How much will people actually pay for a new service, a cleaner water supply, or an improved product? Guessing is expensive. A willingness-to-pay (WTP) survey measures it directly — and it is one of the most valuable tools in pricing, policy, and project design.
What a willingness-to-pay survey does
A WTP survey estimates the maximum amount people are willing to pay for a good, service, or improvement. It is widely used for:
- Pricing new products and services
- Utility and infrastructure planning (water, sanitation, electricity)
- Environmental valuation — pricing non-market goods like clean air or conservation
- Public services — gauging demand and affordability
- Development projects — designing cost-recovery and tariffs
Common WTP methods
Contingent valuation
Respondents are described a scenario and asked directly what they would pay. Question formats include:
- Open-ended — "What is the most you would pay?"
- Bidding game — iterative "would you pay X? Y?" until a threshold
- Payment card — choose from a list of amounts
- Dichotomous choice — "Would you pay X? Yes/No" (statistically robust)
Discrete choice / conjoint analysis
Respondents choose between bundles of features at different prices. Their choices reveal how much each feature is worth — a more indirect but often more reliable approach.
Designing a good WTP survey
- Describe the good clearly — respondents must understand exactly what they are valuing.
- Make the scenario realistic — including a believable payment mechanism.
- Choose the right elicitation format — dichotomous choice reduces certain biases.
- Pilot thoroughly — WTP wording is sensitive; small changes shift answers.
- Sample representatively — WTP varies hugely by income and context.
The biases to watch
WTP surveys are notoriously prone to bias:
- Hypothetical bias — people overstate what they would really pay
- Strategic bias — answering to influence an outcome (e.g. understating to keep prices low)
- Anchoring — starting bids influence the final answer
Good design — realistic scenarios, careful formats, and follow-up questions — keeps these in check.
Designing the scenario, which is most of the work
A willingness-to-pay survey lives or dies on the scenario — the description of what the respondent is being asked to pay for. Vague scenarios produce vague numbers that cannot be used for pricing anything.
A usable scenario specifies six things:
| Element | Weak | Usable |
|---|---|---|
| The good | "Better water supply" | "A metered piped connection to your house" |
| The quality | Unstated | "Six hours of supply daily, tested monthly" |
| The provider | Unstated | "Operated by the gram panchayat" |
| The payment vehicle | "Would you pay?" | "A monthly bill, payable at the panchayat office" |
| The timing | Unstated | "Starting in six months, ongoing" |
| The alternative | Unstated | "If not, the current handpump remains" |
The payment vehicle matters more than people expect. Respondents answer differently to a monthly bill, a one-time connection charge, a tax increase and a voluntary contribution — sometimes by a factor of two. Choose the vehicle that matches how payment would genuinely be collected, because that is the number you can act on.
The alternative — what happens if they say no — is the other frequently missing piece. Without it, respondents cannot evaluate the offer, and many will answer as though refusal costs them nothing.
Choosing the elicitation format
How you ask determines the shape of the answer and the analysis that follows.
| Format | How it is asked | Analysis | Main weakness |
|---|---|---|---|
| Open-ended | "What is the most you would pay?" | Direct mean and median | High don't-know rates; unrealistic answers |
| Payment card | "Which of these amounts?" | Interval regression | The shown range anchors the answer |
| Bidding game | Iterate up or down from a start | Interval regression | Strong starting-point bias |
| Single-bounded dichotomous | "Would you pay 200 rupees?" | Probit or logit on the bid | Needs a large sample |
| Double-bounded dichotomous | A follow-up bid after the first answer | Interval regression | More efficient; some anchoring |
| Discrete choice experiment | Choose between bundled options | Conditional logit | Cognitively demanding |
For field surveys with mixed literacy, double-bounded dichotomous choice is usually the best compromise. It mimics a real take-it-or-leave-it price decision, which respondents find natural, and the follow-up question extracts substantially more information than a single bid. The cost is a more involved analysis and some sensitivity to the starting bid.
Open-ended questions are the most tempting and the least reliable. Asked to name a maximum price for something they have never purchased, respondents guess, anchor on whatever number is salient, or decline to answer.
Setting the bid vector
For dichotomous formats, the set of bid amounts you offer determines the precision of the whole study. Bids are assigned randomly across respondents.
| Step | Method |
|---|---|
| Estimate the plausible range | From the pilot, or from existing prices for comparable goods |
| Choose 5 to 8 bid levels | Spread across that range |
| Concentrate levels near the expected median | Where the response curve is steepest |
| Randomise assignment | Each respondent gets one starting bid |
| Test the extremes in the pilot | If nobody accepts the highest bid, the range is wrong |
The pilot has a specific job here: an open-ended WTP question on 40 respondents, used solely to set the bid vector for the main survey. Getting the range wrong is the most common technical failure — if every respondent accepts the highest bid, you learn only that true WTP is somewhere above it.
Analysing WTP data
The output should be more than an average. Three things belong in every WTP report:
The demand curve. Plot the share accepting against the bid amount. This is the most useful single output, because it answers the question a decision-maker actually has: at price X, what share of households take it up?
Mean and median WTP. The median is generally more robust; the mean is influenced by a small number of very high responses and is more relevant when aggregate revenue is the question.
WTP by subgroup. By income quintile, landholding, household size and location. A single average conceals whether a proposed tariff is affordable for the poorest, which is usually the actual policy question.
| Output | Decision it supports |
|---|---|
| Demand curve | Pricing at a target uptake level |
| Median WTP | Whether the scheme is viable at all |
| Mean WTP × households | Aggregate revenue potential |
| WTP by income quintile | Whether cross-subsidy or a lifeline tariff is needed |
| Share with zero WTP, and why | Whether refusal is affordability or objection |
Separate genuine zeros from protest zeros. A household that cannot afford the service is a real zero. A household refusing because it believes the government should provide it free is registering an objection, not a valuation, and including it as a zero biases the mean downward. Ask a follow-up on the reason for any zero response.
Testing whether the answers mean anything
Stated WTP is hypothetical, and hypothetical answers exceed real payments — commonly by 20 to 50%. Three internal checks belong in the design:
A scope test. Offer a larger version of the good to a random subsample. WTP should rise. If respondents state the same amount for six hours of water and twelve, they are not valuing the good and the study has failed.
A budget reminder. Explicitly remind respondents of their limited income before eliciting. This reduces overstatement measurably and costs one sentence.
A certainty follow-up. After a "yes", ask how certain they are on a 1 to 10 scale. Treating anything below 7 or 8 as a "no" brings hypothetical results substantially closer to observed behaviour.
For the underlying method taxonomy and the theory behind each approach, see contingent valuation and other willingness-to-pay methods. For the sampling and field procedures around it, how to conduct a socio-economic survey step by step and how to reduce bias in field surveys. Our socio-economic and baseline survey service designs and fields WTP studies. The NOAA panel guidance on contingent valuation remains the reference standard for scenario design.
The takeaway
A well-run willingness-to-pay survey replaces guesswork with evidence, telling you what the market or community will genuinely bear. But WTP is only as good as its design: describe the good clearly, pick the right format, pilot hard, and treat the numbers with appropriate caution.
Frequently Asked Questions
How do you conduct a willingness-to-pay survey?
Define a specific scenario covering the good, its quality, the provider, the payment vehicle, the timing and the alternative. Pilot an open-ended question to set your bid range, then field a dichotomous-choice format with randomly assigned bids, and analyse as a demand curve rather than a single average.
What is the best question format for WTP surveys?
Double-bounded dichotomous choice for most field surveys. It mirrors a real take-it-or-leave-it price decision, which respondents handle naturally, and the follow-up bid extracts considerably more information than a single question.
Why should I avoid open-ended WTP questions?
Because respondents asked to name a maximum price for something they have never bought tend to guess, anchor on a salient number, or decline to answer. Don't-know rates are high and the resulting distribution is unstable.
What is a protest zero in a WTP survey?
A zero response driven by objection rather than inability to pay — for example, a belief that the government should provide the service free. Including protest zeros as genuine valuations biases the mean downward, so always ask why a respondent answered zero.
Do people actually pay what they say they would?
Usually less. Hypothetical WTP typically exceeds real payment by 20 to 50%. Budget reminders, certainty follow-up questions and treating low-certainty acceptances as refusals all narrow the gap.
What should a WTP survey report include?
A demand curve showing uptake at each price, median and mean WTP, breakdowns by income quintile and other relevant subgroups, the share of zero responses with reasons, and the results of a scope test demonstrating that respondents valued the good rather than answering arbitrarily.
How DSTechnoverse Runs Your Survey
DSTechnoverse is a full-service socio-economic and baseline survey agency based in Indore. We do not just advise — we run the whole study end to end and hand you evidence you can defend:
- Design — objectives, indicators, and sampling (random, stratified, cluster, or multi-stage), with a piloted questionnaire translated into local languages.
- Field data collection — trained enumerators using CAPI (tablet) tools such as KoBoToolbox, ODK, and SurveyCTO-style apps, with GPS stamping, audio audits, and built-in skip-logic validation.
- Quality control — back-checks, spot-checks, and daily high-frequency data monitoring so problems are caught in the field, not after.
- Analysis — SPSS, R, Python, and Excel for indicators, cross-tabulations, and impact estimates, including baseline–endline difference-in-differences.
- A detailed report — cleaned datasets, indicator tables, charts, GIS maps, and clear interpretation, with interactive Power BI / Tableau dashboards on request.
Our survey experience in Indore includes Smart City citizen-feedback surveys, Startup India & MSME surveys, Swachh Survekshan (Swachh Bharat) cleanliness surveys, and skill-development impact evaluations — delivered for government, municipal, corporate, and NGO clients.
Planning a socio-economic, baseline, endline, or willingness-to-pay survey? Talk to our survey team for a scope and quote.