F7. Survey Development for Validating a Business Idea Step by Step
After this lesson you can
- Decide when a survey is the right tool and when interviews should come first
- Write single-idea, neutral, past-behavior questions with balanced options
- Design surveys that reduce satisficing and acquiescence
- Pretest a survey with think-aloud and verbal probing
- Explain measurement and representation error
- Treat stated willingness to pay as weak evidence
Video coming soon
Videos are planned to publish from December 2026.
A step-by-step lesson on building a survey that measures a known problem instead of collecting a friendly yes. It covers when to survey, question wording and order, balanced response options, satisficing, cognitive pretesting and total survey error.
Questions and answers
Questions a learner would ask about this lesson. Think through your own answer before you open each one.
- When should I use a survey instead of interviews?
Show answer
Interviews come first: they discover problems and customer language. Surveys then measure how common and how strong a known problem is, how people solve it today, and how segments you defined differ. A survey cannot find problems you have not discovered yet, show whether people will actually pay, or describe your whole market from a convenience sample.
- What makes a survey question strong?
Show answer
One idea per question, neutral wording, and a concrete time frame about past behavior. For example, "In the last 12 months, did you seek statistical help for a project?" is stronger than "Would you hire a statistician?" Response options should be balanced, should not overlap, and should include "None" or "Other" when needed.
- What is satisficing?
Show answer
Krosnick (1991) describes satisficing as giving an answer that seems good enough, instead of the best answer, to reduce the effort of responding. Shortcuts include choosing the first reasonable option, agreeing with statements, giving the same rating to every item, saying "don't know," and answering at random. The fix is design, such as short surveys, concrete wording and fewer long rating grids.
- How do I pretest a survey?
Show answer
Use cognitive interviews, which Willis (2005) describes as a pretest where people from the target group answer draft questions while an interviewer learns how they understood them. The lesson suggests recruiting three to five target respondents, using think-aloud plus verbal probes one person at a time, noting every misunderstanding, revising, and retesting changed questions.
- If most respondents say they would pay, is that evidence of demand?
Show answer
It is weak evidence. Saying yes to "Would you pay 600 dollars for this package?" costs nothing. Behavioral evidence, such as pre-orders or deposits with clear terms or signed letters of intent, is stronger and is tested later in the course's pricing lesson.
Practice quiz
- Question 1Satisficing, as described by Krosnick, means respondents do what?
- Question 2Cognitive interviewing is mainly used to do what?
- Question 3Founder B surveys their department mailing list and gets 80 percent yes. What is the main total survey error concern?
- Question 4What is wrong with "How often and how much do you pay for statistical help?"
- Question 5Which response scale is balanced?
- Question 6Why does the lesson prefer "How easy or difficult was it to find statistical help?" over "Statistical help is hard to find. Agree / Disagree"?
- Question 7According to the lesson, what order should survey questions follow?
- Question 8The total survey error framework splits error into which two sides?
- Question 9A founder doubles the number of responses from the same convenience sample. What does the lesson say?
- Question 10In Founder B's rewritten survey, which item was removed?
Saved in your browser only: no account, no server.
Entrepreneurship for Graduate Students is education, not legal, tax, financial, or immigration advice. Bring questions about your own situation to your university technology transfer office, a licensed professional, or the right campus office.