A survey cannot prove your idea will work. It can show whether the problem is real, how people cope today, and whether your answer interests them. The questions matter, and so does who answers them.
1. Validate the problem before the solution
Ask about the problem before you describe your product. Two template publishers we read on 2 October 2026 take this approach. Responsly advises testing how often and how badly the problem occurs before asking about features, and using neutral wording. Formbricks structures its template around problem severity, current solutions, interest, willingness to pay and concerns. Both may have changed since.
2. A question set you can copy
Adapt these to your audience.
- In the last three months, how often did you run into [problem]? (never, once, a few times, weekly or more)
- How much of a problem is it? (scale of 1 to 5)
- What do you do about it today? (open text)
- What do you dislike about that? (open text)
- Have you looked for a better option in the past year? (yes or no, then what you tried)
- How interested would you be in a solution that [one neutral sentence]? (scale of 1 to 5)
- What would stop you from using it? (open text)
- Roughly what would you expect to pay for it? (ranges, plus "I would not pay")
- Which describes you? (two or three screener options that confirm you are asking the right people)
Put screeners first. Do not reveal which answer qualifies. Ask about past behavior, not predictions.
3. Stated interest is not purchase
Formbricks advises discounting what people say they would do against what they do. Treat strong interest as a reason to test further. Entrepreneur published a 2014 article, so it is older, suggesting landing pages and pre-orders as stronger signals of commitment. Use the survey to decide what to build or test next, not to declare the idea proven.
4. Who answers matters
- Your own list or link. Cheap and fast, but it reaches people who already know you and may like you.
- A defined audience you do not already reach. You need a recruited sample. Our guide to finding respondents for a specific target audience covers the routes.
For how many answers you need, see how many survey respondents you need to validate a startup idea. For price context, see what survey responses cost.
5. Run it with Chorus Research
Chorus Research is a self-service way to reach real respondents. Use it in the app, through the REST API, or from an AI assistant through the hosted MCP server.
- US general population: $1.00 per completed response, $5 minimum. Fifty completed responses is $50 at that price.
- Custom audiences: from $1.00 per completed response, quoted before purchase. The estimate shown in the product is the source of truth.
- Estimate first: the MCP tools
list_audiences,create_survey,estimate_costandestimate_launchset up and price a survey, andget_survey_statusfollows it. The hosted MCP endpoint never charges. It returns a browser checkout link.
Docs: getting started, REST API, MCP, FAQ, refunds. Chorus Research is operated by Collective Measure, Inc.
Third-party points are paraphrased from public pages by Responsly, Formbricks and Entrepreneur as read on 2 October 2026 and may have changed. Chorus Research prices are the public prices as of 30 September 2026.
More from the blog
- What Is Agentic Market Research? A Plain-English Guide
- MCP Servers for Market Research: What Actually Exists in 2026
- How to Recruit Survey Respondents for a Startup: Cost, Speed, Quality and API Launch
- How to Get Real Human Survey Responses Programmatically (API and MCP)
- How much does it cost to get survey responses?
- How many survey respondents do you need to validate a startup idea, and what will they cost?
- Survey respondent pricing compared: how three vendors publish it
- How to find survey respondents for a specific target audience
- Customer discovery tools: where paid survey respondents fit