Most pricing frameworks assume you already know your customer's willingness to pay. In practice, product teams rarely do. You're entering a new market. Launching a feature no one has asked you to price before. Negotiating a B2B deal without benchmarks. The demand curve is invisible and the clock is ticking.
Pricing under uncertainty is the real operating condition for most product managers. This lesson gives you a structured approach for making defensible pricing decisions when data is thin.
Start with an anchor, not a number
Before you can set a price, you need an anchor — a reference point that frames what "reasonable" means for your buyer. Without an anchor, negotiation or adoption behavior becomes unpredictable.
Two anchoring techniques work in product contexts:
Internal anchor (cost + desired margin): Start with what it costs you to deliver the product or feature, add the margin you need to be sustainable, and use that as a floor. This is not your price — it is your walkaway point. Below this number, the deal is economically irrational regardless of strategic intent.
External anchor (comparable value): Find a proxy. What does your buyer currently spend to solve this problem manually? What does a competing or substitute solution cost? What is the measurable output your product enables — in time saved, revenue generated, or risk avoided — and what is that worth? This ceiling-of-value becomes your upper anchor.
Your actual price lives somewhere between floor and ceiling, calibrated by competitive pressure and how clearly your buyer can see the value.
Define your walkaway number before every pricing conversation. If you don't know it going in, you will almost certainly give ground you shouldn't.
Willingness-to-pay research: structured approaches
When you have time to research before pricing, three approaches are worth knowing:
Direct questioning (least reliable): Ask customers what they would pay. This produces optimistic answers because there is no real commitment. Use it only to triangulate, never as primary data.
Van Westendorp Price Sensitivity Meter (PSM): A four-question survey that maps a price acceptability range. You ask respondents: at what price would this feel too cheap to trust? Too expensive to consider? Expensive but worth considering? A good deal? The intersection of responses produces an "acceptable price range" and a "point of marginal cheapness." PSM is most useful for consumer or SMB products where you can survey a representative sample. It tells you about psychological range, not absolute willingness.
Conjoint analysis: Present buyers with bundles of features at different price points and ask them to choose. The statistical output reveals how much each feature contributes to perceived value, and what price thresholds cause drop-off. This is rigorous but slow — appropriate for major pricing architecture decisions, not rapid iteration.
When none of these are practical, use a proxy: interview five to seven existing customers and ask them to describe the last time they bought something that solved a similar problem. The price they paid and the mental model they used to evaluate it tells you more than a direct pricing question.
Pricing experiments: what works, what backfires
If you have enough traffic or pipeline to run experiments, pricing tests are powerful. But the design matters.
What works:
- Testing price tiers against each other (e.g., ₹999/mo vs. ₹1,499/mo for a SaaS product) on cold traffic where buyers have no prior anchor.
- Gradual price increases to existing cohorts, with clear communication of the new value being added.
- Geographically isolated tests where markets don't talk to each other (applicable in multi-city India rollouts).
What backfires:
- Showing different prices to different users of the same product in the same market simultaneously. If buyers compare notes — and they will — the trust damage outweighs any signal you collected.
- Testing price points so far apart that the result tells you nothing about the real optimum.
- Running a test without a clear hypothesis. "Let's see what happens at ₹2,000" is not an experiment.
The safest framing: treat early pricing as a hypothesis with a stated falsification condition. "We believe ₹1,200/month will achieve 15% trial-to-paid conversion in the first 90 days. If conversion is below 10%, we will revisit." This forces you to define success criteria before results are visible, which protects the decision from motivated reasoning.
The case for skim-then-cut
When you genuinely don't know where the market will settle, skim-then-cut is the lower-risk default strategy for a product with any differentiation.
The logic: starting high and cutting is easier to manage than starting low and raising. Price cuts are welcomed; price increases erode trust and trigger churn. A high launch price also signals quality and filters for buyers who are most committed — giving you better early customer data than a low-price flood of marginally interested users.
The practical constraint: skim-then-cut only works if you have a credible path to cutting. If your cost structure or brand positioning makes it impossible to ever serve a lower-price segment, you are not skimming — you are simply choosing a niche. That can be a valid strategy, but it should be a conscious choice, not a default.
For most new B2B products without established price benchmarks, starting at the high end of your defensible range and cutting after 6–12 months of learning is a more recoverable position than starting too low and trying to raise prices on an already-anchored customer base.
Three questions to ask before you set a price
Before finalizing any pricing decision under uncertainty, pressure-test it against these three questions:
-
What is our walkaway number, and is this price above it? If you cannot define the minimum economically viable price, you are not ready to price.
-
What does the buyer believe they are comparing this to? Pricing does not exist in isolation — buyers always have a reference point, even if you don't know what it is. Try to surface the anchor they are using before the conversation happens.
-
If we need to adjust in six months, which direction is easier to move? If raising would be catastrophic and cutting would be trivial, start higher. If you are in a market where being caught raising prices destroys the relationship, start lower but model the path to profitability explicitly before you do.
Pricing under uncertainty is not about getting it right the first time. It is about making a defensible decision with the information you have, building in a review mechanism, and moving quickly enough that you learn before the market moves.