Payback beyond the blended average CAC

Blended acquisition costs can hide which customers actually generate sustainable returns. Compare contribution and acquisition costs across segments and channels, while keeping observed results separate from uncertain lifetime estimates.

Build & Validate
Venture FinanceDistribution

By the EWOR Team3 min read

Your customer acquisition cost looks good. Your lifetime value estimate looks even better. So growth must be working, right?

Not necessarily. A blended average can hide customers who require expensive onboarding, generate little contribution or leave after a few months. It can also make an inefficient acquisition channel look better than it is.

To understand whether your growth is sustainable, look beyond the averages and examine how acquisition costs and customer economics vary across segments and channels.

Understand who you acquire and how you reach them

Customer-segment economics tell you whether particular customers generate enough value to justify the cost of serving them. Channel economics tell you how much it costs to acquire customers through a particular route. These questions are related, but they are not interchangeable.

Imagine your product serves both self-service teams and organisations that need extensive implementation. Paid search and partner referrals might each bring in both groups. Comparing channels alone could hide the cost of serving different customers. Comparing customer segments alone could hide an expensive way of reaching them.

Break down the numbers where meaningful differences exist, rather than creating more segments just to make the analysis look sophisticated. And keep small samples visible: a handful of customers is rarely enough to establish a reliable pattern.

Count the costs that growth creates

For each customer cohort, track revenue, direct delivery costs, onboarding and support effort, retention and attributable acquisition costs. Where possible, examine the customer segment and acquisition channel together. Use consistent cost definitions and avoid counting the same expense twice.

Advertising spend is not the whole cost of acquisition. Sales time and other marketing costs matter, too. Track unpaid founder sales effort separately so a low cash cost does not disguise a process that depends on hours of work and may be difficult to scale.

Timing matters as well. If customers take months to move through your sales process, this month's spending may relate to customers acquired later. Dividing it by this month's new customers can give a misleading picture. Explain how you allocate shared costs and attribute customers, especially when the evidence is uncertain. A precise spreadsheet does not guarantee an accurate conclusion.

Separate what you know from what you estimate

For newer customer cohorts, start with observed contribution over a defined period: revenue minus the direct costs of delivery. Show acquisition spending alongside it. Do not assume a few months of data tell you how much a customer will be worth over their lifetime.

Lifetime value depends on what happens in the future: whether customers stay, how much they spend and what it costs to serve them. Simple churn-based estimates can be misleading when cohorts are young or customer behaviour is changing. Use scenarios, make your assumptions explicit and keep estimates separate from observed results.

There is no universal LTV threshold that proves a business is healthy. Payback period and available cash matter, too. A customer can look profitable over several years on paper while taking too long to generate the cash you need today.

Let the evidence guide your next move

Once you understand the differences, you can investigate the underlying problem. Are certain customers expensive because onboarding takes too much time? Does one channel attract customers who churn quickly? Is the acquisition process itself too costly?

If a channel attracts very different customer types, its blended acquisition cost may not tell you much about how efficiently it reaches your best-fit customers. If retention data is still immature, focus on observed contribution and be honest about what you cannot yet estimate.

Use what you learn to test one change at a time. The analysis can steer your distribution decisions, but it will not automatically identify the best channel. Positive contribution still needs to cover acquisition and fixed costs, and a promising lifetime value estimate cannot pay today's bills.

Get the next signal before everyone else.

Frameworks and contrarian theses for outlier founders, straight to your inbox. No fluff, no recycled advice, weekly.

The EWOR Signals newsletter