What's the right CAC payback target — 12, 18, 24 months in 2027?

The three anchors compared

The three numbers people argue about are not arbitrary. Each corresponds to a recognizable financial profile, and choosing between them is really choosing which profile you are — or intend to become.

Twelve months is the efficiency anchor. At a 12-month payback, one year after landing a customer you have recovered the entire fully-loaded cost of acquiring them, and every gross-margin dollar after that funds the next cohort or drops to the bottom line. The practical consequence is that growth becomes substantially self-funding. Take a cohort that cost $1M in fully-loaded acquisition spend and returns $1M of gross-margin-adjusted revenue over the following twelve months. By month 13 the original outlay is back in the bank. If 90% of that cohort renews into year two — and at businesses running this profile gross retention is usually higher — the cohort throws off close to another $1M, which can finance the next cohort outright. External capital becomes an accelerant rather than life support. That is why private equity and growth-equity firms gravitate to this number: it implies the business can compound without an open-ended equity requirement, and it minimizes the bridge financing the growth plan depends on.

Eighteen months is the venture-scale default. An 18-month payback says you are deliberately spending ahead of pure efficiency to capture a market faster than self-funding would allow, inside a band that Series B and C investors will underwrite without argument. The underwriting logic is specific: an investor wants evidence that the incremental sales-and-marketing dollar buys durable ARR rather than renting temporary revenue. An 18-month payback paired with gross retention around 88-92% and net retention above 105-110% demonstrates exactly that — the customer is still there long after the recoupment window closes, and the account base grows. The implied LTV-to-CAC lands in the healthy 3x-5x range, which clears most return thresholds with room. Eighteen months also gives the go-to-market organization room to build motions with longer horizons: outbound enterprise, brand, partnerships, category creation. A company squeezed to a strict 12-month bar at Series B usually has to starve exactly those motions.

Twenty-four months is the outer edge. It is the boundary of defensibility, not a default. It works when a narrow set of conditions all hold at once: annual contract values in the high five figures or above, contracts that run two to three years, gross revenue retention in the 92-96% band, net revenue retention comfortably above 115-120%, and gross margins in the high 70s or better. Under those conditions the customer relationship runs six, eight, ten years and grows every year, so recouping acquisition cost in twenty-four months still leaves the overwhelming majority of a long, expanding, high-margin lifetime as pure value creation. Outside those conditions, twenty-four months is where businesses quietly die.

The single most important idea in this whole debate is that the same number means opposite things depending on the retention underneath it. A vertical SaaS company selling three-year contracts to hospital systems at 92% gross margin and 96% gross retention can run a 22-month payback and compound beautifully. An SMB project-management tool selling month-to-month at 78% gross margin and 80% gross retention will go bankrupt on the identical 22 months, because a large share of each cohort is gone before month 24 ever arrives. Anyone who hands you a payback target without first asking about your gross retention is giving you advice that is, at best, a coin flip.

There is a capital-environment layer on top of the profiles. During the zero-rate years that ran roughly through 2021, a 24-to-30-month payback was routinely waved through in Series B and C diligence, because the capital to bridge the gap was nearly free and growth rate was the only thing being priced. When rates repriced risk starting in 2022, public SaaS multiples compressed hard, the exit window narrowed for inefficient companies, and the efficiency bar moved up the funnel — private investors now demand at Series B and C roughly what the public market demands, because the public market is the exit they are underwriting. The operating consequence for a median company between $5M and $200M ARR is that a payback target set in a cheap-capital environment becomes a liability in an expensive one. Set your target for the environment you will actually be raising into.

How to decide between 12, 18, and 24

The decision is a function of five inputs. Run them in order; each one either buys you room toward the long end or forces you toward the short end.

Input one: gross revenue retention. This is the master variable, because it sets the ceiling on how long a payback you can survive. The viability condition is blunt: payback must finish comfortably before the period over which the cohort actually pays. GRR above 92% earns room toward 24 months. GRR between 85% and 92% supports a 12-to-18-month target. GRR below 85% forces 12 months or less, and GRR under 80% means SMB-grade discipline is non-negotiable — six to twelve months, no exceptions.

Input two: gross margin. This scales the whole calculation, and skipping it is the most common self-deception in the metric. If you carry 80%+ gross margins, your adjusted payback sits close to your headline number and you have full latitude. If you run 60-70% — common for AI-native products carrying real inference costs, or for services-heavy SaaS — your true payback is materially longer than the raw number, so you must target a *shorter* raw payback to land at a healthy real one. Be honest that your structural ceiling is lower than a high-margin peer's.

Input three: net revenue retention. NRR shortens the effective payback because the customer's spend grows during the recoupment window. Above 120%, expansion is materially funding your own recovery and you can tolerate a longer gross payback. Around 100%, new-logo economics have to stand entirely on their own. Below 100%, gross churn is actively working against you and the target must tighten.

Input four: growth rate. This is the Rule of 40 interaction. A long payback depresses near-term margin — you are spending cash now for customers who repay later — but if the spend is productive it buys growth. Growing 60% a year at a 24-month payback can still clear 40 even at a deeply negative margin, and the long payback is a rational land-grab bet. Growing 15% a year at the same 24 months cannot clear 40; the payback is not buying enough growth to justify itself, and you have an efficiency problem you can no longer outrun. This is why "is 24 months acceptable?" is genuinely unanswerable without the growth rate attached.

Input five: runway and the raise you are walking into. Short runway, or a near-term round into a disciplined market, forces a shorter target — you cannot run a payback that depends on bridge capital you may not get on acceptable terms. A long runway or a strong balance sheet buys room to run a deliberately longer payback for market capture.

The output of that walk is not a bare number. It is a number plus the sentence that defends it: *"Our target is a 15-month fully-loaded, gross-margin-adjusted new-business CAC payback, because we hold 90% GRR, 80% gross margin, 116% NRR, and are growing 45% — and that target drops to 12 months if GRR falls below 87%."* That conditional framing is the deliverable. A target without the rationale is a guess. A target with it is a management decision you can defend to any board, and a RevOps team can operationalize the trigger condition directly into the reporting cadence.

There is a mirror-image discipline that belongs in the decision: never report a payback figure without gross retention next to it. A board that sees "CAC payback: 16 months" has learned almost nothing. A board that sees "CAC payback: 16 months | GRR: 91% | NRR: 118%" can actually evaluate the business, because they can see the recoupment window sitting inside a durable, expanding relationship. The pairing is not context around the metric — it *is* the metric.

The concrete numbers behind each option

Targets only mean something once the calculation underneath them is honest. Two arithmetic disciplines separate a real payback number from a flattering cousin of it.

The gross-margin adjustment. The correct denominator is not new MRR — it is new MRR multiplied by gross margin, because the only revenue available to repay acquisition cost is what survives the cost of delivering the service. Hosting, support, customer success headcount, third-party data and API costs, payment processing: all consumed before a dollar is free to pay back CAC. The distortion is not small. Take $10,000 of fully-loaded CAC per customer against $1,000 of new MRR. Against raw MRR the payback reads 10 months — comfortably inside any target. Apply a 75% gross margin and the adjusted MRR is $750, so the true payback is 13.3 months. Apply 65%, common where infrastructure or services load the cost of revenue, and it is 15.4 months. The "10-month business" was a 15-month business the whole time, and a team reporting the raw number to its board was operating with a 50% error baked into its most important efficiency metric.

Fully-loaded CAC. Every excluded cost makes payback look shorter than it is, and exclusions compound. Fully loaded means: base salaries for account executives, SDRs and any sales-assist roles; commissions, accelerators and SPIFs recognized to the right period; all marketing program spend — paid acquisition, events, content, webinars, ABM; marketing team salaries across demand gen, product marketing, content and marketing ops; the fully-loaded cost of the SDR and BDR layer, which sits "before" the AE and is the classic omission; the entire sales and marketing tool stack — CRM, sales engagement, intent data, marketing automation, attribution, call recording; sales and marketing leadership plus the RevOps headcount supporting acquisition; and a reasonable allocation of shared facilities, IT and recruiting attributable to the GTM org. A team counting only program spend and AE commissions might report a 9-month payback on a business whose fully-loaded number is 17 months, then build comp plans and board expectations on the fiction.

Two guardrails keep it honest. First, reconcile to the P&L: the sum of everything you call CAC should tie, within allocation tolerance, to the actual sales and marketing line over the same period. If your CAC implies spend far below the income statement, you are excluding real costs. Second, document the definition and freeze it — write down what is in and out, get finance and GTM leadership to agree, and do not quietly redefine it between board meetings.

Segment baselines. Because ACV, retention, and cost-to-acquire all rise together as you move up-market, the target should differ by segment:

• SMB: 6-12 months. Low ACVs, short cycles, and the weakest retention — GRR commonly 75-85%. The cohort decays fast, so recoup fast. SMB motions that cannot get under 12 months usually have a pricing problem (ACV too low for the cost to acquire and serve), a channel problem (running a sales motion where a self-serve motion is required), or both.
• Mid-market: 12-18 months. Higher ACVs, stickier usage, GRR often 85-92%. A more involved motion with real AEs and sales engineering pushes cost-to-acquire up, but retention and deal size absorb it. Twelve to fifteen months is genuinely strong; fifteen to eighteen is fine when NRR is doing real work.
• Enterprise: 18-24 months. Highest ACVs, longest and most expensive cycles — multi-stakeholder, procurement, security review, sometimes a year-plus — and the best retention, with GRR frequently 92-96% and NRR well above 115%.

Channel baselines, which are often more actionable because mix is something a GTM leader can shift quarter to quarter:

• PLG / self-serve: under 6 months, often under three. No AE comp on the deal, minimal human touch. A PLG motion that cannot clear this is a sales-assisted motion wearing PLG pricing — an unprofitable combination.
• Inbound, marketing-sourced and sales-closed: 9-15 months. Warm leads, so no prospecting cost, but a closing AE. This is usually the efficiency backbone of a mid-market business.
• Outbound, SDR-sourced: 15-24 months. The full cost stack with no warm-lead discount. That is not a flaw — outbound reaches buyers who will never raise their hand — but it must be run as a deliberate long-payback investment and scrutinized past 24 months.
• Partner / channel: variable. A mature ecosystem delivering pre-qualified deals can pay back very fast, because you pay margin share instead of a full GTM stack. An immature program needing enablement and co-marketing looks terrible for two years and then inflects. Evaluate it on its maturity curve.

Three different paybacks, and the one that hides problems. *Blended* payback divides total S&M by total new gross-margin revenue across everything. It is the easiest to compute and the least useful, because it averages businesses with opposite economics — a three-month self-serve motion can subsidize a broken 30-month outbound motion, or a healthy enterprise segment can be dragged down by a structurally unprofitable SMB one. *New-business* payback isolates new logos from expansion and renewal, which matters because a CSM driving an upsell costs a fraction of an AE landing a new account; folding expansion in makes new-logo acquisition look more efficient than it is. *Segment* payback breaks the number by the dimensions with genuinely different economics. Report all three, set targets at segment level, and never let the blend be the only number anyone looks at. A 16-month blend is fine if it is 60% mid-market at 14 and 40% enterprise at 20. The same 16-month blend is alarming if it is 50% SMB at 22 propped up by 50% enterprise at 10 — the SMB half is broken and the enterprise half is suspiciously good.

Two adjustments worth computing, carefully. Multi-year prepaid contracts create a large gap between revenue-recognized payback and cash payback. An enterprise customer signing $300K across three years and prepaying against $180K of CAC is cash-positive on day one, even though the recognized-revenue calculation reads 18-22 months. That cash-timing benefit is real and belongs on the dashboard next to prepaid mix — but it does not change whether the customer is economically profitable, which still depends on retention and lifetime value against total CAC. Report both views: the recognized one tells you about durability, the cash one about the financing burden of growth. Similarly, *expansion-adjusted* or net payback accounts for cohort revenue growing at the NRR rate during the window; at 130% NRR the cumulative revenue crosses the CAC line meaningfully sooner than the flat-revenue calculation implies. That view is legitimate when expansion is durable, broad-based and proven across many cohorts — and misleading when 105% NRR is masking high gross churn offset by aggressive upsell to survivors. Keep the conservative gross-margin-adjusted number as primary and label the net view clearly as secondary.

One cross-check. The SaaS Magic Number — net new ARR in a period divided by prior-period S&M spend — is the same efficiency measured in different units. Roughly 0.75 and up is efficient; above 1.0 is strong; below 0.5 means the engine is struggling to convert spend into growth. Because the two metrics are mathematically linked through gross margin, they should tell a consistent story. When they diverge, investigate rather than explain away: a great payback alongside a weak Magic Number usually means your CAC definition is too narrow, since the Magic Number is anchored to the actual P&L. A strong Magic Number alongside a weak new-business payback usually means expansion is carrying the company and new-logo acquisition is the real problem.

Implementing the target and sequencing the work

Choosing the number is the easy half. Making it a live operating constraint takes a defined sequence, and skipping steps is how companies end up managing to a metric they have not actually instrumented.

Step one — freeze the definitions, in writing. Before any target is set, finance and GTM leadership agree on exactly what is in CAC, which gross-margin figure is used (fully burdened with COGS, including customer success if you treat CS as a delivery cost — not a flattering "platform margin"), and which payback view is primary. Reconcile the CAC total to the sales and marketing line in the financials. Publish the definition. This step exists because a target built on an unstable numerator is unfalsifiable.

Step two — instrument by segment and channel before setting targets. Build the reporting so payback can be cut by SMB / mid-market / enterprise and by PLG / inbound / outbound / partner, with new-business isolated from expansion. Until that exists, you have one blended number that is structurally incapable of telling you where a problem lives. Early-stage companies often cannot compute anything meaningful yet — at seed the sample is tiny and CAC is dominated by founder-led selling that does not scale — and the correct posture there is to instrument honestly without managing to a hard number, since premature optimization starves the experimentation that finds the motion.

Step three — derive the target per segment using the five-input walk, then write the defending sentence and the trigger condition for each one. Different segments get different targets; a single company-wide target is a fiction.

Step four — set the stage-appropriate gradient. Series A can carry 18-24 months while the motion is still finding channel-market fit and reps are ramping; what matters there is that the number trends the right way and the unit economics close. Series B is where the 18-month default becomes a real operating bar, because the motion is supposed to be proven and now scaling. Series C and beyond, and anything a growth-equity or PE firm underwrites, should be at 12-15 months or better, since capital at that stage is supposed to be accelerant rather than life support. Trajectory matters as much as level: a board wants to hear that payback was 22 a year ago, is 16 now, and is heading to 13.

Step five — wire the target into the operating cadence. Payback belongs in the monthly business review alongside GRR and NRR, in the board deck as a pair rather than a bare number, and in the quarterly planning conversation as a constraint on channel mix. When the trigger condition fires — GRR slips below the threshold you named — the target tightens automatically rather than being renegotiated.

The levers, in order of speed. When payback needs to come down, the fastest lever is usually channel mix — leaning harder into PLG and inbound and tightening the outbound bar moves the blended number without cutting total spend. Next is pricing and packaging: raising ACV improves payback directly, and moving to annual-prepaid billing improves the cash view immediately. Then retention, which does not change the payback arithmetic but changes what payback you can afford — every point of GRR buys headroom. Then sales efficiency: ramp time, win rates, and rep productivity all move CAC. Cutting spend is the crudest lever and usually the one with the worst second-order effects.

The failure mode to design against. Having built the case for discipline, the opposite error deserves equal weight, because it is common and in some ways more damaging: optimizing payback too hard permanently caps a company's ceiling. The fastest way to improve the number is to stop doing everything with a long payback — cut outbound, cut brand and category investment, cut partnerships, cut the enterprise motion, and lean entirely on the cheapest channels. The dashboard improves next quarter. But the motions you cut were the ones reaching the largest accounts, building the brand that was quietly generating your inbound, and capturing the parts of the addressable market a competitor will now take. The recognizable pathologies: starving outbound and discovering two years later that a rival owns upmarket; killing thought-leadership spend because it has the longest and least-attributable payback; refusing a new product line, geography or segment because the initial payback looks ugly, when every such investment looks ugly at the start; and re-rigging sales comp to reward only fast easy deals, so reps rationally stop chasing the big slow valuable ones.

Hold both ideas at once. Payback is a guardrail, not the optimization target — a range you stay inside so the business is financeable and the unit economics close, while growth, market capture and terminal value are what you actually optimize *within* it. "We will not let blended payback exceed 18 months, and inside that constraint we will invest as aggressively as we can to capture the market" gets the relationship right. "We will drive payback as low as it will go" confuses the seatbelt for the steering wheel, and arrives very efficiently at a smaller destination.

Related questions

Should we use blended or new-business CAC payback for the board?

Report both. Blended is the headline summary; new-business payback is what actually drives growth investment decisions, because it isolates the cost of landing a logo you do not have from cheap expansion revenue that averages the number down and flatters new-logo efficiency.

Does a 24-month payback ever make sense for SMB?

Practically never. With SMB gross retention around 75-82%, a large share of each cohort churns before month 24, so the cohort never repays in cash. The faster you grow at that profile, the faster you burn. Target 6-12 months.

How does payback relate to LTV-to-CAC?

They measure the same economics on different horizons. Payback asks how fast you recoup; LTV-to-CAC asks how much total value the customer returns. A short payback with weak retention still yields a poor ratio, so a healthy business needs both, not either alone.

What payback should an AI-native product with 60% gross margins target?

A shorter raw one. Since the gross-margin adjustment is larger, a raw 12-month number is really 20 months. Target a raw payback tight enough that the adjusted figure lands where your retention supports — and accept a structurally lower ceiling than an 85%-margin peer.

Should prepaid multi-year contracts change our target?

They change the cash view, not the target. A prepaid enterprise deal can be cash-positive on day one while the recognized payback reads 20 months. Track prepaid mix alongside payback, but set the target on recognized, gross-margin-adjusted economics.

FAQ

What exactly goes into fully-loaded CAC?

Sales base salaries and commissions including accelerators and SPIFs, all marketing program spend, marketing team salaries, the fully-loaded SDR and BDR cost, the entire sales and marketing tool stack, sales and marketing leadership plus supporting RevOps headcount, and a reasonable allocation of shared overhead attributable to the GTM organization. If your CAC total does not roughly reconcile to the sales and marketing line in your financials, you are excluding real costs and your payback is understated.

Why do I have to multiply by gross margin?

Because the only revenue that can repay acquisition cost is what survives delivery. Hosting, support, customer success, third-party APIs and payment processing are consumed first. At 75% gross margin a raw 10-month payback is really 13.3 months; at 65% it is 15.4. Teams that skip the adjustment are not measuring CAC payback — they are measuring a more optimistic cousin of it under the wrong name.

Should the target be the same across all segments?

No. A single company-wide target is a fiction, because ACV, retention, and cost-to-acquire all differ by segment. SMB should sit at 6-12 months, mid-market at 12-18, enterprise at 18-24. Use the blend as a board headline and the segment targets as the actual management tool. Segment mix should explain the blend, never hide inside it.

How does the Rule of 40 change the answer?

It makes the growth-efficiency trade-off explicit. A 24-month payback at 60% growth can still clear 40 even at a deeply negative margin, so the long payback is arguably a rational land grab. The same 24 months at 15% growth cannot clear 40 — the payback is not buying enough growth to justify itself. Ask not "is my payback good" but "is my payback good given how fast I am growing."

What if my Magic Number and my payback disagree?

Treat the divergence as a signal to reconcile inputs, not something to explain away. A strong payback alongside a weak Magic Number usually means your CAC definition is too narrow, since the Magic Number is anchored to actual P&L spend. A strong Magic Number alongside a weak new-business payback usually means expansion is carrying the company while new-logo acquisition is the real problem.

Can we report expansion-adjusted payback instead?

Only as a clearly labeled secondary metric, and only if expansion is durable, broad-based and proven across many cohorts. At 130% net retention the cohort genuinely crosses the CAC line sooner than a flat-revenue calculation implies. But 105% NRR sitting on 80% gross retention is a fragile expansion story, and leaning on the net view there hides the gross-churn problem rather than illuminating it.

Sources

• https://www.bvp.com/atlas/state-of-the-cloud
• https://openviewpartners.com/blog/saas-metrics/
• https://a16z.com/16-startup-metrics/
• https://www.forentrepreneurs.com/saas-metrics-2/
• https://www.klipfolio.com/resources/kpi-examples/saas/cac-payback-period
• https://www.mckinsey.com/industries/technology-media-and-telecommunications/our-insights/the-rule-of-40-the-key-to-long-term-saas-success
• https://www.saastr.com/
• https://tomtunguz.com/
• https://www.kbcm.com/technology-group/saas-survey
• https://www.profitwell.com/recur/all/cac-payback-period

Related on PULSE

• How do you calculate fully-loaded customer acquisition cost?
• What's a healthy LTV-to-CAC ratio for SaaS?
• Gross revenue retention vs net revenue retention — which matters more?
• What is the SaaS Magic Number and how do you use it?
• How does the Rule of 40 apply at different growth stages?
• How should RevOps report unit economics to the board?