Most DTC brands quote a customer lifetime value figure built on gross revenue, then scale ad spend against it and run out of cash 90 days into the ramp. The number is wrong by 25 to 50 percent before the first dollar goes into Meta, because revenue is not what a customer is worth. This page walks the three legitimate ways to calculate LTV, anchors every one of them in contribution margin, lists the five inputs you have to get right, and names the six mistakes that wreck the figure on the way to the board deck.
The Short Answer: How to Calculate Customer Lifetime Value
LTV = AOV x annual purchase frequency x customer lifespan in years x contribution margin %. The non-negotiable is contribution margin, not revenue. Using revenue inflates LTV by 25 to 50 percent and is how DTC brands talk themselves into unaffordable CAC.
That is the simple historic formula. It has two siblings that exist for different reasons:
- Simple historic LTV. Averages the past 12 to 24 months and projects forward. Use it for directional planning, not media decisions.
- Cohort-based LTV. Tracks the actual cumulative margin of a specific acquisition group at month 1, 2, 3, 6, 12, and 36. Use it for finance and channel quality.
- Predictive LTV. Machine learning forecasts who a customer will become by Day 30. Use it for daily bidding and segmentation.
Quick worked example using the simple formula. An apparel brand runs a $90 AOV, 2.4 orders per customer per year, a 2-year lifespan, and a 55 percent contribution margin.
$90 x 2.4 x 2 x 0.55 = $237.60 LTV.
Every methodology on this page converts to contribution margin, or it lies to you.
Why Contribution Margin Is the Only Honest Basis
There are three strata of DTC profit, and they are not interchangeable.
- Gross revenue. Total cash collected from the customer.
- Gross margin. Revenue minus Cost of Goods Sold.
- Contribution margin. Gross margin minus every variable cost required to deliver that order: fulfillment, outbound shipping, packaging, payment processing, variable marketing.
The number that matters: top-quartile DTC brands maintain a 54 to 56 percent contribution margin, while the median sits dangerously around 25 percent. SaaS uses revenue and margin almost interchangeably because gross margins clear 90 percent. Physical product does not. Revenue LTV overstates a customer by the margin gap, and that gap is where the DTC carnage of the last few years lived.
Worked contrast. A customer places 3 orders at $80 each. The brand runs a 50 percent gross margin and pays $12 in fulfillment plus $8 in processing per order, leaving $20 of contribution per order.
- Revenue LTV: $240. Against a $55 CAC, that is a 4.4x ratio. Looks healthy.
- Contribution-margin LTV: $60. Against the same $55 CAC, the real ratio is 1.1x. The brand is breaking even on variable costs and will bleed cash once fixed overhead is layered on.
The Casper case is the cautionary version of this at scale. A $765 AOV, an estimated $302 CAC, gross margins of 43 to 51 percent, and a one-and-done category that structurally caps repeat revenue. Retrospective analysis put the real LTV:CAC near 1.4x; once heavy marketing, fulfillment, and returns were factored in, the company was effectively losing hundreds of dollars per mattress shipped. It raised over $340 million chasing a number that was a mirage.
For the definition of contribution margin itself and the breakeven ROAS that falls out of it, the deeper read is on contribution margin.
The Five Inputs You Have to Get Right
Garbage inputs yield garbage LTV. These are the five.
Average Order Value. Total DTC revenue divided by DTC orders. Strip wholesale, B2B, and corporate-gifting anomalies before you calculate it. Single transactions above $2,000 or carts above 50 units will distort the average enough to break every downstream decision. AOV is a direct-to-consumer-only number.
Purchase frequency. Total orders divided by unique customers across a 12-month window. Frequency is category-bound, not aspirational. A mattress runs about 1.1 over three years. A daily supplement runs around 6.0 over the same period. Use your category's actual rhythm.
Contribution margin percent. (Gross profit minus variable costs) divided by revenue. Recall the spread: top-quartile DTC is 54 to 56 percent, median is around 25 percent. If you do not know your number to the point, you are not ready to set a CAC ceiling.
Customer lifespan. In subscription, churn is binary and observable. In non-subscription DTC, a customer never tells you they are gone, they simply stop showing up. Statistically, define a customer as churned when they have not repurchased within 2 to 3 standard deviations beyond the average inter-purchase time. For the subscription-specific churn mechanics, route to subscription LTV.
Time horizon. "Lifetime" is a fiction beyond a point. Operators cap LTV at three horizons:
- 60-day LTV for cash flow decisions: what CAC you can afford today.
- 12-month LTV for annual planning.
- 36-month LTV as the practical valuation ceiling. Beyond 36 months the math is noise.
Method 1: Simple Historic LTV (The Baseline)
The formula, restated:
LTV = AOV x annual purchase frequency x customer lifespan (years) x contribution margin %
Worked example, line by line:
- AOV: $90
- Purchase frequency: 2.4 orders per year
- Lifespan: 2 years
- Contribution margin: 55%
- LTV: $90 x 2.4 x 2 x 0.55 = $237.60
Where this number belongs: pitch decks, board narrative, internal sanity checks. Not media decisions, and never campaign-level decisions.
The honest limits are three:
- Homogeneity assumption. It treats every customer like the mathematical average. They are not. The average represents nobody.
- No cohort decay. It assumes purchase frequency stays steady across the lifespan. Real engagement spikes in month 1 and decays logarithmically.
- Blends cohort quality. A customer acquired via a 40-percent-off welcome code is averaged in with a full-price customer. Studies of discount cohorts (greater than 30 percent off) show they repeat-purchase 35 to 50 percent worse than average. The blend hides that.
Useful as a baseline. Dangerous as a decision tool.
Method 2: Cohort-Based LTV (The Standard for Real Decisions)
Group customers by the month of their first purchase. Track the cumulative contribution margin that exact group generates at month 1, 2, 3, 6, 12, and 36. Each cohort is its own curve.
The formula is iterative:
Cohort LTV at Month X = (cumulative cohort contribution margin through Month X) / (customers acquired in that cohort)
One cohort walked in plain numbers. A premium supplement brand acquires 1,000 new customers in January at a blended $40 CAC.
- Month 1. The cohort generates $30,000 in contribution margin on first orders. LTV per customer is $30. Against a $40 CAC, the ratio is 0.75x. The brand is cash-negative on the cohort.
- Month 2. 400 of those customers are on a replenishment subscription and renew, adding $12,000 in margin. Cumulative LTV is $42. Ratio is 1.05x. The cohort has officially paid back.
- Month 12. Cumulative margin reaches $90,000. LTV is $90. Ratio is 2.25x.
What the cohort lens surfaces that an average hides:
- Channel quality. Black Friday cohorts often carry a 6-month LTV roughly 30 percent below spring cohorts, because the discount selects for one-time buyers.
- Creative quality. Two ads at the same $35 CPA can produce 90-day LTVs of $62 and $38, a 60 percent gap in real profitability.
- Product gateway effect. A skincare starter bundle can drive a 6-month LTV three times higher than a standalone moisturizer first-order, even at higher CAC.
The honest limit is the time-delay paradox. Proving a January cohort's 12-month LTV requires waiting until December. Excellent post-mortem tool. Slow real-time optimizer.
For the curve construction itself, the channel comparison table, and the budget reallocation logic that follows, the deeper read is on cohort analysis.
Method 3: Predictive LTV (Closing the Time-Delay Gap)
Predictive LTV uses machine learning to forecast who a customer will become by Day 30, so you can bid against future value instead of waiting a year for cohort data to mature.
The formula scaffold mirrors the historic version:
pLTV = predicted transactions x predicted AOV x contribution margin %
The two predicted values come from a statistical pair the industry has standardized on:
- BG/NBD (Beta-Geometric / Negative Binomial Distribution) predicts purchase frequency and churn probability. It watches recency and frequency to decide whether a customer is still active or has quietly churned.
- Gamma-Gamma predicts the monetary value of future orders. It assumes each customer has an unobserved typical order size, and the order-by-order variance is separate from that underlying average.
Worked contrast. A skincare brand spends $100k a month on Meta, runs a $45 CAC and a $62 first-order AOV, leaving roughly $17 of first-order contribution margin. Two new buyers arrive:
- Customer A clicks a high-fidelity video ad, buys a starter kit at full price, opens the post-purchase email within 12 hours.
- Customer B clicks a retargeting ad with 30 percent off, buys the cheapest single SKU, unsubscribes immediately.
Same CAC. Same first order. A simple average treats them as identical. Cohort analysis forces the brand to wait a year to see the difference. Predictive LTV flags the divergence by Day 30: a 3-year pLTV of $400+ for Customer A, stuck at $62 for Customer B. The brand can confidently raise allowable CPA to $80 for A-lookalikes and defund campaigns acquiring B-lookalikes.
The hard volume floor matters. GA4 requires at least 1,000 returning users with purchase events across 28 days to even fit a model. Below 500 customers, the models hallucinate, with error rates routinely between 30 and 40 percent. Predictive LTV is also brittle to product or shipping changes; historical signals miscalibrate when the underlying offer shifts.
Tool landscape, briefly:
- GA4, Shopify native, Klaviyo. Free or included. Rigid, revenue-based not margin-adjusted. Entry-level scoring.
- Lifetimely (AMP). Mid-market Shopify. Pricing recently moved up the stack to roughly $999 to $1,999 per month.
- Pecan AI. Mid-market custom predictive models, starting around $760 per month.
- Voyantis. Enterprise value-based bidding integration, reportedly around $25,000 per month and only sensible above roughly $150k monthly ad spend.
The three-method comparison:
| Method | Complexity | Formula / approach | Primary use | Primary limitation |
|---|---|---|---|---|
| Simple historic LTV | Low | AOV x purchase frequency x lifespan x CM% | Pitch decks, directional baselines, sanity checks | Homogeneous-customer assumption; ignores cohort decay |
| Cohort-based LTV | Medium | Cumulative cohort CM / customers in cohort | Quarterly finance, channel and creative quality calls | Time-delay paradox: 12-month proof takes 12 months |
| Predictive LTV | High | pLTV = predicted transactions x predicted AOV x CM%, via BG/NBD + Gamma-Gamma | Daily media buying, value-based bidding, VIP segmentation | Needs 1,000+ active users; below 500 customers, error rates 30 to 40 percent |
The Number Most Founders Skip: Payback Period
LTV without a payback period is a fantasy. The payback period is the number of months it takes for cumulative contribution margin to equal CAC.
A brand can post a 4:1 LTV:CAC at 12 months and still go bankrupt if 60-day payback is 0.4x and ad spend is scaling. The ratio is a long-term truth. Payback is whether you survive to collect it.
2026 DTC benchmarks:
- 1 to 3 months. Exceptional. Self-funded reinvestment cycle, spend recycles fast.
- 3 to 6 months. Standard for brands with proven LTV and stable working capital.
- 6 to 12 months. Tolerable with subscription MRR or committed credit lines. Brittle to cash-flow shocks.
- 12+ months. A red flag for bootstrapped DTC. Scaling paid media against this profile strangles cash within a quarter.
Levers that pull payback forward without touching CAC:
- Raise first-order AOV. Multi-pack minimums on low-AOV items, hero-SKU pricing increases.
- Post-purchase one-click upsells before the thank-you page, while purchase intent is at its peak.
- Raise the free-shipping threshold so customers add high-margin accessory items.
- Unbundle the starter kit to lower the front-end barrier, then sell accessories through zero-CAC email flows.
For the full CAC formula and the payback math expanded, go to customer acquisition cost. For the ratio interpretation and what 3:1 actually means for physical product, go to LTV CAC ratio.
LTV:CAC Benchmarks for DTC in 2026
The 3:1 rule SaaS investors cite was built on 90 percent gross margins. Applying it cleanly to a DTC brand running 25 to 56 percent contribution margin produces either a number you cannot hit or a number that does not protect the cash account.
The brand standard in 2026:
- 60-day contribution-margin payback at 1.2:1. This is the budget discipline.
- 24-month LTV:CAC at 2:1 or 2.5:1. This is the maturation target.
The attribution caveat matters here. Meta permanently removed the 7-day and 28-day view-through windows in January 2026 and narrowed click attribution to a 1-day engage-through window for social interactions. Platform-reported conversions dropped 15 to 40 percent overnight. The CAC you read in Ads Manager is no longer the CAC you spent.
The fix is a blended CAC built from Multi-Touch Attribution, Marketing Mix Modeling, and server-side tracking via the Conversions API. The deeper reads on the attribution stack are attribution and MER.
The Six Mistakes That Wreck the LTV Number
- The revenue delusion. Treating top-line revenue as LTV inflates the customer by the full margin gap, often 25 to 50 percent, and routes brands into CACs the math will never pay back.
- Aspirational lifespan. Assigning 24 months because the product is "excellent." When real cohort data shows month-6 drop-off, the brand is deep into spend against revenue that will never arrive.
- Ignoring the payback period. A 4:1 12-month ratio with a 12-month payback kills a scaling brand inside a quarter. Healthy ratio and healthy payback solve different problems.
- Skipping the discount rate. A dollar three years out is not worth a dollar today. Future cash on multi-year LTV needs a 10 to 15 percent discount applied, or long-term retention looks falsely competitive with immediate acquisition.
- Calculating LTV too early. Under 500 customers, one whale or a few early churners moves the model 30 to 40 percent. Optimize for first-order profitability and 60-day payback until you have roughly 12 months of cohort data.
- Trusting predictive models pre-significance. pLTV under 1,000 active users produces wildly volatile forecasts. Cohort data stays the ground truth until you cross the volume threshold.
Which Method to Use When
The methods are not in competition. A mature stack runs all three on the same contribution-margin definition.
- Simple historic belongs in pitch decks, board narrative, and quick sanity checks.
- Cohort-based belongs in quarterly financial reviews, channel and creative quality decisions, and the actual scale-or-pause call.
- Predictive belongs in daily media buying, value-based bidding signals fed back to ad platforms, VIP segmentation, and proactive retention flows.
Finance runs cohort. Marketing runs predictive. Both run on contribution margin or both lie.
What This Means for Your Acquisition Spend
The right move is sequential. Rebuild the LTV number on contribution margin. Re-derive the allowable CAC against the 60-day payback rule. Most brands discover they have been overspending against a phantom customer for quarters.
From there, three downstream moves, each owned by the page that goes deepest on it:
- Build the channel cohort curves and compare media sources on real 90-day retention: cohort analysis.
- If you run subscription, model retention on the actual churn curve, not an assumed lifespan: subscription LTV.
- Pull the operational levers that raise LTV directly (bundles, post-purchase, retention flows): raise LTV.
For the broader metric stack this LTV number plugs into, the entry point is paid media metrics.
Get the LTV Number Right Before You Scale
Most teams chasing LTV problems do not need another dashboard. They need a margin-honest LTV figure and a payback discipline tied to it. That is the audit.
If your CAC has been creeping and your contribution margin has not been measured to the order, start with the proactive read in our paid media audit. If your ads stopped scaling six weeks ago and you are trying to find the leak, start with why ads stopped working.
The performance creative work we do only pays back when the LTV math underneath it is real. Get the number right, then we scale against it.