“I would buy that” is an encouraging sentence, but the circumstances of a research session are unusually forgiving. The product is in front of the participant, the explanation is complete and no competing purchase is demanding their money at that moment. Purchase intent records the stated intention under those conditions.
It is commonly measured with a question about likelihood of buying and a response scale running from definitely would to definitely would not. Researchers may report the percentage choosing the highest response, or the two highest responses together. Those summaries need their labels, question wording and audience definition to remain meaningful.
Make the imagined purchase concrete
Specify the offer, price, relevant availability and time period. Buying a service “sometime” is a different proposition from subscribing next month at an annual commitment. Ipsos’ discussion of purchase intent also highlights the influence of category participation: a sample containing many people who rarely buy the category can produce a different result from a study of active buyers.
This does not mean researchers should recruit only the most enthusiastic audience. The sample should match the decision. If the aim is to assess a broad market opportunity, excluding less frequent buyers may create an equally misleading picture.
In an illustrative study of a meal service, high intent could depend on participants assuming that delivery is included. If the actual offer charges separately, the finding concerns a different proposition. Concept testing should establish comprehension before the team interprets willingness to buy.
Use intent as one piece of evidence
Ask what makes the offer relevant, what would prevent a purchase and what it would replace. These explanations can reveal conditions hidden by an overall score. Someone may like the service but need another household member to agree, or face a purchasing cycle that makes immediate adoption unlikely.
Comparisons are more useful when the concepts, samples and study conditions are comparable. A difference between scores from unrelated studies may reflect the measurement as much as the offers. Avoid applying a universal conversion factor that turns a stated percentage into expected sales without evidence that the model fits the category and setting.
Where feasible, examine later behaviour through a trial, a real offer or follow-up research. Actual purchasing still has context, including availability and promotion, but it can test assumptions made in the survey. Willingness-to-pay research adds another perspective when price is uncertain.
Purchase intent helps describe the response to an offer. A useful report preserves that distinction, explaining what people said they would do and what remains to be established about the decision they will eventually face.