How do I find and pay people to take a survey online?

Find Survey Participants and Pay for Responses

Last reviewed 2026-09-18

Short answer

To find people to take a survey, post a multi-seat bounty on RentAHuman that describes the survey, the eligibility criteria, the time it takes, and the pay per completed response. Workers apply, you accept those who pass screening, they complete the survey in your own survey tool and submit the unique completion token it shows, and you release payment from escrow. RentAHuman supplies the respondents and the payment rail. The questionnaire, unique completion tokens, attention checks, and duplicate-response checks live in your survey tool and your review. This is hiring people to answer a survey; hiring someone to design or analyze one is a different, single-seat bounty.

Best for

Short online surveys that need a set number of completed responses from screened respondents
Respondents in a particular country or city, or people who own a device or use a product you name
Pilot surveys used to test wording and timing before a larger fielding
Repeated survey waves posted by a team or an AI agent through the REST API or MCP

Limits and responsibilities

Respondents are self-selected marketplace workers, not a quota-balanced panel. Report results as a convenience sample.
RentAHuman does not host the survey, generate completion tokens, or detect duplicate or low-effort responses. Your survey tool and your review do that.
Screening is self-reported unless you require verification. People who want the pay have an incentive to claim eligibility, so use knockout questions and consistency checks, and disclose the screening and pay terms before anyone applies.
Each marketplace account can hold one application and one seat per bounty. That does not prove unique identity or rule out duplicate accounts, so treat it as a convenience rather than an anti-fraud control. To keep accounts that took an earlier wave out of a new one, exclude your earlier bounties through the API or MCP.
Responses arrive as accepted applicants complete the survey, so fill time depends on criteria, pay, and time of day.

How the workflow works

01
Build the survey in your own tool, add an attention check and a final screen that shows a completion token unique to each response, and time it with a few test runs.
02
Post a bounty with one seat per response you need, the pay per completed survey, eligibility questions, and a completion window so abandoned seats reopen.
03
Accept applicants who pass screening, send each one a personal survey link or invitation ID through the bounty conversation, and require the completion token as the submission.
04
Match each token to one response in your tool, review duplicates and failed attention checks against the criteria you disclosed, release payment for completions that met them, and message the rest with the reason.

Example tasks

Product feedback survey

Collect 50 responses from people who bought a named category of product in the last three months, paid per completed response.

Country-specific pricing survey

Run a 10-minute pricing survey limited to respondents in one country and accept only submissions whose unique completion token matches one response.

Survey pilot

Field the draft with 10 people, pay for the pilot, and use their completion times and comments to fix the instrument before the main wave.

Practical guide

Respondents versus a survey designer

This page is about hiring people to answer a survey you already have. A survey designer, programmer, or analyst is a different hire: post a single-seat bounty that describes the deliverable, the tool, and the review steps. Keep the two roles in separate bounties, because the pay, screening, and completion evidence differ.

Hypothetical budget

Every number here is illustrative. Suppose you need 100 completed responses to a 10-minute survey and decide to pay $3 per completion, which is the minimum per-seat price.

Respondent pay: 100 × $3 = $300. The $10,000 per-bounty cap applies to this pool before the fee.
Standard platform fee, 18% added on top of respondent pay: $54. Some accounts have fee exemptions.
Total funding when you post: $354.
Rejected or abandoned responses: budget 10 to 15 extra seats, or reopen seats as they expire.
Pilot wave first: 10 seats × $3 = $30, plus the standard 18% fee, is $35.40 total.
If the pilot shows the survey takes 20 minutes, double the per-seat pay rather than hoping people finish for less.

Completion-code workflow

RentAHuman does not host questionnaires, so the survey runs in your own tool. The link between a response and a paid seat is a completion token unique to that response. A shared code that every respondent sees cannot prove a unique completion.

Send each accepted participant a personal survey link or invitation ID through the bounty conversation.
Have the survey tool generate a completion token unique to each response and show it only on the final screen.
Require that token as the bounty submission, and match it to one response and one invitation in your tool.
Keep the mapping between invitation IDs and marketplace accounts in a separate file rather than asking for usernames inside the survey.
Approve a seat only after its token matches one complete response that met the criteria you disclosed.

Screening criteria that hold up

Write criteria as facts a person can answer yes or no, and put the knockout question first.
Use country or city targeting on the bounty for geography instead of asking.
Repeat one screening question inside the survey and drop inconsistent answers.
Leave automatic review on for clear matches and mismatches, and review uncertain applications yourself.

Fraud and duplicate checks you must run

One account can hold one seat per bounty, but that does not verify identity or prevent duplicate accounts. Response quality and duplicate detection are your call, and the checks below are review signals rather than automatic proof of fraud.

Duplicates: one unique completion token per response, matched to one invitation and one seat. Repeated device, network, or free-text patterns are reasons to look closer, not proof.
Speeders: flag completion times below about half of your pilot median for review.
Attention: include one instructed-response item and one open-ended question, and review a single failure in context instead of rejecting on it alone.
Straight-lining: flag identical answers across grid questions as a review signal.
Repeat participants: exclude earlier survey bounties when posting a new wave through the API or MCP, knowing this works at the account level only.
Rejections: disclose the criteria in the brief before anyone applies, message the reason, keep records, and release payment to everyone who met the agreed terms, including any withdrawal or screen-out pay you promised.

RentAHuman vs Prolific vs CloudResearch vs SurveyMonkey Audience for survey respondents

This is a fit comparison based on each provider’s published documentation, not a ranking. Amazon Mechanical Turk, which many teams used for survey respondents, closes on September 30, 2026, so it is not listed as an option.

Service	Best for	Consider
RentAHuman	Screened respondents paid per completed response, with country or city targeting, on the same marketplace used for other human tasks and reachable through REST or MCP.	No built-in survey hosting, demographic quotas, or duplicate detection. You bring the survey tool and run the quality checks.
Prolific	Academic studies, surveys, and human data with a participant pool built for research.	Optimized for structured online studies rather than open-ended operational or in-person tasks.
CloudResearch Connect	Verified research participants and longitudinal studies.	Designed around research and human-data workflows, not general local or physical-world tasks.
SurveyMonkey Audience	Packaged consumer and market-research panels delivered inside a survey platform.	Focused on survey responses rather than flexible human task execution.

Survey participant FAQ

Where can I find people to take my survey?

Post a multi-seat bounty on RentAHuman that names the topic, eligibility, time required, and pay per completed response. Workers apply, you accept those who pass screening, and they complete the survey in your own tool and submit the unique completion token it shows.

How much does it cost to pay survey respondents?

You set the pay. As a hypothetical example, 100 completed 10-minute surveys at $3 each is $300 in respondent pay. The standard 18% platform fee adds $54, so total funding is $354. Pay for time at a rate you would accept yourself, and budget extra seats for responses you reject.

Does RentAHuman host the survey?

No. Build the survey in any tool that can issue a completion token unique to each response on its final screen, then send accepted participants their personal link. RentAHuman handles recruitment, screening questions, messaging, and escrow payment.

How do I stop duplicate or low-quality responses?

Give each accepted participant a personal link or invitation ID, require the unique completion token the survey shows at the end, and match it to exactly one response. Add an attention check and compare completion time against your pilot, but treat those as signals to review against criteria you disclosed up front, not as automatic rejections. When you post a new wave, exclude your earlier survey bounties through the API or MCP.

Is this the same as hiring a survey designer?

No. This page covers paying people to answer your survey. To hire someone to write, program, or analyze a survey, post a single-seat bounty that describes that deliverable, the tool, and the review steps.

Is a RentAHuman survey sample representative?

No. It is a convenience sample of marketplace workers who chose to apply. Use screening to reach the people you need, report the recruitment method, and do not present results as population estimates.

Product and integration links

Post a survey bounty Create an API key