CAC Payback Period: Why a Healthy Number Doesn't Mean You'll Ever See the Cash Back
A 14-month CAC payback target means nothing if a chunk of that cohort churns in month 7. Here's how churn timing quietly breaks payback math.
Every SaaS finance deck has a slide with CAC payback period on it, usually next to LTV:CAC ratio, usually presented as a single settled number: "our payback period is 14 months." What that slide doesn't show is how many of the customers behind that average were still paying in month 14. If a meaningful chunk of the cohort churned in month 6 or 8, the CAC spent acquiring them was never actually recovered — it just got averaged into a number that still looks acceptable.
That's the number most teams report. Almost nobody reports the companion number: what share of that same cohort was gone before the 18-month mark. Those are two different questions, and only one of them shows up in the standard formula.
What CAC payback period actually measures
The formula is simple: CAC payback period = CAC ÷ (ARPA × gross margin %). Spend $6,000 to acquire a customer paying $500/month at 80% gross margin, and you recover that spend in 15 months ($6,000 ÷ $400 = 15). It's a cash-flow metric, not a profitability metric — it answers "how long until this customer has paid back what it cost to get them," not "how much are they worth over their lifetime." That second question is what LTV answers, and the two get confused constantly because they're built from the same three inputs (CAC, ARPA, margin) plus one more that only LTV uses directly: churn rate.
That's the tell. LTV is built with churn baked into the formula from the start — LTV = (ARPA × gross margin) ÷ churn rate. The standard payback formula has no churn term at all. It's a static division that assumes the customer is still around at month 15, without checking whether that assumption holds for your actual base.
The 2026 benchmarks
| Stage | Typical CAC payback | Investor bar for "efficient" |
|---|---|---|
| Bootstrapped | ~5 months | N/A — self-funded |
| Seed | 4-5 months | Under 6 months |
| Series A | 10-12 months | Under 12 months |
| Series B ($10M-$50M ARR) | 14-18 months | Under 14 months |
| Series C+ | 18-24 months | Under 18 months |
Compiled from the Bessemer BVP Cloud Index, KeyBanc / Sapphire SaaS Survey, and High Alpha SaaS Benchmarks (2025-2026).
Payback also moves a lot with contract size, which is worth knowing before you compare your number to a peer's. OpenView's 2026 data puts sub-$5K-ACV businesses at an 11-month median, while $50K-$100K enterprise deals run closer to 22 months, because longer sales cycles and field-sales cost eat further into the recovered margin before the deal even closes.
Source: OpenView SaaS Benchmarks 2026.
The assumption baked into every payback number
Say your blended payback period is 14 months and you consider that healthy — it beats the Series B bar above. Now suppose your monthly logo churn runs 3%, which is a perfectly normal number for a mid-market SaaS product. Over 14 months, compounding monthly churn at 3% means roughly 35% of a starting cohort is already gone before that 14-month mark arrives. Those customers' CAC was never recovered. The 14-month number isn't wrong, exactly — it's an average across everyone, including the customers who paid back their CAC twice over and the customers who churned in month 4 and paid back none of it. The average can sit at a comfortable 14 months while a third of the spend behind it never comes back at all.
This is the same mechanism we cover in our piece on early-lifecycle churn: most SaaS churn isn't evenly spread across a customer's tenure, it's front-loaded into the first 90 days. That front-loading is exactly what breaks a blended payback number, because it means the customers churning early — the ones whose CAC is never recovered — are disproportionately represented in months 1 through 3, not spread evenly across the full payback window the formula assumes.
Why the blended number hides this
Blended CAC payback averages together customers acquired across many months, at whatever their churn rate happened to be at each point. A cohort acquired eighteen months ago that's mostly paid back looks the same in the blended average as a cohort acquired three months ago that's already lost 15% of its customers to early churn. Both get folded into one trailing number, and the number can hold steady even as the newer cohorts are getting structurally worse, because the older, mostly-recovered cohorts are still propping up the average.
The fix is to stop measuring payback as a company-wide average and start measuring it per cohort. Take the customers acquired in a single month, track their cumulative revenue against what it cost to acquire them, and separately track what percentage of that specific cohort has already churned by the month your target assumes they'd still be paying. If 20% of a cohort is gone by month 6 and your payback target is month 14, you don't have a 14-month payback period for that cohort — you have a 14-month payback period for the 80% who are still around, and a permanently unrecovered CAC balance for the 20% who aren't. Our guide to reading cohort retention curves covers building this kind of month-by-month comparison in more depth.
What this means for how you fix it
Once you're looking at cohort-level payback instead of a blended number, there are really only two levers, and they behave differently:
- Shorten the payback period itself — raise ARPA, improve gross margin, or lower CAC. This helps every customer equally, whether they were going to churn early or not, and it's the lever most finance teams reach for first because it's the one the formula directly exposes.
- Reduce early churn specifically — this is the lever that actually protects the CAC you already spent, because it targets the customers most likely to leave before payback completes rather than improving the economics of everyone including customers who were never at risk.
The second lever is usually the higher-leverage one for a company whose payback number looks acceptable on paper but whose cash position doesn't match it, because it's fixing the mismatch between the formula's assumption and reality rather than just making the formula's number smaller. Catching a subscriber who's about to cancel in month 3 or 4 — before their CAC has come close to paying back — is worth structurally more to your payback math than catching one who's about to cancel in month 20, even though both show up identically as "one lost customer" in a blended churn rate. A cancellation flow that captures a cancel reason and responds with the right save offer, covered in our breakdown of why customers actually cancel, is one of the few interventions that specifically targets pre-payback churn, because it intervenes at the exact moment a customer who hasn't yet recovered their CAC is deciding whether to leave.
If you want to see how sensitive your own payback number is to churn timing, our CAC payback period calculator and LTV calculator are both worth running side by side — plug in a lower early-churn assumption and watch how much faster the real, cohort-level payback period gets, even holding CAC and ARPA constant. That gap between the two numbers is roughly what a cancellation flow focused on early-tenure saves is worth to you. It's also the reason "reduce churn" and "improve unit economics" aren't two separate initiatives at most SaaS companies — they're the same fix showing up on two different slides in the same deck.
Frequently asked questions
What is a good CAC payback period for SaaS in 2026?+
The blended median for $5M-$25M ARR SaaS companies is 18 months as of 2026, flat versus late 2025 but up from 15 months in 2023, per OpenView's SaaS Benchmarks. Investors now look for growth-stage companies to run under 12 months as a capital-efficiency signal, and a Series B company at $10M-$50M ARR is considered healthy under 14 months. Contract size moves the number a lot: sub-$5K ACV businesses post an 11-month median, while $50K-$100K enterprise deals run closer to 22 months because of longer sales cycles and field-sales cost.
What's the difference between CAC payback period and LTV:CAC ratio?+
LTV:CAC ratio tells you whether a customer is worth more than they cost to acquire over their entire relationship with you — a long-term profitability check. CAC payback period tells you how many months it takes to get that acquisition cost back in cash, independent of how much more the customer is worth after that point. A business can have a healthy 4:1 LTV:CAC ratio and still run out of cash, because LTV:CAC says nothing about timing and payback period is entirely about timing.
How does churn rate affect CAC payback period?+
The standard payback formula (CAC ÷ (ARPA × gross margin)) assumes the average customer sticks around long enough to reach that month. It doesn't account for customers who churn before then. If your churn is front-loaded — concentrated in the first few months, which is where most SaaS churn actually happens — a meaningful share of the cohort behind your "14-month payback" number churns before month 14 and never pays that CAC back at all. The blended average can look fine while a real chunk of spend is never recovered.
Should I calculate CAC payback using blended or cohort-based numbers?+
Cohort-based. A blended payback period averages together customers acquired months or years apart, at different churn rates and different points in their lifecycle, into one number that can look stable even while the underlying cohorts are getting worse. Track a single acquisition cohort's cumulative revenue against its CAC month by month, and separately track what percentage of that cohort has already churned by the month your payback target assumes they'd still be paying. That second number is the one blended payback hides.
Stop losing subscribers today
One script tag. One function call. A live cancellation flow in under 10 minutes.
Start free trial →