Estimating Customer Lifetime Value Without Churn Data

Estimating customer lifetime value before you have customers requires borrowing retention data from competitors and accepting that your early math will be wrong by a factor of three. You cannot measure churn when your user base is zero. You have to synthesize it from category benchmarks and competitor pricing models.

The standard formula for Customer Lifetime Value is Average Revenue Per User divided by Customer Churn Rate. If you charge $20 a month and 5 percent of your customers cancel every month, your LTV is $20 divided by 0.05, which equals $400. This assumes a constant churn rate and a constant revenue stream. Both assumptions are false for a new business.

The Easy Variable and the Missing Variable

Average Revenue Per User is the easy part. You look at the market. If you are building a social media scheduling tool, Buffer charges $6 a month per channel. Hootsuite charges $99 a month. You decide to charge $15. That is your ARPU. You can adjust it later based on annual plans or discounts, but $15 is a concrete starting point.

Churn is the silent killer of early LTV estimates. A monthly churn rate of 5 percent means an average customer stays for 20 months. A churn rate of 10 percent means they stay for 10 months. Without historical data, founders often guess 5 percent because it sounds reasonable. This is a mistake.

You must borrow category benchmarks. Public market data is your first resource. Look at the public filings of traded SaaS companies in your broader category. When project management tools go public, their filings reveal net dollar retention rates. While net dollar retention includes upsells, it gives you a ceiling for your own retention estimates.

For private companies, search for open startup dashboards. Companies like Buffer publish their exact churn rates in real time. If you are building a social media tool, you do not need to guess. You can look at their live dashboard and see their monthly churn fluctuates between 4 and 6 percent. B2C mobile apps often see 70 percent churn in the first 30 days. B2B SaaS targeting small businesses typically sees 3 to 7 percent monthly churn. Enterprise SaaS with annual contracts sees 1 to 2 percent monthly churn. If you are building a consumer habit tracker, using a 5 percent monthly churn assumption will bankrupt your marketing budget.

The Worked Example

Let us build a hypothetical B2B SaaS application. It is an invoicing tool for freelance designers. You price it at $12 a month. You look at public data from open startup pages for small business SaaS. You see an average monthly churn of 8 percent.

Your basic LTV equals $12 divided by 0.08. Your estimated LTV is $150. This means you can spend up to $50 to acquire a customer and maintain a healthy three to one ratio between lifetime value and acquisition cost.

This estimate will be wrong by a factor of three.

New products are buggy, lack features, and attract the wrong early users. Your first cohort will not churn at the category average of 8 percent. They will likely churn at 24 percent or more. Why exactly does a new product suffer this penalty? It comes down to product market fit and acquisition channels. Established companies have optimized their onboarding over years. They know exactly which marketing channels bring in users who stay. On day one, you have zero optimization. Your onboarding sequence will have friction. Your core features will have edge case bugs.

More importantly, your early marketing will attract the wrong people. You will acquire users who thought your product did something else, or who only need it for a one off project. If your early churn is 24 percent, your actual LTV is $12 divided by 0.24, which is $50. If you are spending $50 to acquire a customer based on your borrowed benchmark of $150 LTV, you are breaking even on gross revenue and losing money after operational costs. Your estimate was off by exactly a factor of three.

Sensitivity Analysis

Sensitivity analysis is not an academic exercise. It dictates your survival. Let us model a consumer subscription app priced at $10 a month. You plan to use paid social ads to acquire users at $15 per install.

If your monthly churn is 10 percent, your LTV is $100. A $15 acquisition cost gives you a staggering 6.6 ratio. You print money. But consumer apps rarely see 10 percent monthly churn. If your actual churn is 30 percent, your LTV collapses to $33. Your ratio is now 2.2. After app store fees and server costs, you are losing money on every ad click. A 20 percent absolute shift in churn destroyed the entire business model. You must build a spreadsheet that calculates LTV across a spectrum of churn rates, from wildly optimistic to historically catastrophic.

Refining Estimates with Demand Signals

You can reverse engineer competitor churn assumptions from their pricing and acquisition channels. If a competitor charges $10 a month and relies heavily on paid search ads that cost $3 per click, they need a high LTV to survive. If they convert 5 percent of clicks, their acquisition cost is $60. To maintain a healthy ratio, their LTV must be at least $180. Therefore, their churn must be around 5.5 percent. You can use this deduced churn rate as a baseline for your own estimates.

You can adjust your borrowed benchmarks by looking at demand signals across the web. When users leave a competitor, they leave a trail. They post in niche communities about why they canceled. They leave app store reviews detailing missing features. These are demand signals. If a competitor has a high volume of demand signals related to poor customer support, you can assume their churn is higher than the category average. If you build a product with superior support, you can justify estimating a slightly lower churn rate for your own model. The Unthinkable iOS app surfaces these exact business and app ideas from real demand signals across the web, allowing you to spot where incumbents are bleeding users.

The Two Stage Model

The basic LTV formula assumes churn is linear. It is not. Churn follows a power law. A massive percentage of users will leave in the first 30 days. Those who survive the first month are much more likely to stay for a year. For a new business, you should estimate LTV in two stages. Stage one is the first 90 days. Stage two is the residual lifetime.

Assume 100 users sign up for a $20 a month product. In month one, 40 percent churn. You have 60 users left. In month two, 20 percent of the remaining users churn. You have 48 users left. In month three, 10 percent churn. You have 43 users left. Your revenue for the first three months is $2000 plus $1200 plus $960, totaling $4160.

Now you have 43 stabilized users. You can apply a standard category benchmark churn rate to them, say 5 percent. Their residual LTV is $20 divided by 0.05, which is $400 each. For 43 users, that is $17200. Add the initial $4160, and your total cohort LTV is $21360. Divide by the original 100 users, and your true estimated LTV is $213.60.

If you had just used a blended guess of 15 percent churn on the standard formula, LTV would be $20 divided by 0.15, or $133. The two stage model is more accurate because it reflects how humans actually adopt software. They try it, most leave immediately, and the rest stay for a long time.

Discount Rates and Gross Margins

For a new business, money today is worth vastly more than money in three years. The standard LTV formula ignores the time value of money. If your churn is 2 percent, the formula says the customer stays for 50 months. That is over four years.

You must discount future cash flows. A common discount rate for startups is 12 percent annually, which is roughly 1 percent a month. Instead of dividing ARPU by Churn, divide ARPU by the sum of Churn plus the monthly discount rate. Back to our $20 ARPU and 5 percent churn. Unadjusted LTV is $400. Discounted LTV is $20 divided by the sum of 0.05 and 0.01. That is $20 divided by 0.06, which equals $333. You just lost $67 of perceived value simply by acknowledging that future money is less useful to a cash starved founder.

Gross margin is also frequently ignored by founders who want to see a large LTV number. This is a fatal error. Lifetime value is a measure of the cash a customer contributes to your business to cover fixed costs and generate profit. Revenue is not cash. If you run a computationally heavy AI tool, your cost of goods sold will be massive.

Let us say you charge $30 a month. Every active user costs you $12 a month in API tokens and database hosting. Your gross margin is $18. If your churn is 10 percent, your revenue based LTV is $300. Your actual, margin based LTV is $180. If you set your marketing budget based on the $300 figure, you will run out of cash before the end of your second quarter. Always multiply your ARPU by your gross margin percentage before dividing by churn.

Validating the Estimate

The moment you launch, your estimate becomes a hypothesis to be tested. You will not have true churn data for months, but you will have proxy metrics.

Track your activation rate. If your category benchmark assumes a 40 percent first month churn, but only 20 percent of your new signups complete the core onboarding task, your first month churn will be catastrophic. You do not need to wait 30 days to know your LTV estimate is wrong. You know on day one.

Track how often users log in. If your product is a daily habit tracker, and the average user logs in twice a week, they will churn. Your borrowed LTV estimate is already invalidated.

Estimating LTV for a new business is an exercise in risk management, not accounting. You borrow benchmarks to set a baseline. You reverse engineer competitors to check your logic. You apply two stage models and discount rates to prevent dangerous optimism. You accept that early product friction will divide your estimate by three. And you watch your early activation metrics closely to see just how wrong your initial math really was.