| name | willingness-to-pay-research |
| description | Design a willingness-to-pay study — Van Westendorp PSM, Gabor-Granger, conjoint/MaxDiff, or a live price A/B test — choosing the method by the decision it serves and the data available, with sample-frame and bias guards. Reach for this before setting a price number, when defending a price, or when stated WTP and behavior disagree. Routes statistical significance to applied-statistics. |
Skill: Willingness-to-Pay Research
A price with no WTP evidence is a guess wearing a number. This skill picks the
research method, designs it with the right guards, and routes the significance read
to applied-statistics.
Step 0 — Two guards that apply to every survey method
- Stated WTP overstates real WTP. People say they'll pay more than they do.
Discount survey numbers toward any behavioral data you have.
- Ask the buyer, not just the user. In B2B the budget holder and the user are
often different; a study that surveys only users mis-reads the budget.
Step 1 — Pick the method by the decision
Run ../../knowledge/pricing-decision-trees.md §3:
- Need a rough acceptable range, early & cheap? → Van Westendorp PSM (the four
price-perception questions → acceptable range + too-cheap/too-expensive points).
Returns a range, not a price.
- Picking a number once you have a band? → Gabor-Granger (purchase intent at
set price points → demand + revenue curve).
- Question is about packaging / which features drive WTP? → Conjoint / MaxDiff
(choice-based feature-price tradeoffs).
- Have live traffic and can experiment? → Live price A/B test — observed
behavior beats every survey. Route significance to
applied-statistics.
Step 2 — Design the instrument
- Sample frame: who qualifies, how you reach them, and whether they represent the
target segment (not just whoever answers).
- Who to ask: buyers (budget) for the price decision; users for the value
perception. Often both, segmented.
- Bias guards: avoid anchoring (don't lead with your preferred price), randomize
price-point order, neutral framing.
- Sample size: size it for the precision you need — hand the power/sizing math to
applied-statistics rather than guessing.
Step 3 — Run / spec and read
- Van Westendorp → report the acceptable range + the optimal-price-point indicators
as a band and a sanity check, never as "the price."
- Gabor-Granger → plot demand and revenue curves; the revenue-maximizing point is a
candidate, not a mandate (consider strategy, not just the curve peak).
- Conjoint → translate part-worths into which feature belongs in which tier (feed
packaging-and-tiering).
- A/B test → report effect size + CI; route significance to
applied-statistics.
Step 4 — Discount toward behavior & bound the claim
Apply the Step-0 discount. State the confidence the evidence supports — a 40-response
Van Westendorp is a directional band, not a validated price.
Output
A study design (method + sample frame + who-to-ask + bias guards + sample size) or,
if already run, the WTP band/curve with its confidence bound and the behavioral
discount applied. Hand the actionable band to pricing-strategist; hand significance
to applied-statistics.