Academics · Study Skills

Ask customers
the right questions.

A survey is only useful if it helps you decide something. Start with the decision, add each question with its answer type and the reason you are asking, and the generator flags leading wording, two questions in one and “would you buy” traps, estimates how long it takes and how precise your sample can be, then prints the survey ready to hand out.

Write a surveyWhat you get

Surveys

The worked example is made up: a popsicle stand at a farmers market, a phone case resale idea whose questions need fixing, and a homework help app with a large sample. Saved on this device only.

Loading the survey form. If nothing appears, turn on JavaScript.

What you get

Academics · Study SkillsFree

What you finish with
A record per survey: the idea, the customer and the decision; how many answers and how you will reach people; each question with its answer type, choices and reason; five checks. Estimates the minutes to answer and the margin of error, flags biased or risky questions, and prints the survey as respondents see it.
Who it is for
Middle and high school students running a business project, a market day stand or an entrepreneurship class, and the teachers who guide them.
Format
Browser survey builder with a saved record per survey. Prints the survey itself, the full record, every survey in a table or a blank form; downloads a spreadsheet (.csv); exports and imports a file.
Access
Free. No account, no sign-in and nothing to install. Everything you type stays in this browser; nothing is sent to Caveat.
Where it lives
The Study Skills space in Caveat Academics.
Support
Questions about access, an order or a correction: team@caveatacademic.com.

How to use it

  1. Write the decision first.
  2. Plan who you will ask and how many.
  3. Add questions with a reason for each.
  4. Fix what the flags point out.
  5. Try it on three people, then print it.

Write a survey

How market surveys go wrong, and right

Fixing four common question problemsProblemBeforeBetterLeading“Don’t you love spicy snacks?”“How do you feel about spicy snacks?”Two in one“Is it cheap and tasty?”Ask price and taste separatelyFuture buying“Would you buy this?”“What did you buy last time?”Vague time“Do you snack often?”“How many times last week?”Fixing four common question problemsLeadingBefore: “Don’t you love spicy snacks?”Better: “How do you feel about spicysnacks?”Two in oneBefore: “Is it cheap and tasty?”Better: Ask price and taste separatelyFuture buyingBefore: “Would you buy this?”Better: “What did you buy last time?”Vague timeBefore: “Do you snack often?”Better: “How many times last week?”
Ask about what people did, not what they imagine they would do.

Margin of error, plainly

With 100 answers from a random group of customers, a result like “40 percent prefer mango” is good to within about 10 points either way. The figure is 1.96 × √(0.25 ÷ n). It shrinks slowly: four times the answers only halves it. And it means little if you asked only friends.

Who you ask matters more

Fifty answers from people who actually shop at the market tell you more than 200 from classmates who never go. Ask where your customers already are.

Behavior beats intention

People are polite about new ideas, so “would you buy this?” gets more yes answers than real sales. Questions about what people did last week or pay now are more honest guides.

Sources

  1. Dillman, D. A., Smyth, J. D., & Christian, L. M. (2014). Internet, Phone, Mail, and Mixed-Mode Surveys: The Tailored Design Method (4th ed.). Wiley.
  2. Pew Research Center. Questionnaire design (methods guide on question wording and order).
  3. Fitzpatrick, R. (2013). The Mom Test: How to Talk to Customers and Learn If Your Business Is a Good Idea When Everyone Is Lying to You.
  4. Margin of error for a proportion at 95 percent confidence, worst case p = 0.5: 1.96 × √(0.25 / n); checked with Wolfram|Alpha for n = 60.

The worked example is made up. The margin of error assumes a simple random sample of your customers; answers from friends, classmates or a posted link do not meet that, so treat it as the best case. Timing is a rough guide. Nothing you type is sent to Caveat. Support: team@caveatacademic.com.