CAC (customer acquisition cost)

By Lasse Schou, Founder & CEO of GrowPanel

Reviewed 7 October 2026 · Updated 16 October 2026

Customer acquisition cost (CAC) is the average cost of winning one new paying customer. The formula is total sales and marketing cost in a period ÷ new paying customers acquired in that period. "Fully loaded" CAC includes salaries, commissions and tools, not just ad spend.

Founders rarely run out of ideas; they run out of efficient acquisition. CAC is the metric that tells you whether growth is getting easier, staying stable, or becoming painfully expensive. If CAC drifts up without you noticing, you can wake up six months later with "growth" that quietly destroyed your cash position.

What CAC actually includes

CAC sounds simple, but teams get misaligned because they use different "cost buckets." Decide which version you're using, and use it consistently.

The two CAC definitions worth tracking

1. Paid (or variable) CAC
   Useful for channel optimization and experiments.

   • Ad spend
   • Sponsorships
   • Affiliate commissions
   • Freelancers/agencies tied to acquisition

2. Fully loaded CAC
   Useful for budgeting, hiring, and runway planning.

   • All paid CAC items, plus:
   • Sales team payroll
   • Marketing team payroll
   • Sales commissions and spiffs
   • Sales development (SDR/BDR) costs
   • Tools: CRM, sequencing, intent, data providers
   • Events (if acquisition-driven)

Fully loaded CAC is the classic definition. David Skok's "Startup Killer" essay describes it as your entire cost of sales and marketing over a period, "including salaries and other headcount related expenses," divided by the number of customers acquired in that period.

Two more terms you'll hear:

• Blended vs paid CAC. Andreessen Horowitz's "16 Startup Metrics" distinguishes blended CAC (total acquisition cost divided by all new customers, including organic ones) from paid CAC (divided only by customers from paid channels), and says investors consider paid CAC more important than blended CAC when evaluating the viability of a business.
• Marginal CAC. Fully loaded CAC is an average. Marginal CAC is the cost of the next customer: the extra spend divided by the extra customers it brings in. As you push a channel harder, marginal CAC usually climbs above the average, which is why an average that looks fine can hide spend that no longer pays back.

What to typically exclude from CAC (but document your policy):

• Core R&D/product payroll
• Finance/legal
• Broad G&A rent and admin (unless your model requires it to acquire customers)

The Founder's perspective: If you're deciding whether to hire two AEs or double paid spend, you need fully loaded CAC. If you're deciding whether to scale LinkedIn ads or cut them, you need paid CAC. Confusing the two produces confident decisions with bad inputs.

Fully loaded CAC depends on both the spend mix (numerator) and new customers (denominator); you need a consistent definition before you compare trends.

How to calculate CAC correctly

The core formula is straightforward:

$$\text{CAC} = \frac{\text{Sales and marketing cost in period}}{\text{New paying customers acquired in period}}$$

The "gotchas" are almost always about timing and counting.

Step 1: Define the denominator precisely

Most SaaS teams should use:

• New paying customers (new logos) created in the period
  Not trials started, not leads, not opportunities.

Be explicit about edge cases:

• Free to paid conversions: count them as new customers, but make sure the costs that created them are included.
• Reactivations: generally exclude from "new customers" and track separately (they're not new acquisition).
• Expansion-only closes: not new customers; they affect retention metrics like NRR (Net Revenue Retention), not CAC.

If you're sales-led, your denominator is often "new customers closed-won." If you're product-led, it might be "new self-serve paying subscribers."

Step 2: Decide the time window (and handle lag)

A common mistake is calculating CAC monthly when your sales cycle is 45–120 days. That makes CAC look volatile and misleading because costs happen before closes.

Two practical approaches:

• Lagged CAC: attribute costs to customers acquired 1–3 months later (choose a lag that matches your typical sales cycle).
• Cohort CAC: attribute the full cost of a cohort's acquisition program to the cohort of customers that resulted.

If you don't correct for lag, you'll mistakenly conclude "CAC spiked" when in reality closes simply slipped into next month.

When the sales cycle is longer than your reporting period, same-month CAC can swing for timing reasons rather than true efficiency changes.

Step 3: Don't mix "cash" and "P&L" views

CAC is usually a P&L metric (expense incurred), not a cash metric. But founders often care about both:

• P&L CAC helps you understand unit economics.
• Cash timing matters for runway and payback.

If you're paying annual contracts upfront for tools or events, consider whether you amortize those costs for CAC. The key is consistency month to month.

Once you have CAC, see how fast it pays back with the free CAC payback calculator.

What drives CAC up or down

CAC changes because either costs changed, or conversion/volume changed. In practice, it's usually both.

1) Funnel conversion and sales execution

CAC improves when conversion improves at any stage:

• Higher lead-to-trial or lead-to-demo conversion (see Conversion Rate)
• Higher SQL rate (see Qualified Pipeline)
• Higher close rate (see Win Rate)
• Shorter cycle (see Sales Cycle Length)

Even if spend is flat, CAC rises when:

• win rate drops
• sales cycle lengthens (fewer closes per month)
• lead quality declines

Practical diagnostic: if CAC jumped, check whether new customers fell and then trace backward: closed-won → pipeline created → qualified leads → lead volume.

2) Channel saturation and competition

Paid channels often get more expensive as you scale:

• you exhaust the easiest audiences
• competitors bid up the same keywords
• marginal placements convert worse

This shows up as a higher cost per lead and lower lead-to-customer conversion, which is double damage to CAC (see cost per lead and ROAS below).

3) Pricing and packaging changes

CAC is "cost per customer," not "cost per dollar of revenue," so pricing changes don't directly change CAC. But they can change the quality of customers and the conversion rate:

• Raising price can reduce conversion and increase CAC (fewer customers for same spend)
• Better packaging can increase conversion and reduce CAC
• Heavy discounting can preserve conversion while damaging payback (more on that below)

If you're changing pricing, monitor ASP (Average Selling Price) and ARPA (Average Revenue Per Account) alongside CAC to avoid false conclusions.

4) Brand, word-of-mouth, and product-led loops

As your product and reputation improve, you often see:

• more direct traffic
• higher conversion from the same traffic
• lower CAC even without spending less

This is why CAC should be interpreted together with retention and satisfaction signals like NPS and CSAT (see NPS (Net Promoter Score)). Strong retention and advocacy reduce future acquisition pressure.

The Founder's perspective: CAC rarely "improves" because you got better at arithmetic. It improves because you built a stronger acquisition engine: tighter ICP, better positioning, higher conversion, faster sales cycles, or more durable word-of-mouth.

Cost per lead and ROAS

CAC sits at the bottom of the funnel, so it reacts slowly. Two upstream metrics give you faster feedback on paid acquisition: cost per lead (CPL) and return on ad spend (ROAS). Both are only useful once you translate them back into CAC.

Cost per lead (CPL)

Cost per lead is what you spend to generate one lead in a channel, campaign or period. Ad platforms define it the same way; LinkedIn's Campaign Manager, for example, reports it as the amount spent per lead collected:

$$\text{CPL} = \frac{\text{Lead acquisition cost}}{\text{Number of leads}}$$

Include the costs required to create the lead (media spend, agency or contractor fees, channel-specific creative) and leave out sales payroll unless you're deliberately measuring cost per SQL. Only compare CPL across channels when "lead" means the same thing: a demo request, a trial signup and an ebook download are very different leads.

CPL turns into a paid CAC for that channel once you divide by your lead-to-customer rate:

$$\text{Paid CAC} = \frac{\text{CPL}}{\text{Lead-to-customer rate}}$$

A $50 CPL with a 5% lead-to-customer rate implies $1,000 of paid CAC per customer, before any sales salaries. Turned around, the same relationship gives each channel a spending ceiling:

$$\text{Max CPL} = \text{Target CAC} \times \text{Lead-to-customer rate}$$

With a $6,000 target CAC and a 2% lead-to-customer rate, you can pay up to $120 per lead. The classic trap is optimizing CPL on its own: shorter forms, gated content and broader targeting all lower CPL, but if conversion falls faster than CPL does, CAC goes up. Rank channels by the CAC they imply, not by CPL.

ROAS (return on ad spend)

ROAS is the revenue you attribute to ads divided by what you spent on those ads. Google Ads reports it as conversion value divided by cost:

$$\text{ROAS} = \frac{\text{Attributed revenue}}{\text{Ad spend}}$$

In SaaS, ROAS is slippery because subscription revenue arrives month by month and churn takes some of it back. A 30-day ROAS on monthly plans will look poor even for a good channel, while annual prepay pulls a year of revenue into the first month. Three rules keep it honest:

• Measure it on cohorts. Tie a month's spend to the customers it acquired and follow their revenue at 30, 90 and 180 days. That's the version that catches channels which buy customers who churn quickly.
• Use gross profit, not revenue. Multiply attributed revenue by your gross margin so a high-COGS product doesn't look better than it is.
• Treat platform-reported ROAS as a tuning signal. Each ad platform attributes conversions with its own rules and lookback windows, so use its ROAS to compare creative and audiences inside that platform, and make budget decisions on revenue from your billing data.

ROAS and CAC payback are the same question asked from opposite ends. Measured on gross profit, a cohort's ROAS reaches 1.0 in the month it has paid back its paid CAC:

$$\text{Margin ROAS after N months} = \frac{\text{Gross profit from the cohort in its first N months}}{\text{Ad spend that acquired the cohort}}$$

For example, with $200 ARPA, an 80% gross margin and $960 of paid CAC, each retained customer earns back its acquisition cost in six months ($200 × 0.8 × 6 = $960), so six-month margin ROAS is 1.0 if nobody churns, and lower if some do. If margin ROAS needs longer than your CAC payback target to reach 1.0, the channel is too expensive at its current scale.

Putting the chain together: CPL ÷ lead-to-customer rate gives paid CAC per channel; adding sales and marketing salaries, commissions and tools gives fully loaded CAC; ROAS checks the same spend from the revenue side.

How to tell whether CAC is healthy

CAC has no universal benchmark. A "good" CAC depends on (1) how much gross profit a customer generates per month, and (2) how long they stay and expand.

The three lenses founders should use

1) Payback

Payback answers: how quickly do we earn back CAC from gross profit?
This is usually the first "health" check because it's directly tied to cash efficiency.

• If payback is getting worse, growth is becoming more expensive to finance.
• If payback is improving, you can often scale faster with less risk.

Link this explicitly with CAC Payback Period.

2) LTV relative to CAC

A common unit-economics ratio is:

$$\text{LTV to CAC ratio} = \frac{\text{Customer lifetime value}}{\text{Customer acquisition cost}}$$

Use LTV (Customer Lifetime Value) and LTV:CAC Ratio as the framework, but don't stop at a single number. Two businesses can have the same LTV:CAC while having very different cash profiles depending on payback speed.

3) Margin and retention reality

CAC is only as "safe" as the retention and margin behind it:

• Falling Gross Margin reduces how much profit is available to pay back CAC.
• Rising Logo Churn reduces realized lifetime and makes CAC riskier.
• Improving NRR (Net Revenue Retention) can justify higher CAC because expansion helps repay acquisition cost.

Practical benchmark ranges (use with caution)

Bessemer Venture Partners publishes CAC payback targets by segment, measured against gross margin-adjusted ARR. Use them for orientation, not as goals:

Bessemer's State of the Cloud 2023 adds a single good / better / best scale of 12–18 / 6–12 / 0–6 months. Actual results are often slower, especially at higher price points: Benchmarkit's 2026 survey of 342 B2B SaaS and AI companies found a median CAC payback of 16 months, from 11 months for ACVs under $5k to 22 months for $50k–100k. David Skok, who originally proposed recovering CAC in under 12 months, has since written that around 20 months is now common for healthy SaaS businesses, and that above 24 months you should work to bring it down.

If you're early-stage, stable payback matters more than "perfect" payback. Volatile CAC plus long payback is what creates unpleasant surprises.

How founders use CAC in real decisions

CAC becomes powerful when it informs a specific decision: where to allocate headcount and dollars.

1) Setting growth budgets and hiring plans

If you know your target CAC and payback, you can set a sane acquisition budget:

• If CAC is stable and payback is within target, scaling spend is less risky.
• If CAC is rising and payback is stretching, adding spend often makes the problem worse.

Tie this to cash discipline metrics like Burn Rate and Burn Multiple. A company can "grow" while becoming less capital efficient if CAC is drifting up.

The Founder's perspective: The question isn't "Can we buy growth?" It's "Can we buy growth at a CAC that keeps payback inside our runway?" CAC is a runway management metric as much as a marketing metric.

2) Comparing channels

Channel comparisons fail when:

• costs are tracked by month, but closes happen later
• channels influence each other (content supports paid, brand supports outbound)
• you attribute based on last touch only

A workable founder approach:

• Use paid CAC to optimize channels weekly.
• Use fully loaded CAC to validate that the overall machine is healthy monthly/quarterly.
• If a channel looks "too good," sanity-check: does it still look good when you include the people and tooling required to run it?

3) Pricing and discount policy

CAC interacts with discounting indirectly –  but brutally –  through payback.

If you discount heavily:

• CAC per customer may stay the same
• gross profit per customer falls
• payback stretches and the business becomes harder to finance

That's why discounting should be evaluated with Discounts in SaaS and with ARPA/ASP trends. If you "save" conversion rate with discounts, you may be quietly trading away unit economics.

4) Segment strategy (SMB vs mid-market)

Blended CAC hides the truth. Segment CAC reveals where you actually have leverage.

Two segments can look like this:

When founders say "CAC is fine," the follow-up is: fine for which segment? Segmenting by plan, industry, ACV, or channel often changes the strategy.

Segmenting CAC into payback outcomes prevents blended averages from hiding which customers you can scale profitably.

When CAC "breaks" (and what to do)

CAC isn't just a performance score –  it's an early warning system. Here are common "break" patterns and the founder actions they call for.

Pattern 1: CAC rises while volume falls

What it usually means

• channel saturation
• messaging mismatch
• lead quality drop
• sales execution regression

What to do

• Diagnose conversion by stage (lead → MQL → SQL → close)
• Review ICP fit and disqualify faster
• Tighten positioning and sales enablement
• Consider reallocating to higher-intent channels

Related metrics to pull in:

• Qualified Pipeline (MQLs and SQLs)
• Win Rate

Pattern 2: CAC looks stable, but payback gets worse

What it usually means

• price realization dropped (more discounting)
• gross margin fell (higher COGS or support burden)
• churn increased (customers don't stay long enough)

What to do

• Audit discounting and packaging
• Review COGS (Cost of Goods Sold) and Gross Margin
• Look at retention trends with Retention and Cohort Analysis
• Investigate churn drivers with Churn Reason Analysis

Pattern 3: CAC improves "too fast"

What it usually means

• you cut acquisition spend and starved the top of funnel
• you benefited from a short-term channel anomaly
• attribution shifted rather than real efficiency

What to do

• Verify pipeline creation didn't collapse
• Check lagged CAC and next-month closes
• Confirm retention quality didn't deteriorate (cheap customers who churn)

Common mistakes and edge cases

Mistake 1: Counting expansions as "new customer revenue"

CAC is about acquiring new logos. Expansion belongs in retention metrics like Expansion MRR and Net MRR Churn Rate.

Mistake 2: Using signups or trials as the denominator

That turns CAC into something closer to cost per signup, which is not what you need for unit economics. Use leads/trials for funnel optimization, but keep CAC tied to new paying customers.

Mistake 3: Ignoring refunds and chargebacks

Refunds and chargebacks don't change CAC directly, but they can destroy payback and distort "effective acquisition." If refunds are material, review Refunds in SaaS and Chargebacks in SaaS.

Mistake 4: Comparing CAC across months with different definitions

If you changed what's included (tools, payroll allocation, commissions), you changed CAC. Document the change and consider recalculating historical CAC to keep trends comparable.

A simple CAC operating cadence for founders

If you want CAC to drive action (not debates), run it on a cadence:

• Weekly (channel view): paid CAC proxy metrics (lead volume, CPL, conversion rates)
• Monthly (management view): blended paid CAC and fully loaded CAC, with lag adjustment
• Quarterly (strategy view): CAC by segment + payback + retention quality

Pair CAC with:

• ARPA (Average Revenue Per Account) to understand revenue per customer
• Logo Churn and NRR (Net Revenue Retention) to validate lifetime value
• Burn Multiple to ensure growth remains capital-efficient

The goal isn't to force CAC down at all costs. The goal is to build an acquisition engine where CAC, payback, and retention align –  so you can scale without financing every dollar of growth twice.