our comparison of Performance Max and standard Shopping campaigns covers that choice separately at https://oxfordimpulse.com/performance-max-vs-standard-shopping/ .

How long to give it

This is where the numbers scatter most. Start with learning, which is the period when the system is adjusting to a change. Google says learning status never lasts more than 7 days. Major structural changes can trigger it, and after you set a new goal, performance can take 1 to 2 conversion cycles to settle[4].

Those are two different clocks. Learning is the short one, capped at 7 days. Settling is the longer one, tied to how long your customers take to convert. A strategy can leave the learning status and still be finding its feet.

Now the evaluation windows. Google’s pages give these:

  • 2 conversion cycles or 2 weeks, on the Smart Bidding learning page[4]
  • 1 to 2 full conversion cycles, on the page about evaluating automated bid strategy performance[6]
  • 30 days with at least 30 conversions, on the Target CPA page[2]

They overlap but they’re not identical. If your conversion cycle is short, \”2 conversion cycles\” is small next to 30 days. If your cycle is long, it’s the other way round. We’re not picking one for you. What matters is that you pick a window before you change anything, write it down, and don’t judge the result a few days in. Judging too early risks cutting a strategy off before it has settled.

Tip

While you wait, avoid major structural changes. They can trigger learning again[4], and then you’ve restarted the clock without noticing.

The June 2026 name change

You may have seen the labels \”Maximize conversions with a Target CPA\” and \”Maximize conversion value with a Target ROAS\”. From June 2026, Google relabels them as plain Target CPA and Target ROAS[5].

This is a change of labels only[5]. It’s not a new strategy. Where your documents, training notes or reports use the old long names, update them so your team isn’t searching for options under words that no longer appear.

How to choose: Target ROAS vs Target CPA

Don’t ask which is better. Google’s pages don’t rank them, and neither will we. Ask two questions instead: what’s your goal, and what data do you have?

Start with the goal

  • If you want conversions at a steady average cost and each conversion is worth about the same, Target CPA matches that goal[2].
  • If you want a particular return from spend and your conversions differ in value, Target ROAS matches that goal[1].

Then check the data

  1. Is conversion tracking working? Both strategies need it[1][2].
  2. Do your conversions report values? If not, you don’t meet the stated requirement for Target ROAS[1].
  3. Do you have 15 or more conversions in the last 30 days? For Search and Shopping, that’s Google’s stated floor for Target ROAS[1].
  4. Do you have around 3 to 4 weeks of value data? The pages say 3 weeks[3] and 4 weeks[1], so aim for the longer one if you can, and know that’s your own call.

If you fail step two or three, Target CPA is the one you can start with, since Google says it can begin without conversion history[2]. That’s about what you can use now, not a verdict on quality. You can move to value-based bidding later when your data supports it.

FAQs

What is the difference between Target ROAS and Target CPA?

Target CPA aims for as many conversions as possible at an average cost per conversion that matches your target[2]. Target ROAS aims for a set return on ad spend using predicted conversion value[1]. The first is about cost, the second is about value.

Does Target CPA need conversion tracking?

Yes. Google says conversion tracking must be set up before you use it[2]. The page doesn’t mention conversion values.

Can I use Target CPA with no conversion history?

Google says you can start Target CPA with no conversion history[2]. For judging how it’s doing, its page suggests looking at the last 30 days and having at least 30 conversions[2].

How many conversions does Target ROAS need?

For Search and Shopping, the Target ROAS page asks for 15 or more conversions in the last 30 days, plus conversion tracking with values[1]. Other pages give different guidance on data timing, so check the current help page for your campaign type.

How long is the Smart Bidding learning period?

Learning status never lasts more than 7 days[4]. Major structural changes can trigger it. After you set a new goal, performance can take 1 to 2 conversion cycles to settle[4].

How long should I wait before judging a bid strategy?

Google’s pages give different windows: 2 conversion cycles or 2 weeks[4], 1 to 2 full cycles[6], and 30 days with 30 or more conversions for Target CPA[2]. Pick one before you start and stick to it. Avoid major structural changes while you wait.

How much value data do I need before switching to Target ROAS?

The value-based bidding page says about 3 weeks, or 1 to 2 conversion cycles[3]. The Target ROAS page says 4 weeks[1]. The pages don’t match, so treat it as a range.

Are the June 2026 name changes a new strategy?

No. From June 2026, \”Maximize conversions with a Target CPA\” and \”Maximize conversion value with a Target ROAS\” are relabelled Target CPA and Target ROAS[5]. Google describes it as a change to labels only.

Where to go from here

Open your account, check whether your conversions carry values and count your conversions over the last 30 days, then pick your evaluation window before you touch the bid strategy. If you’d like a second pair of eyes on the setup, our Google Ads page at https://oxfordimpulse.com/google-ads/ has more.

Sources

Share this article

Written by

Ömer S. Ekinci

CEO, Oxford Impulse Google Partner

Ömer S. Ekinci is the director of Oxford Impulse, a London-based Google Partner agency. He works hands-on in Google Ads, Meta Ads and Google Merchant Center accounts for e-commerce, healthcare and B2B brands across the UK, Europe and the US.

Free audit

Want a second pair of eyes?

Get a free audit and a clear list of what to fix first.

Get a free audit
0% read 7 min left

Target ROAS vs Target CPA: What Each Needs and How Long to Wait

Get a free audit