How the two formulas work
Repeat purchases. Order value × orders per year × years gives lifetime revenue. Multiplying by gross margin turns it into the gross profit that can actually pay for advertising.
Subscriptions. Divide monthly gross profit per customer by the share of customers who cancel each month.
Dividing by churn works because of a simple series. With 4% monthly churn, a customer pays in month one, is still there in month two with a 96% chance, in month three with a 96% × 96% chance, and so on. Those chances add up to 1 ÷ 0.04 = 25 months, the expected lifespan, and $56 of monthly gross profit × 25 months = $1,400.
The “first 12 months” figure stops the same series after a year: monthly gross profit × (1 − (1 − churn)12) ÷ churn, or $56 × (1 − 0.9612) ÷ 0.04 ≈ $542 at the defaults. As a cross-check, Twilio’s CLV guide (dated Mar 10, 2026, last verified 2026-10-05) uses the same structure: $100 a month, 80% margin and 5% churn gives ($100 × 0.80) ÷ 0.05 = $1,600. Subscription mode returns the same answer. Twilio divides by revenue churn rather than customer churn; the two match when every customer pays the same.
Monthly churn, lifetime and LTV
Churn does most of the work in subscription mode. Each row runs the subscription defaults through the calculator’s formula; the 4% row is the worked example above.
| Monthly churn | Average lifetime (months) | LTV on margin | LTV on revenue |
|---|---|---|---|
| 1% | 100 | $5,600 | $8,000 |
| 2% | 50 | $2,800 | $4,000 |
| 3% | 33.3 | $1,866.67 | $2,666.67 |
| 4%, worked example, matches the calculator’s Subscription mode | 25 | $1,400 | $2,000 |
| 5% | 20 | $1,120 | $1,600 |
| 7% | 14.3 | $800 | $1,142.86 |
| 10% | 10 | $560 | $800 |
Halving churn doubles both the lifetime and the LTV, so the low-churn rows are the ones to doubt. A one-point drop from 2% to 1% adds $2,800 of LTV on margin, while the same drop from 5% to 4% adds $280. That is why the calculator warns when churn is under 0.5% a month.
Why this calculator starts from margin, not revenue
A common shortcut stops at order value × frequency × lifespan. That is lifetime revenue, and it overstates what a customer is worth by 1 ÷ margin. Here is the repeat-purchase example both ways, with a $120 acquisition cost. The numbers are illustrative, not a benchmark.
| Example | Revenue LTV | Gross-margin LTV |
|---|---|---|
| Lifetime value | $720 | $288 |
| LTV:CAC at $120 per customer | 6:1 | 2.4:1 |
| Most you can pay per customer and break even over the lifetime, before overhead | $720 | $288 |
| Subscription example ($80/mo, 70% margin, 4% churn) | $2,000 | $1,400 |
At a 40% margin the revenue figure is 2.5 times the real one. Set ad targets from $720 and you can pay several hundred dollars for a customer who will only ever return $288 of gross profit.
Where to find each input
- Average order value: revenue ÷ orders for the same period, after refunds and returns.
- Orders per year: orders in the last 12 months ÷ customers who ordered in that time.
- Years retained: the hardest number. If your business is younger than the lifespan you want to type, you are guessing. Use 1 and read the 12-month figure.
- Gross margin: revenue minus product, shipping and payment costs, ÷ revenue. Leave ad spend out; it belongs on the acquisition side.
- Monthly churn: customers who cancelled in a month ÷ paying customers at its start. Average several months.
A platform’s CLV is not this number. In its help article (updated Aug 5, 2025, last verified 2026-10-05), Klaviyo defines Historic CLV as “the total value of all previous orders an individual has made, taking into account any refunds and returns” and Predicted CLV as “a prediction of how much money a particular customer will spend in the next year.” Both are revenue, and the predicted one covers a year, not a lifetime. Klaviyo only shows predictions once an account has at least 500 customers with orders, 180 days of order history with orders in the last 30 days, and some customers with 3 or more orders, and says its predictions “are not expected to be exact for any single individual.”
Google Ads’ New Customer Value setting is a bidding option, not an LTV calculator. Its help page (no date shown, last verified 2026-10-05) says “We recommend New Customer Value for all advertisers with purchase conversion goals” and requires at least one Purchase conversion goal. It does not say it works out lifetime value for you.
Eight ways LTV goes wrong
- Revenue instead of margin: $720 instead of $288 in the example.
- Trusting 1 ÷ churn when churn is tiny or new. At 0.5% a month the formula implies 200 months, over 16 years. With under about six months of churn data, we would not trust it.
- Mixing units: annual order frequency with monthly churn, or 0.04 in a field that expects 4%.
- One average for different customers. Deep-discount customers and ones who found you on their own can behave differently. Split them if you can.
- Calling LTV profit. It is before acquisition cost and overhead.
- Copying a platform’s CLV. Those are revenue, and historic or next-year.
- Ignoring refunds and returns. They come straight off order value.
- No time frame. A $60 CPA looks fine next to a $288 LTV, but at $24 of gross profit per order it takes 2.5 orders to earn back, most of a year at 3 orders a year. The CAC, LTV and payback calculator works out payback.
What this number can’t tell you
Unless your gross margin is 100%, this calculator will always say you can afford less per customer than a revenue-based one does. That points to smaller ad budgets, which is not what a company that sells ad management usually tells you. We think the smaller number is the right one.
LTV is only as good as the lifespan or churn you type in, and nothing AdFlint does makes those guesses better. With a few months of data, a lifetime figure is a forecast dressed up as a measurement. The 12-month figure is easier to defend: $72 for the repeat-purchase example and about $542 for the subscription one. Even Google’s LTV calculator for Android developers (no date shown, last verified 2026-10-05) says its model “makes various assumptions which aren’t guaranteed or confirmed.”
LTV to CAC, worked through
LTV:CAC divides lifetime value by customer acquisition cost (CAC), what it took to win the customer. At a $120 CAC, the repeat-purchase example reads 2.4:1 on margin and 6:1 on revenue. Type $120 into the calculator’s optional acquisition-cost field to see both.
The usual yardstick, 3:1, is investor David Skok’s rule of thumb from his 2010 SaaS Metrics post (last verified 2026-10-05). It rests on his experience with SaaS companies, not a study, so treat it as a starting point, especially outside SaaS (what we could not confirm). Against it, the margin figure falls short and the revenue figure clears it twice over: same customer, opposite answers.
Working back from 3:1, a $288 margin LTV means CAC has to stay at or under $96; at a $120 CAC, margin LTV has to reach $360. The ratio also says nothing about time. At $24 of gross profit per order, $120 takes 5 orders to earn back, roughly 20 months at 3 orders a year. The CAC, LTV and payback calculator works out the payback period for you.
Using LTV to set ad targets
LTV caps what a customer can cost. Four checks connect it to your campaigns:
- First-order break-even: gross profit per order ($24 here) is the most a sale can cost before you rely on repeat orders. Check it with the CPA calculator.
- Break-even ROAS: on one order, the break-even ROAS is 1 ÷ margin, 2.5 at 40%. The ROAS calculator compares it with your real spend.
- Cost per click: a target CPA multiplied by your conversion rate is the most you can pay per click before the CPA target breaks. The conversion rate calculator gives you the rate.
- Budget: the ad budget calculator shows how many customers a monthly budget buys.
If you bid to a target CPA or ROAS, set it from margin LTV you can afford to wait for, not revenue LTV. See the guides to Google Ads bidding strategies, Meta objectives and bidding and ad budgets, pacing and pricing, or the glossary entries for LTV and CAC.