Defining the core metrics

The question that changes the room

The investor didn’t ask about growth or market size. He asked: why do you believe your customer will stay for 3.5 years?

I had just walked him through PizzaPortal’s unit economics. He had the numbers in front of him. He could see the payback math. And instead of questioning our marketing spend or our competitive position, he went straight to the assumption buried inside the formula. How long does a customer stay? Everything else is arithmetic. That one variable is a bet.

He was right to ask. Because if our average customer churned after eighteen months instead of forty-two, the entire model collapsed. Same acquisition cost. Same contribution per order. Completely different business.

That question sits at the center of every business I have built, invested in, or advised since. Customer Lifetime Value (LTV) tells you how much gross profit a single customer generates over the time they stay with you. Customer Acquisition Cost (CAC) tells you how much you spend to get that customer through the door. The gap between them is where your company lives or dies.

Most founders can define both terms on a whiteboard. Very few can calculate them honestly for their own business. The difference between knowing the concept and running the math with real inputs is the difference between a pitch deck and a plan.

Let me show you how the formulas actually work.

Customer Lifetime Value (LTV)

LTV is the total gross profit one customer generates over their lifetime. Not revenue. Gross profit. If you build your model on revenue, every projection will be too optimistic, and you will overspend on acquisition without understanding why the cash keeps disappearing.

The base formula:

LTV = ARPU x Gross Margin % x Customer Lifetime

ARPU (Average Revenue Per User) is the average amount one customer brings in during a given period. For a subscription business, take your monthly recurring revenue and divide it by the number of paying customers. For a transactional business, it is revenue per transaction averaged across the customer base. At PizzaPortal, our monthly ARPU was roughly PLN 3, because the average customer ordered once a month at about PLN 30 on a 10% commission.

Gross Margin % is what you keep from every euro of revenue after paying the direct cost of delivering your product or service. For a software company, that cost includes hosting and customer support. For an e-commerce business, it includes production, packaging, fulfillment, and shipping. At PizzaPortal, the margin on our commission left us with PLN 2.40 per order. If I had built projections on the PLN 3 commission instead, every number would have been 25% too optimistic.

The revenue-versus-profit distinction matters here. A customer who generates €100 in revenue but costs you €80 to serve is worth €20. If you build your LTV on the €100, you will overspend on acquisition and never understand why the cash keeps disappearing. Use contribution margin if that fits your model better, but always subtract the direct cost of delivery. You want to know what’s left after serving the customer, not what comes in the door.

Customer Lifetime is how long a customer stays. The standard shortcut: divide 1 by your monthly churn rate. If 5% of your customers leave every month, the average customer sticks around for 20 months (1 / 0.05). If monthly churn is 2%, average lifetime jumps to 50 months.

The problem with this formula is that it assumes churn is constant. In practice, most businesses see higher churn in the first few months, then a flattening curve as loyal customers settle in. Many businesses overestimate lifetime by looking at early adopter behavior, people who love you and tell their friends, and assuming everyone will behave the same way. Cohort data corrects for this. If you don’t have cohort data yet, be conservative. Assume shorter lifetimes and be pleasantly surprised rather than building a business on hope.

Here’s a simple example. A SaaS product charges $10 per month. Gross margin is 80%. Monthly churn is 5%, giving an average lifetime of 20 months.

LTV = $10 x 0.8 x 20 = $160.

Each customer delivers $160 in gross profit over their lifetime. If your CAC is under $160, you’re making money. If it’s above that, you are paying for the privilege of losing money on every customer you acquire.

A founder I advise ran this calculation for the first time last year. He had been in business for two years. He sat back from his laptop and said nothing for about ten seconds. Then: “So we’ve been losing money on every customer since we launched.” He had the revenue to prove otherwise. He did not have the gross profit.

Real businesses are messier. Take a B2B SaaS company with $250 monthly ARPU, 70% gross margin, and 3% monthly churn. The quick formula gives LTV = $250 x 0.7 x 33.3 = roughly $5,830. But that number assumes every customer behaves identically for the same amount of time.

In practice, some customers upgrade after month six and their ARPU jumps from $250 to $400. Others churn in month two, having never fully adopted the product. Gross margin shifts as you scale: support costs per customer drop, but infrastructure costs change too. A more honest approach is to project each month’s revenue for a cohort, apply churn probabilities at each stage, include expected upsells and account expansions, and optionally discount future cash flows if the lifetime stretches beyond two years. Cohort-based LTV reflects how customers actually behave rather than how you wish they would.

Track LTV by acquisition channel and by customer segment. A blended average is better than nothing, but it hides the customers who carry your business and the ones who quietly drag it down. A SaaS company I worked with discovered that customers from their webinar funnel had an LTV of $4,200, while customers from paid social had an LTV of $1,100. The product and pricing were identical. The unit economics were completely different depending on how the customer found them.

I had a version of this conversation with a Warsaw-based founder over coffee in 2022. He had the segmented data on his phone. He showed me the paid social column first, then scrolled right. His face did not change, but he stopped talking mid-sentence. The next month, his paid social budget dropped by 60%.

Customer Acquisition Cost (CAC)

How much did you actually spend to get that customer? Not just the ad click. Everything. Customer Acquisition Cost (CAC) is the total cost of turning a stranger into a paying customer.

The formula is deceptively simple:

CAC = Total Acquisition Spend / Number of New Customers

The catch is in what you include in “Total Acquisition Spend.” Most founders count ad spend and stop there. That’s like calculating the cost of running a restaurant and forgetting to include the staff.

A real CAC calculation includes every cost involved in bringing a customer from awareness to first payment. Paid advertising on Google, Meta, LinkedIn, TikTok, and whatever else you run. Agency and freelancer fees for anyone outsourced to work on acquisition. The gross salaries, employer taxes, social security contributions, and commissions of everyone who touches the acquisition process: marketers, SDRs, BDRs, account executives. The CRM, the attribution software, the email automation platform, and any other tool primarily used for lead generation and conversion. And the content that converts: the videos, the landing pages, the email sequences.

If a tool or a team member splits their time between acquisition and retention, allocate a percentage of their cost based on time or output. Precision here is not perfectionism. It’s the difference between understanding your business and guessing.

Let me show you the difference this makes. A startup spends €20,000 on ads, €12,000 on sales team salaries, and €4,000 on marketing software. They acquire 500 new customers. CAC = €36,000 / 500 = €72.

That looks manageable. Now add the costs they left out. €6,000 for video production and creative assets. €3,000 for an agency retainer. €7,000 in commissions and employer-side social security. The real total is €52,000. The real CAC is €104.

That’s a 44% gap. If your pricing and LTV model were built around €72, every strategic decision based on that number is built on faulty math. And you won’t know it until the cash runs out.

Early-stage companies underestimate CAC in predictable ways: they count only ad spend, forget salaries and tools, treat marketing KPIs like cost-per-lead or CPM as proxies for actual CAC, or stop updating the number as their team and cost structure evolve. Each of these shortcuts produces a ratio that looks healthier than it is.

Track CAC monthly, by channel, and in two versions: a simplified channel-specific CAC for tactical decisions, and a fully loaded CAC for strategy. The first tells you which campaigns to scale. The second tells you whether the business works.

The LTV:CAC ratio

Once you have both numbers, the next step is obvious. Divide LTV by CAC.

LTV:CAC = LTV / CAC

This ratio is the single clearest measure of whether acquiring customers makes you money or costs you money. It’s the number every investor will ask about, and the number most founders calculate wrong, usually because the inputs are wrong.

If your LTV is $600 and your CAC is $150, your LTV:CAC ratio is 4.0. For every dollar you spend acquiring a customer, you get four back in gross profit over that customer’s lifetime. Here’s what the ranges mean in practice.

Below 1.0 means every new customer is a net loss. Between 1.0 and 3.0, the business is not yet self-sustaining. Between 3.0 and 5.0 is where most healthy companies operate. Above 5.0 can signal efficiency or under-investment. I will walk through each range in detail, with specific actions for each, in Chapter 4.

I once reviewed a pitch deck where the ratio sat at 6.8. The room was impressed. I asked when they had last calculated it. The founder checked his notes. The number was fourteen months old. The market had shifted, the ad costs had doubled, and no one had rerun the math.

CAC payback period

The LTV:CAC ratio answers whether acquisition is profitable. The payback period answers when. These are different questions, and founders who ignore the second one run out of cash while the first one still looks great on paper.

Payback = CAC / (ARPU x Gross Margin)

This gives you the number of months it takes to recoup the money you spent acquiring a customer. The difference between a 6-month payback and a 15-month payback is the difference between a company that funds its own growth and a company that needs a new funding round every year.

Take a SaaS product. CAC is $240. Monthly ARPU is $30. Gross margin is 75%, so monthly contribution is $22.50. Payback = $240 / $22.50 = 10.7 months. If the average customer stays longer than 10.7 months, the acquisition pays for itself. Everything after that is profit. Everything before that is capital you’ve advanced and are waiting to get back.

Now take a different model. A company sells solar roofs. CAC is €1,200. Sale price is €15,000. Gross margin is 35%, producing €5,250 in profit on the first transaction. If the customer pays upfront, the full CAC is recovered immediately. Payback is instant.

At SunRoof, this was roughly our reality. The sales cycle could be long, weeks or months of consultations and proposals. But once the contract was signed and the roof installed, the economics resolved in a single transaction. In businesses like this, conversion rate and sales velocity matter more than monthly contribution. Your capital is tied up in fewer, higher-value deals, and the risk is not slow payback but slow sales.

And then there are low-margin, high-frequency businesses. Apply the payback formula to PizzaPortal. PLN 100 to acquire a customer, PLN 2.40 per order in contribution. That works out to about 42 orders to break even. If the customer ordered once a month, the payback stretched to roughly three and a half years.

This is the reality of many marketplace and delivery businesses. The numbers only work with high retention and high order frequency. Accept a long payback only if your cohort data shows customers actually stick around long enough. If it doesn’t, no financial model will fix the problem.

As a rough guide: payback under 6 months is excellent. You can reinvest quickly without external capital. Between 6 and 12 months is acceptable for most subscription and SaaS models, and usually the upper limit investors want to see. Above 12 months is risky unless your LTV:CAC ratio is strong and churn is near zero. At that point, you are betting on patience, and patience requires either deep pockets or very favorable credit terms.

The target depends on the business model. One-time, high-value sales like solar installations or enterprise software licenses should recover CAC in the first transaction. Recurring-revenue businesses like SaaS and media subscriptions should target payback under 12 months. Low-margin transactional models like delivery platforms and e-commerce need to track cohort behavior obsessively, because averages will hide the customers who never come back. Enterprise B2B with long sales cycles can tolerate 12 to 18 months if churn approaches zero and annual contract values are large enough to justify the wait.

If your growth depends on raising money every 12 months and your payback period is 14 months, you are structurally mismatched. Either improve the payback or reduce the burn. There is no third option.

The numbers that lie

The most dangerous version of bad unit economics is the one that looks good on a slide. I have seen it happen dozens of times. A founder walks into a board meeting, presents a healthy-looking LTV:CAC ratio, and everyone nods. Then someone pulls a thread, and the whole thing unravels. The numbers were technically correct. They were also deeply misleading.

Here are three patterns that catch the most founders off guard.

Rising CAC with flat or declining LTV. As you scale, audiences saturate and ad platforms get more expensive. CAC tends to go up over time. If LTV does not also improve, whether through better retention or higher spend per customer, the ratio deteriorates quarter by quarter. You might still be growing revenue while slowly destroying the business underneath. Monitor both numbers together. Growth only works if both sides of the equation move in a direction that keeps the ratio healthy.

LTV inflated by a small segment. Your top 5% of customers might generate ten times the value of everyone else. If your “average” LTV is pulled up by that group, it does not represent your typical customer. It represents your best customer. Build strategy on the median, not the mean. If value is concentrated in a small group, you have two problems: how do you find more customers like the 5%, and what happens if they leave?

Optimizing for cheap instead of profitable. A team celebrates because they cut CAC by 40%. Three months later, they discover that the cheap-to-acquire customers churn at twice the normal rate. Their LTV is half the average. The “saving” on CAC destroyed value downstream. The goal is not to find the cheapest customers. The goal is to find the ones who stay and pay.

Every one of these patterns looked correct on a dashboard. The problem was never the math. The problem was the inputs. Go beneath the averages. Track what customers actually do, not what your model assumes they will.

The next question every founder asks is: what does “good” actually look like for my business?

Definitions without benchmarks are a ruler with no markings.