Skip to main content

Calculating CPA targets for SaaS

Cost per acquisition (CPA) is what one conversion costs in paid spend. The correct CPA target depends on what a conversion is worth to your business. This article shows how to calculate that target from your business economics.

The basic question​

For every dollar you put into Google Ads, how many dollars come back, and over what time?

If you spend $200 to acquire a customer, and they pay you $20,000 in their first year, that is a good CPA. If you spend $200, and they pay you $400 over their lifetime, that is a bad CPA.

The math depends on which conversion you measure (sign-up, demo request, paid customer) and the value behind it.

Step 1: pick the conversion you optimize for​

Three common choices:

  • Customer. This is the closest to revenue. It is the hardest to track in Google Ads, because of the time delay.
  • Trial start or demo request. This is closer to the click. It tracks well, but you must know your conversion rate from trial to customer.
  • Sign-up. This has the cleanest tracking. It is the furthest from revenue.

For most B2B SaaS in the first year of paid search, trial start or demo request is the correct anchor. Track it in Google Ads. Then multiply by your downstream conversion rate to estimate the customer CPA.

Step 2: calculate customer lifetime value​

This is the maximum amount a customer is worth to you over the full relationship.

Simple formula: average revenue per customer per year × average customer lifetime in years.

For a B2B SaaS with $10,000 average annual contract value (ACV) and a 4-year average customer lifetime, LTV = $40,000.

You can refine it:

  • Subtract the gross margin reduction. If the customer costs you 20 percent of their revenue to serve, multiply LTV by 0.80. So the example becomes $32,000.
  • Discount future cash flows. A dollar in year 4 is worth less than a dollar today. Most operators ignore this for early-stage paid search math. It matters more at scale.
  • Add expansion revenue. If customers grow over time (they add seats or upgrade plans), include that.

For lean teams, keep it simple. Use ACV × lifetime, and optionally multiply by gross margin. Do not add complexity until the simple version works.

Step 3: pick your CPA ratio​

CPA cannot be your full LTV, because then you have zero margin. It must be a fraction.

Common start points for B2B SaaS:

  • CPA = 20 to 30 percent of LTV. Fast-growth mode. You spend a lot on acquisition for fast revenue growth, and you sacrifice efficiency.
  • CPA = 10 to 20 percent of LTV. Balanced mode. It is common for series A through C SaaS that want healthy growth with reasonable burn.
  • CPA = 5 to 10 percent of LTV. Efficient mode. It suits mature, profitability-focused SaaS, or a tight capital environment.

For our example with $32,000 LTV at balanced mode (15 percent), max CPA = $4,800. That is the most you pay for a customer.

Step 4: convert customer CPA to your tracked CPA​

Google Ads probably tracks something earlier than the customer (trial start, demo request, sign-up). You must convert it.

If you measure trials in Google Ads, and your trial-to-customer conversion rate is 20 percent, then 1 customer = 5 trials.

Customer CPA target / trials per customer = trial CPA target.

To continue the example: $4,800 / 5 = $960 max trial CPA.

If your campaign delivers trials at $300, you have margin and can scale. If it delivers trials at $1,500, you lose money on every customer the campaign produces.

Step 5: factor in payback period​

LTV math assumes you have time. If you need cash back fast (most early-stage SaaS does), the CPA target gets tighter.

Payback period is how long until cumulative customer revenue is more than CPA.

For our example with $10,000 ACV, a $4,800 max CPA is a 5.8-month payback. This is acceptable for most growth stages.

If you need a 12-month payback or shorter, the max CPA drops:

  • 12-month payback: max CPA = ARR (so $10,000 in our example).
  • 6-month payback: max CPA = ARR × 0.5 ($5,000).
  • 3-month payback: max CPA = ARR × 0.25 ($2,500).

Pick a payback target based on your runway and growth stage. A tighter payback means a tighter CPA.

Put it together​

For our example B2B SaaS:

  • ACV: $10,000
  • Customer lifetime: 4 years
  • Gross margin: 80 percent
  • LTV: $32,000
  • Mode: Balanced (15 percent ratio)
  • Max customer CPA: $4,800
  • Trial to customer rate: 20 percent
  • Max trial CPA: $960
  • Payback at this CPA: ~5.8 months

So the goal in Google Ads is to deliver trial conversions at $960 or below.

Test campaigns that run at $1,500 trial CPA are not profitable. Test campaigns at $400 are good, and you should scale them.

Adjustments by industry​

The example numbers are reasonable for mid-market B2B SaaS. Adjust them for your context:

High ACV, low volume​

If you sell $100,000+ contracts to a small total addressable market, your max CPA can be $10,000+. Google Ads might not be your primary channel (sales-led growth often dominates). But for the volume you do get, you can afford a lot.

Low ACV, high volume​

If you sell $50/year subscriptions, your max CPA is in the tens of dollars. In many B2B SaaS clusters, Google Ads CPCs can be more than your max CPA. This makes the channel too expensive. You run only the cheapest clusters with the best conversion.

Long sales cycles​

If the average sales cycle is 9 months, your "trial to customer" data is late. For a while, you optimize on guesses. Wait for a longer evaluation window before you call a campaign profitable.

Freemium​

If you have a free tier, the "conversion" might be a sign-up. Your conversion rate from free to paid is the connector. Free-to-paid is usually 1 to 5 percent for self-serve SaaS, and lower for broad-category products.

When CPA is not the right metric​

Some campaigns optimize on something other than CPA:

  • Brand campaigns. Optimize on impressions, share of voice, branded query growth.
  • Top-of-funnel content. Optimize on email captures or subscribers, not direct revenue.
  • Retargeting. Optimize on a conversion further down the funnel, because the audience is already qualified.

For these, CPA framing works, but the inputs are different.

A quick framework​

If the math feels like too much:

  1. Pick a target CPA from a rough rule: 10 to 15 percent of the expected first-year revenue from that customer.
  2. Run campaigns to that target.
  3. Adjust every quarter as data builds up.

For a SaaS where customers pay $1,000 a year on average, that is a $100 to $150 max CPA. It is rough but usable.

Next​