What Is Cost Per Acquisition (CPA)? Formula, Profitable Limits and Target CPA

What is cost per acquisition (CPA)?

Cost per acquisition (CPA) is the average amount you pay in advertising for each conversion, such as a sale, a lead form, a call or a sign up. You calculate it by dividing total ad spend by the number of conversions. For example, $2,000 in spend and 40 sales give you a CPA of $50.

The definition looks simple. In practice, it is the metric I see misread most often. I have managed ad accounts since 2012, and the pattern repeats: two businesses report the same CPA, one makes money and the other loses it. That happens because CPA means little on its own. You have to read it next to order value, gross margin and close rate.

In this guide I cover the CPA formula, how to find a profitable CPA limit, how Target CPA bidding works in Google Ads and the steps we use to bring acquisition costs down. The numbers are worked examples, so you can swap in your own data and follow the same logic.

Is CPA the same as cost per conversion?

Yes, in day to day use they mean the same thing. Google Ads shows the metric as "Cost / conv." in its reports. Google also expands CPA as "cost per action" in some help pages, while marketers usually say "cost per acquisition". Action or acquisition, the maths stays identical.

Confusion usually starts inside the team. The agency report says CPA, the ad platform says cost per conversion, and finance says "ad cost per customer". All three can refer to the same number, but they can also hide different definitions. For instance, one person may count leads while another counts paying customers.

So my first advice is simple: agree on one term and one definition in the first meeting. Write down which action counts as a conversion and which date range you use. If you want the wider vocabulary in one place, the digital marketing glossary covers the main terms across Google and Meta.

How do you calculate cost per acquisition?

The core formula is: CPA = total ad cost / number of conversions. Google Ads Help defines cost per conversion the same way: total cost divided by the figure in the "Conversions" column.

A few quick examples make it concrete:

• You spent $4,000 in a month and received 80 lead forms. CPA = 4,000 / 80 = $50.
• You spent $700 in a week and closed 14 sales. CPA = 700 / 14 = $50.
• You spent $1,200 on a campaign and got zero conversions. In that case there is no CPA to calculate, and the report shows a blank or a zero.

The last example matters. A campaign with zero conversions is not "cheap"; it is the most expensive campaign in the account. Also remember that CPA is an average. Within the same month, one sale might cost $15 and another $140. Therefore, I recommend working with at least two or three weeks of data before you make a decision.

If you prefer not to do the maths by hand, start with the conversion rate calculator and then read the result next to your spend.

Which conversions does Google Ads use for CPA?

Google Ads counts only primary conversion actions in the "Conversions" column. According to the official conversion tracking help page, this column generally feeds bidding optimisation. The "All conversions" column, on the other hand, includes primary and secondary actions plus special sources such as view through conversions.

The practical consequence is huge. Your cost per conversion depends entirely on which actions you mark as primary. For example, if "add to cart" counts as a primary conversion, the number of conversions rises, CPA drops and the report looks great. Yet no extra money reaches the bank. That is why the conversion actions list is the first screen I open in any new account.

Google also notes that this column may include modelled conversions. As more users decline cookies, the modelled share grows. In other words, the CPA in your ad account and the CPA you derive from your CRM may not match exactly. Setting up enhanced conversions is a good first step to close that gap.

How do CPC and conversion rate drive CPA?

CPA is the result of two inputs: cost per click and conversion rate. The formula is CPA = average CPC / conversion rate. For example, a $2 average click and a 4% conversion rate give you 2 / 0.04 = $50.

This formula always reminds me of one lesson: there are two doors to a lower acquisition cost. The first is cheaper clicks, and the second is a site that converts better. Most advertisers push only on the first door. However, lifting conversion rate from 4% to 6% at the same click price cuts the cost from $50 to roughly $33.

Cheaper clicks also tend to bring weaker traffic. You loosen match types, irrelevant searches creep in and conversion rate falls. As a result, CPC goes down while CPA goes up. So track both numbers together, never one in isolation. The CTR calculator helps if you also want to check click through rate on the same data.

What is the difference between CPA, CPL, CAC and ROAS?

These four abbreviations get mixed up because they all link cost to an outcome. Still, each one answers a different question. The table below shows the difference at a glance:

In short, CPA looks through the lens of the ad platform, while CAC looks through the lens of the company. A $50 CPA can sit next to a $120 CAC because CAC also includes salaries, software and sales time. I collected the full set of metrics in the article on digital marketing KPIs.

How do you convert CPA into ROAS?

When every conversion has roughly the same value, the two metrics convert into each other. The formula is CPA = average order value / ROAS. For example, a $200 average order and a ROAS of 4 mean you pay $50 per conversion.

You can also work backwards: ROAS = average order value / CPA. That way a team that says "our CPA is $60" and a team that says "our ROAS is 3.3" can talk at the same table. For a quick check, use the ROAS calculator.

There is a limit, though. If basket sizes vary widely, an average cost misleads you. A $20 T shirt and an $800 coat both count as "one conversion". In that situation, value based management with ROAS gives a truer picture. For single price services and lead generation, CPA stays the more practical metric.

How do you find a profitable CPA?

A profitable CPA is any acquisition cost that stays below the gross profit you earn from one conversion. The first step is the break even point: break even CPA = average order value × gross margin. If ad cost per sale rises above that figure, every sale loses money.

Here is a worked example:

1. Your average order value is $150.
2. After product cost, shipping, payment fees and a returns allowance, your gross margin is 40%. One order therefore earns $60 in gross profit.
3. Your break even CPA is $60. At that level, advertising leaves zero profit.
4. If you want at least $20 net profit per order, your target CPA should be $40.

The most common mistake is to treat revenue as profit. Spending $90 in ads on a $150 order feels fine at first glance, but the margin maths shows a loss. On the other hand, repeat purchases change the picture. If customers come back, you can base the limit on gross profit over their first 6 or 12 months instead of the first order. Google's own return on investment example also measures what remains after costs, not raw revenue.

How should lead generation businesses set a target CPA?

In service businesses, ads do not produce sales directly; they produce leads. So you build the calculation backwards from the sale. The formula is: maximum cost per lead = gross profit per customer × lead to customer close rate.

For instance, suppose one customer brings $2,000 in revenue with $600 gross profit, and your sales team turns 20% of leads into customers. Then you can pay up to 600 × 0.20 = $120 per lead. Anything above that number loses money.

The weakest link here is the close rate. Most businesses guess it instead of measuring it. If the real rate turns out to be 10% instead of 20%, the limit drops to $60. Therefore, I recommend logging every lead in a sheet or CRM and marking which ones turn into sales for at least two months. I explain how to choose the right goal in the article on setting website conversion goals.

How does Target CPA bidding work in Google Ads?

Target CPA is an automated bid strategy in Google Ads. You set the average cost per conversion you want, and the system adjusts bids in each auction based on the likelihood of a conversion. The goal is to win as many conversions as possible at that average cost.

The official Target CPA help page adds an important caveat. Some conversions will cost more than your target and some will cost less; overall, Google Ads tries to keep the average close to it. Factors outside Google's control, such as changes to your website or rising competition, can also move the result.

There is a naming update too. From June 2026, Google relabels "Maximize conversions with a target CPA" as simply "Target CPA". According to Google, the behaviour of the strategy stays the same and you do not need to take any action in the account.

The strategy works in Search, Display, Video, Shopping, App and Demand Gen campaigns. You can select it when you create a campaign, switch to it from campaign settings or set it up under bid strategies in the shared library.

How many conversions does Target CPA need?

According to Google, you can start using Target CPA with no conversion history at all. However, Google recommends that you evaluate performance over the last 30 days, with at least 30 conversions. So starting is easy; reading the results needs data.

My approach in accounts with very few monthly conversions is to feed the system first. For example, I also measure a meaningful action close to the sale, not just the sale itself. That way the algorithm collects signal faster. Whether that extra action becomes primary is a separate decision, though. If the wrong action becomes primary, the reported CPA falls for the wrong reason.

Google also explains how it builds the recommended target: the system uses the average CPA from the last 30 days, adjusted for conversion delays. If no history exists, the recommendation follows your business goals. Consequently, starting from your real 30 day average is safer than picking a number that "feels right".

What happens if your target CPA is too low?

If the target sits too low, the system passes on clicks that could have converted. In Google's words, this can lead to fewer total conversions. Put simply, you ask for cheaper conversions and end up with fewer of them.

I see this most in accounts under budget pressure. The advertiser says "$50 is too much, let's move to $30" and cuts the target in one step. Then impressions fall, spend stays well below budget and conversion volume halves. The average cost on the report may improve, but the business wins fewer customers.

I prefer to move in stages:

• Start the target at your real 30 day average.
• Change it in small steps and collect a few weeks of data after each one.
• If spend stays far below budget, loosen the target a little.
• Watch conversion volume and cost together, not the average alone.

Google also advises against setting bid limits with Target CPA, because limits can restrict automatic optimisation.

Should you use a standard or a portfolio bid strategy?

You can apply Target CPA in two ways. A standard strategy runs on a single campaign. A portfolio strategy groups several campaigns under one shared target; campaigns, ad groups and keywords then optimise together toward it.

Which one should you pick? A portfolio works well when campaigns sell the same product with similar margins. That way low volume campaigns benefit from the shared data pool. On the other hand, one shared target makes no sense when campaigns sell products with different margins. A product with $20 margin and a service with $200 margin should not share the same target.

My practical rule is simple: I group campaigns whose target costs sit close together and keep the others separate. For budget planning across campaigns, the Google Ads budget calculator gives you a solid starting point.

Why does CPA suddenly jump?

A sudden rise usually comes from measurement or the website, not from the ads. These are the first things I check:

• Conversion tag: If a site update broke the tag, sales continue but the platform stops seeing them, so the cost spikes overnight.
• Conversion delay: If users click today and buy three days later, the last few days look expensive for a while.
• Site problems: A checkout error or a slow page lowers conversion rate.
• Competition and seasonality: Click prices rise in busy periods, and conversion cost follows.
• Campaign changes: New targeting, new keywords or a strategy switch can make the system unstable for a while.

Rule these out one by one before you touch the target. Otherwise you may treat a tracking error as a bidding problem and break a campaign that works. For a quick health check, try the Google Ads audit tool.

How do you lower your cost per acquisition?

The lasting way to lower your cost per acquisition is to improve the ads, the targeting and the website together. Here are the steps I follow in order:

1. Fix measurement: Make sure your primary conversion actions really bring money.
2. Clean search terms: Add searches without buying intent to your negative lists.
3. Speed up landing pages: A slow, cluttered page lowers conversion rate directly.
4. Match offer and message: The promise in the ad and the offer on the page should be the same.
5. Shorten forms: Every unnecessary field reduces the number of people who finish.
6. Isolate weak segments: Review devices, locations or hours that stay expensive over time.

All of these steps either raise conversion rate or cut wasted clicks. In other words, they act on both sides of the formula. If you want to see what works on the page side, read about conversion focused web design.

How does a wrong conversion setup distort CPA?

A wrong conversion definition lowers CPA on paper while it hides the real cost. The three errors I meet most often: counting page views as conversions, counting the same form twice and treating a click on a phone number as a real call.

Imagine every click on a WhatsApp button counts as a primary conversion. The platform then reports a low cost. Yet some of those people never send a message. The real cost per lead can end up several times higher than the reported figure. So before you make any action a conversion, ask one question: when this action happens, does the chance of money reaching the business rise meaningfully?

Also check the counting setting. For sales, where every repeat has value, counting every conversion makes sense. For forms, where a second submission from the same person adds nothing, counting one per click gives a truer result. This small setting changes the report noticeably.

How do you read CPA in Meta Ads?

The logic in Meta Ads stays the same, but the label differs. Ads Manager usually shows "cost per result", and the result depends on the campaign objective: a purchase in a sales campaign, a form in a leads campaign.

The key point is not to compare Google and Meta CPA side by side. The two platforms use different attribution windows and different counting methods. Both may claim the same sale. Therefore, I recommend calculating your blended acquisition cost from your own sales data rather than from any single ad platform.

New Meta campaigns also deliver unstable costs in their first days because the system is still learning. Frequent edits in that period extend the process. I cover this in detail in the article on the Meta Ads learning phase.

When is CPA misleading?

CPA works best for short sales cycles and a single type of conversion. In some situations, though, it pushes you toward the wrong decision:

• Mixed basket sizes: Cheap and expensive orders both count as one conversion; ROAS reads this better.
• Long sales cycles: In B2B, the form arrives today and the deal closes three months later. A low cost per form can hide weak lead quality.
• High repeat purchase: A cost that loses money on the first order can be profitable over a customer's first year.
• Brand search in the mix: Brand campaigns bring very cheap conversions and make the account average look better than it is.

The last point deserves attention. People who search your brand name would likely have found you anyway. That is why I report brand and non brand costs separately. Otherwise you cannot see what winning a new customer really costs.

Why does CPA rise as the budget grows?

A modest rise in cost as you scale is normal. The system first collects the most likely conversions, which are also the cheapest. Extra budget then reaches users with weaker intent, and they cost more. This effect has a name: diminishing returns.

For example, CPA might sit at $50 on a $100 daily budget and move to $60 when you double the budget. That is not a failure. The real question is whether the additional conversions from the extra spend still come in below your profitable limit. The marginal cost of those extra conversions matters more than the average.

So I recommend raising budgets in steps. After each step, calculate how many extra conversions you gained and what they cost. Once you approach the profitable limit, look for growth in conversion rate rather than in more spend. For the budgeting logic behind this, see the guide on setting a Google Ads budget.

How do we manage cost per acquisition targets with clients?

When my team and I take over an account, we start the CPA conversation with margin, not with a number. In the first week we audit conversion actions, rebuild the primary and secondary split and compare platform data with real sales data.

Next, we calculate the break even CPA together with the business. We set the target below that limit so it leaves room for profit. We start the bid strategy from the real 30 day average and make changes in small steps. Reporting also covers conversion volume, non brand cost and the share of leads that turn into sales, not just the headline average.

The aim is profitable growth, not the lowest possible number. If you would like to set clear acquisition cost limits for your account with us, take a look at our Google Ads management service.