Retention cohort LTV

The LTV number on your pitch deck is probably wrong

Every SaaS company I've ever audited has overstated LTV by 30–80%. The usual culprits: using month-one retention multiplied by 24 months (ignores the steep curve), assuming current churn stays flat (churn usually rises in later cohorts as the easy wins saturate), and not discounting for gross margin. The formal LTV definition most finance teams want — gross-margin-weighted cumulative revenue per customer, summed across their actual retention curve, not a theoretical one — is almost never what marketing teams put in the board deck.

This calculator takes a cohort-based approach: you enter monthly retention rates for months 1–12 (or longer), your average revenue per user per month, and your gross margin. It returns the real LTV based on the actual shape of your retention curve, and separates "year-1 LTV" (the cash you can plan with) from "steady-state LTV" (the north star).

Benchmarks: retention curves that work (2026)

The shape of retention matters more than the average

Two companies can have the same "20% month-12 retention" and radically different LTVs. Company A drops 40% in month 1 (trial-to-paid, early churn) then decays slowly — month 3 retention 45%, month 6 retention 32%, month 12 retention 20%. Company B retains 70% in month 1 but decays steeply — month 3 retention 50%, month 6 retention 30%, month 12 retention 20%. Both hit 20% at month 12, but Company A's cumulative LTV is often 20–35% higher because the later months contribute more cumulative revenue.

Plotting your retention curve (by monthly cohort, overlaid) is the first analytics exercise I make every new client do. In Mixpanel, Amplitude, or a spreadsheet — but not in the aggregated churn number their Stripe dashboard shows. That number hides the cohort dynamics.

Early-LTV vs. steady-state-LTV: the planning distinction

Early-LTV (month-3 or month-6 LTV) is what you can plan against. It's the cash the customer has actually produced. If you're a bootstrapped DTC with 60-day payback, early-LTV tells you whether the campaign ran a profit.

Projected-LTV (12-month or 24-month) is a forecast based on the retention curve. It's the pitch-deck number. VCs care about it. Operators should be skeptical of it.

Steady-state-LTV (asymptotic) is the infinite-horizon sum of revenue from a customer. Only meaningful for categories with true long-term retention like health insurance, utilities, or enterprise SaaS with net-expansion.

Use early-LTV for cash planning. Use projected-LTV for CAC targets and strategic decisions. Never use steady-state-LTV except for high-retention categories where you have genuine 5+ year data.

The math mistake that kills most LTV models

Using month-1 retention (say, 85%) raised to a power (85% ^ 12 = 14% month-12 retention) assumes constant churn — exponential decay. Real retention curves never look like exponential decay. They look like power curves: steep initial drop, flattening over time. Using exponential decay systematically underestimates long-term retention and underprices LTV for customers who have already made it past the early-churn phase.

The fix: fit an actual curve to your cohort data. Power law (C × t^−a) works well for consumer SaaS and DTC. For B2B, use a segmented model — steep first-month "trial" churn separate from long-run enterprise retention. Stripe's Atlas guide has good templates; if you're on Mixpanel or Amplitude, cohort-retention reports are built in.

Monthly retention rates that actually imply good LTV

Gross margin: the LTV multiplier people forget

A $40/month SaaS subscription with 85% gross margin produces $34/month in gross-margin dollars. A $40/month DTC supplement subscription at 58% gross margin produces $23/month. Both have the same nominal monthly revenue but the SaaS customer is worth 47% more in LTV for the same retention curve. Always calculate LTV in gross-margin dollars, not top-line revenue, unless you're preparing a top-of-funnel growth slide.

For payback calculations specifically, use gross-margin LTV. For CAC:LTV ratios, use gross-margin LTV. For forecasting revenue, use top-line LTV. Keep both numbers visible; the mix matters.

Net revenue retention: the expansion lever

In B2B SaaS, the LTV conversation is increasingly about net revenue retention (NRR) rather than retention alone. NRR measures the dollar-weighted change in revenue from existing customers — expansions (upsells, seat additions, price increases) minus contractions minus churn.

Snowflake has reported 169% NRR at various quarters. HubSpot historically runs 105–112%. A company with 115% NRR doubles customer revenue roughly every 5 years through pure expansion. A company with 92% NRR loses 8% of cohort revenue annually even before accounting for logo churn. The difference between those two trajectories dwarfs any CAC efficiency question.

The retention levers that actually move the LTV number

1. Onboarding to 'activation' within 7 days. Mixpanel data shows activation within 7 days predicts 60%+ higher month-6 retention. Obsess over the activation metric more than acquisition.
2. Habit formation in weeks 1–4. B2C SaaS users who log in 3+ times in week 1 retain 2–3x better than those who log in once. Weekly digest emails, smart push notifications, progress reminders.
3. Pricing that rewards annual commitment. 15–25% annual discount in exchange for annual commit. Typically triples month-12 retention for SaaS.
4. Product velocity. Customers churn less when the product is visibly improving. Ship changelogs, monthly product-update emails, visible roadmap.
5. Customer success for enterprise. Dedicated CSM for over $30K ACV. Expansion pipeline depends on this.
6. Winback flows for consumer. Day-30, day-60, day-90 post-churn emails with targeted offers. Recover 8–15% of churned customers.

The analyst's move: plot revenue-retention by cohort, not just user-retention

Logo retention (are they still a customer?) and revenue retention (are they still paying us the same or more?) can diverge sharply. If logo retention is 80% but revenue retention is 65%, your existing customers are downgrading. If logo retention is 80% but revenue retention is 110%, your existing customers are expanding — which is dramatically better. Always report both numbers on a cohort dashboard.

Three retention archetypes — LTV math by company type

DTC CPG subscription (coffee / vitamins / pet food). CAC $38, first-purchase AOV $42, contribution margin 52%, month-1 retention 65%, month-6 28%, month-12 18%. Gross-margin LTV lands near $168 over 12 months, which yields 8-month cash payback and a 4.4:1 LTV:CAC ratio. Klaviyo at $150/mo (5k list) pays back inside 3 weeks at these volumes because the day-3 post-purchase flow alone recovers 12–18% of single-purchase buyers into recurring. Meta prospecting CPM at $19 for DTC apparel drives a roughly $38 CAC if CTR holds 1.5% and landing-page CVR stays above 3.1%.

Horizontal SMB SaaS ($45/seat, 8-seat average). CAC $420 blended across Google Search (CPC $6.50) and LinkedIn retargeting (CPC $10). Monthly ARPU $360, gross margin 82%, month-12 retention 62%, NRR 108%. Gross-margin LTV cycles to roughly $5,900 over 36 months, producing a 14-month CAC payback and 14:1 LTV:CAC on a hunter-farmer motion. HubSpot Pro at $800/mo covers nurture and scoring until MRR clears $60k, where Professional or Enterprise tier becomes cheaper per seat than stacking point tools.

B2B mid-market SaaS ($2,200 ACV per seat, 18 seats). Fully loaded CAC $2,400, ACV $39,600, gross margin 78%, logo retention 92%, NRR 118%. Gross-margin LTV crosses $155,000 over 5 years, 22-month CAC payback against a 65:1 LTV:CAC ratio. LinkedIn Sponsored Content CPC runs $8–$14 for director-and-above targeting, and a $2,000 copy investment that lifts CTR from 0.7% to 1.1% cuts cost-per-SQL by ~35%. Retention modeling here has to segment by ICP — SMB sub-segment churns 2.4x faster than enterprise sub-segment inside the same product.

Tooling cost vs. retention ROI

A $150/mo Klaviyo bill returns 35–70x on DTC subscription retention through winback and replenishment flows — my standard threshold: if Klaviyo revenue attribution (last-click, inside the tool) is under 18% of topline, the account is under-using the channel. HubSpot Professional at $800/mo earns its keep once lead volume crosses 600 MQLs/month and sales needs scored routing; below that, Starter at $20/mo plus Klaviyo is the cheaper and often better stack.