Software Growth

Willingness to pay

Willingness to pay is the most a customer will pay for your product. Estimate it with interviews, Van Westendorp, conjoint analysis and price tests.

Willingness to pay (WTP) is the maximum price a customer would accept for your product before choosing something else or nothing. It differs by person, company size and use case, which is why one price on a page leaves money behind or loses buyers. Researching it is how you replace a guessed price with a defensible one.

Paddle's guide to WTP is blunt about the limits: humans are not entirely rational, so there is no exact formula, and WTP varies by segment and changes over time (source). Treat any method as producing a range, not a number.

Four ways to estimate it

1. Customer interviews

The cheapest method and a good start. Ask about the problem and what it costs them now, who approves spending, and what budget sits behind it. Jason Cohen recommends raising pricing in customer development interviews, including budget expectations, approval chains and ROI (source). Do not ask "would you pay $X?" Listen to what they already pay for the alternatives.

2. Van Westendorp price sensitivity meter

A short survey with four questions about a product: at what price is it so cheap you doubt the quality, a bargain, getting expensive but still considered, and too expensive to consider. You plot the cumulative answers and read the acceptable range from where the lines cross: the lower bound where "too cheap" meets "expensive", the upper bound where "cheap" meets "too expensive", and an optimal point where "too cheap" meets "too expensive". Limits: it prices one product in isolation, respondents tend to under-state prices when asked directly, and the "optimal" price does not necessarily maximize revenue (source). It needs a few dozen to a few hundred responses to be useful.

3. Conjoint analysis

Respondents choose between product bundles that vary in features and price, and the analysis infers how much each feature is worth. It is the right tool for packaging decisions, such as which features belong in which tier, but it needs a sample and tooling that most early SaaS teams lack.

4. Live price tests

The most honest method, because people spend real money. Show different prices to different new visitors, or offer different prices to different email segments, and compare. McKenzie recommends testing on the marketing site or on an email list, and notes that offering identical features at different prices measures willingness to pay and not feature demand (source).

Turning results into a price

For a live test, compare revenue per visitor and not signups alone.

At $29, 1,000 visitors produce 40 customers: $1.16 per visitor. At $49, the same 1,000 produce 30 customers: $1.47 per visitor. The higher price wins despite fewer buyers. See price elasticity for how to read that tradeoff.

Mistakes

  • Asking only your fans. Current customers already accepted your price.
  • Averaging across segments. A hobbyist and an agency should not be averaged into one price.
  • Treating stated WTP as behavior. What people say and what they pay differ.

For a small SaaS, start with five to ten interviews and a price test. Move to a survey method only if you have an audience large enough to fill it.

Sources

Back to the SaaS glossary