Customer Lifetime Value in SaaS: 3 Formulas, and Why the Simple One Understates by 40%

Customer Lifetime Value is one of the most cited and most miscomputed metrics in SaaS. The formula most founders use — ARPU multiplied by gross margin divided by churn rate — is directionally useful but structurally wrong for any business with meaningful expansion revenue, which is most SaaS businesses operating above $1M ARR.

The understatement is not minor. For a product with 15-20% annual net revenue retention above 100% — common for B2B SaaS with upsell motions — the simple formula understates true LTV by 40-80%. This error compounds: if you anchor your CAC budget to an understated LTV, you systematically under-invest in acquisition and leave growth on the table.

This article covers all three LTV formulas, the math behind the expansion-adjusted version, when to use each, and the segmentation requirements that make any LTV calculation operationally useful.

The Three LTV Formulas

Formula 1: Simple LTV

Simple LTV = ARPU × Gross Margin % / Monthly Churn Rate

This is the most widely used formula. It models LTV as the present discounted value of an infinite stream of monthly gross profit payments from a customer who churns at a fixed rate.

The math behind it: at a constant monthly churn rate of c, the average customer lifetime is 1/c months. Multiply by monthly gross profit (ARPU × GM%) and you get LTV.

Example:

• ARPU: $200/month
• Gross Margin: 75%
• Monthly churn rate: 2%
• Simple LTV = $200 × 0.75 / 0.02 = $7,500

This customer is worth $7,500 in lifetime gross profit, using this formula.

What it gets right: captures the churn effect accurately; easy to compute from standard metrics; enables quick LTV/CAC benchmarking.

What it gets wrong: assumes ARPU is constant over the customer's lifetime. It ignores expansion — upsell, cross-sell, and seat additions that increase ARPU over time. This assumption is wrong for virtually every B2B SaaS with a growth motion.

Formula 2: Cohort-Based LTV

Cohort LTV = NPV of actual observed revenue for a defined customer cohort over their measured lifetime

Cohort LTV is the empirical measurement rather than a formula projection. For every cohort of customers who started in a given month, you:

1. Track every dollar of revenue from every customer in the cohort, month by month
2. Apply gross margin to convert revenue to gross profit
3. Apply a discount rate (typically your WACC or a standard 10-15%) to get NPV
4. Divide by the number of customers who started in the cohort

Example: A January 2023 cohort of 50 customers. By January 2025, you have 24 months of actual revenue data. Total gross profit observed from the cohort over 24 months = $280,000. Customers still active at month 24 have a projected remaining LTV of an additional $90,000 NPV (based on their current revenue run rate and churn probability). Total cohort LTV = ($280,000 + $90,000) / 50 = $7,400 per customer.

What it gets right: captures actual expansion, actual churn, actual behavior. This is ground truth.

What it gets wrong (or requires): needs 24+ months of data to be meaningful; computationally intensive without proper cohort analysis infrastructure; projection of remaining LTV beyond observed period still requires assumptions.

Formula 3: Expansion-Adjusted LTV

Expansion-Adjusted LTV = ARPU × Gross Margin % / (Monthly Churn Rate - Monthly Net Expansion Rate)

This formula modifies the simple version by adjusting the denominator for expansion. Instead of dividing by gross churn, you divide by net churn — the effective revenue loss rate after accounting for expansion from surviving customers.

The math: if customers churn at 2% monthly but surviving customers expand at 1.5% monthly, the net revenue decay rate is only 0.5% per month. The effective customer lifetime (in revenue terms) is 1/0.005 = 200 months — six times longer than the simple formula implies.

Example (same customer as above):

• ARPU: $200/month
• Gross Margin: 75%
• Monthly churn rate: 2%
• Monthly net expansion rate: 1.5%
• Expansion-Adjusted LTV = $200 × 0.75 / (0.02 - 0.015) = $150 / 0.005 = $30,000

This is four times the simple LTV estimate of $7,500. The expansion component is that powerful.

What happens when expansion rate exceeds churn rate:

If your monthly expansion rate of 1.5% exceeds your monthly churn rate of 1.2%, the denominator becomes negative: 0.012 - 0.015 = -0.003. LTV is theoretically infinite — the cohort grows in value over time rather than decaying. This is the mathematical representation of net negative churn, and it means your existing customer base is a compounding asset, not a depreciating one.

This is not academic. Companies like Snowflake and Datadog have demonstrated that enterprise SaaS with strong usage-based expansion can generate cohorts that grow in revenue for years after the initial sale. Their "LTV" is not a finite number — it is an ongoing compounding return on customer acquisition investment.

Why Simple LTV Understates by 40%+

The gap between simple LTV and expansion-adjusted LTV is a function of your Net Revenue Retention rate.

At 110% NRR — respectable but not exceptional for B2B SaaS — the simple formula understates LTV by 40%. At 120% NRR, the understatement is 75%.

The practical implication: if you use simple LTV to set your CAC budget (via LTV/CAC ratio), you will systematically under-invest in acquisition by the same percentage. A company with 40% understated LTV that sets its CAC budget at 3x LTV is actually operating at a 2.1x effective LTV/CAC ratio, leaving significant acquisition investment capacity on the table.

When to Use Each Formula

The shortcut decision rule: if your NRR is consistently above 105%, use the expansion-adjusted formula and note the NRR assumption explicitly. If you have 24+ months of cohort data, validate the formula output against actual cohort LTV. If they diverge by more than 20%, re-examine your churn and expansion rate inputs.

The LTV/CAC Connection

LTV is most operationally useful in the LTV/CAC ratio, which determines how much you can sustainably spend to acquire a customer. The standard benchmark is 3:1 — $3 of LTV for every $1 of CAC.

The formula impact on CAC budgets:

If your true LTV is $30,000 (expansion-adjusted) but you calculate $7,500 (simple), you set your CAC budget at $2,500 (3:1 simple). But you could sustainably acquire at $10,000 (3:1 expansion-adjusted) — four times your assumed budget.

The under-investment means slower growth, smaller sales team, fewer acquisition channels, and slower market penetration. The company that correctly measures LTV grows faster with the same capital efficiency — because it is willing to spend what the unit economics actually support.

The NPV adjustment for LTV/CAC:

Any LTV projection beyond 36 months should be discounted. Undiscounted LTV of $30,000 over 10 years is not the same as $30,000 today. Using a 12% discount rate (common WACC for growth-stage SaaS):

• Undiscounted LTV over 200 months: $30,000
• Discounted LTV at 12% annual rate: $24,300

For LTV/CAC purposes, use discounted LTV if projecting beyond 3 years. For quick benchmarks, undiscounted LTV with a shorter assumed lifetime (36-48 months) is more conservative and usually more defensible.

Segmenting LTV: Why Averages Are Useless

Blending LTV across customer segments produces a number that is wrong for every segment.

Illustrative example:

• SMB customers: $200/month ARPU, 5% monthly churn, 0% expansion → LTV = $3,000
• Mid-market customers: $2,000/month ARPU, 1.5% monthly churn, 1% expansion → LTV = $120,000
• Enterprise customers: $8,000/month ARPU, 0.5% monthly churn, 1.5% expansion → LTV = >$1,000,000 (net negative churn)

Average LTV across 100 SMB + 20 mid-market + 5 enterprise customers = ($300,000 + $2,400,000 + $5,000,000) / 125 = $61,600

This average is the right answer for no customer in your portfolio. The SMB segment justifies a CAC of ~$20,000 at 3:1, which is absurd for a $200/month customer. The enterprise segment justifies a CAC of $300,000+, which your current CAC budget probably doesn't approach.

Segment LTV by at minimum:

• Customer tier (SMB / Mid-market / Enterprise)
• ICP match (ICP-fit vs. non-ICP-fit)
• Industry vertical (if metrics diverge significantly by vertical)
• Acquisition channel (PLG-sourced customers often have different LTV profiles than outbound-sourced)

For customer health scoring and expansion prioritization, see customer health scoring — the LTV segmentation should align with your health score definitions.

Red Flags in LTV Reporting

Red Flag 1: Using Simple LTV for Fundraising with a High-Expansion Product

If your NRR is 115%+ and you present simple LTV to VCs, you are understating your unit economics. Investors who understand the metric will adjust the number upward and view your use of the simple formula as either analytical naivety or conservatism that needs explaining.

Red Flag 2: No LTV Segmentation by ICP

Reporting a single average LTV is only acceptable at pre-product-market-fit stage when you have <50 customers and insufficient data to segment. At >200 customers, segmented LTV is a basic analytical requirement.

Red Flag 3: Using LTV as Justification for High CAC Without Cohort Validation

"Our LTV/CAC is 4:1" is only a useful statement if the LTV number is grounded in observed cohort data. Formula-based LTV projections using assumed churn rates are not validation — they are projections that can be manipulated by adjusting the churn assumption. Before investing heavily in acquisition based on LTV/CAC, validate the LTV against actual cohort data from your earliest customer cohorts.

Red Flag 4: Ignoring Churn Timing in LTV Calculations

The simple formula assumes churn is uniform over time. In practice, SaaS churn is heavily weighted toward the first 90 days (early churn from poor activation) with a lower tail for long-tenured customers. A 2% monthly average churn rate that is actually 6% in months 1-3 and 0.8% in months 4+ produces a different LTV profile than uniform 2% churn. Early churn concentrated before value delivery means the simple formula overstates LTV for products with poor activation — the opposite of the expansion issue. See churn rate analysis for how to decompose churn by tenure.

Building LTV Measurement Infrastructure

To move from formula-based LTV to cohort-based LTV, you need:

1. Customer-level revenue tracking — every dollar of revenue attributed to a specific customer and contract, not just aggregate MRR
2. Cohort assignment — each customer tagged with a cohort start date (first payment date, not trial start)
3. Expansion tracking — upsell and cross-sell revenue attributed to the originating customer record
4. Churn event logging — cancellation date, reason, and ARR at time of churn for every churned customer
5. Cohort analysis tooling — segment customers by start cohort, compute cumulative revenue by cohort-month, and display as the standard revenue retention curve

Most product analytics platforms (Amplitude, Mixpanel) handle event cohorts. Revenue cohorts require either your billing platform's native analytics (Stripe Revenue Recognition, Chargebee, ChartMogul) or a custom SQL model on top of your billing data. See cohort analysis guide for the complete infrastructure setup.

Conclusion

The simple LTV formula is a useful first approximation. It is not a useful basis for setting CAC budgets, evaluating unit economics, or presenting to investors — not for any company with meaningful expansion revenue, which is most B2B SaaS businesses above $1M ARR.

The expansion-adjusted formula corrects the structural flaw. The cohort-based measurement validates the projection. Used together — formula for forward modeling, cohort for backward validation — they give you the most accurate picture of what your customers are actually worth.

The implication for acquisition investment is significant: if your NRR is above 110% and you have been budgeting CAC based on simple LTV, you have been under-investing in growth. Recalculate. Segment by ICP. And align your LTV/CAC thinking with the customer health scoring and NRR analysis that tell you which customer segments are driving your highest lifetime value.

Use our calculator to compute expansion-adjusted LTV with your actual churn and expansion rates, and review our pricing guidance for how contract structure affects long-term LTV.

Frequently Asked Questions

Why does the simple LTV formula understate customer lifetime value?

Simple LTV (ARPU × Gross Margin / Churn Rate) assumes revenue per customer is flat forever. In reality, customers who stay also expand — they upgrade, add seats, and buy add-ons. A customer paying $100/month who expands to $150/month after 6 months generates significantly more lifetime value than the simple formula predicts. The expansion component adds 40-80% to true LTV depending on your net revenue retention rate.

What is the expansion-adjusted LTV formula?

Expansion-adjusted LTV = ARPU × Gross Margin / (Monthly Churn Rate - Monthly Net Expansion Rate). If your monthly churn is 2% and your monthly net expansion rate is 1.5%, the effective denominator is 0.5%, which triples LTV relative to the simple formula. If net expansion exceeds churn, LTV is theoretically infinite — the cohort grows in value over time, not decays.

When should I use cohort-based LTV instead of formula-based LTV?

Use cohort-based LTV (NPV of actual observed revenue) when you have 24+ months of customer data and want the most accurate measurement for fundraising, unit economics reporting, or pricing decisions. Cohort LTV is the ground truth; formula-based LTV is a projection model. If they diverge significantly, your formula assumptions (churn rate, expansion rate) are wrong.