How to find survey respondents for a specific target audience

By Chorus Research, operated by Collective Measure, Inc. · 2 October 2026

You know who you need to hear from: parents of toddlers in one region, people who switched banks this year, developers who build with AI assistants. Writing the questions is the easy part. Getting those specific people to answer them is harder. This guide covers it in order: define the audience, choose a route, pick a sample size, and run a small first launch.

Define the audience before you pick a channel

Write down three things.

Criteria. What must be true of someone for their answer to count? Keep it to the two or three attributes that really change your result. Each extra criterion shrinks the pool of people who qualify.

Screening questions. Give each criterion one question, asked before the survey proper, so people who don't qualify are screened out instead of counted. Word the screeners so they don't reveal which answer qualifies.

Why this comes first. The more specific the audience, the more work it is for any supplier to find it. One guide we read says panel providers price according to how complex your criteria are (GrapeData, read 1 October 2026). A broad audience is simpler to define and source than a narrow one.

Route A: your own channels

Your own email list, social accounts, website or app, and personal networks are the usual first places to look (SurveyMonkey and GrapeData, read 1 October 2026). Local organisations and online communities where your audience already gathers are other options to consider.

• Fits when people matching your criteria already follow you or gather somewhere you can reach.
• Trade-off: it costs little money, but you can't guarantee how many answers you'll get or who they come from. People who already follow you may lean toward liking you. Screen them anyway.

Route B: paid panels and recruiters

When your audience isn't already around you, you can pay a panel or recruiter to find people who match your criteria (SurveyMonkey and GrapeData, read 1 October 2026). Panel providers typically price by how hard the audience is to reach (GrapeData, read 1 October 2026).

• Fits when your criteria are narrow and you need a defined number of completed responses by a certain date.
• Trade-off: you pay per respondent or per project, and you may have to ask for a quote before you know the price. Ask for the price per completed response and any minimum order in writing.

How many responses you need

Work out a minimum sample size before choosing a channel. One top-ranked guide treats 95% confidence as the usual standard when computing it (SurveyMonkey, read 1 October 2026). If you are testing an idea and not publishing a statistic, you may need fewer responses than you fear. Our companion article on how many survey respondents you need to validate a startup idea goes through it, and the one on what survey responses cost covers price.

The Chorus Research path

Chorus Research is a self-service way to reach real respondents. You can use it from the app, the REST API, or an AI assistant through the hosted MCP server. The prices below are the public prices published on the homepage.

• US general population: $1.00 per completed response, with a $5 minimum. Fifty completed responses is $50 at that price.
• Custom audiences: from $1.00 per completed response, quoted before purchase. Narrower criteria can change the quote, so check the estimate for your audience.
• Estimate first: the MCP tools list_audiences, create_survey, estimate_cost and estimate_launch follow the steps above, and get_survey_status follows fielding. The hosted MCP endpoint never charges itself. It returns a browser checkout link for you to complete.

Docs: getting started, REST API, MCP, FAQ, refunds. Chorus Research is operated by Collective Measure, Inc. The estimate shown in the product is the source of truth for any price.

A checklist for a small first launch

1. Write the audience in one sentence, with at most three criteria and a screener for each.
2. Choose a sample size for the decision you're making, not for publication.
3. If a list or community of yours plainly matches the criteria, try it first and count what comes back.
4. Get a price before you commit: run the estimate, or ask a vendor for the price per completed response and the minimum in writing.
5. Start small. A first launch at the $5 minimum (five US general-population completes) shows whether your screener works and the answers are usable.
6. Read the open-text answers as well as the percentages.
7. Scale only what worked, and keep the audience definition fixed so results stay comparable.

Sources

Third-party points are paraphrased from public guides by SurveyMonkey and GrapeData as read on 1 October 2026 and may have changed. Chorus Research prices are the public prices as of 30 September 2026.