Zurück
How to Find Survey Respondents: 9 Sources That Work

Matthieu Saussaye

A good questionnaire sent to the wrong people is worse than no survey at all: it gives you a confident wrong answer.
Most guides on how to find survey respondents list channels and stop. The harder questions are which channel fits your study, what each one really costs per usable answer, and how to keep bots and speeders out. This guide answers those with prices from the vendors' own pages, and shows how SmartInterview lets your customers and a recruited audience answer the same study.
The Short Answer
SmartInterview: the easiest route for most teams. Send a reusable link to your own customers and add a recruited audience in Switzerland and Europe in the same study, with every answer tagged by source. The first 50 respondents are free.
Your own customer list, website and email: the cheapest and most relevant respondents, but they only tell you about people who already know you.
Self-serve consumer panels such as Pollfish (from $0.95 per response) or Centiment (from about $1 per response): fast reach to people who have never heard of you.
Prolific: a vetted pool of 300,000+ participants, popular for academic and behavioural studies, priced on participant pay plus a platform fee.
B2B recruiters such as User Interviews or large sample suppliers like Cint and Dynata: for professionals and hard-to-reach audiences, at a clearly higher price per person.
Where to Find Survey Respondents: Sources Compared
Source | Price entry point | How respondents arrive | Analysis included | Best for |
|---|---|---|---|---|
SmartInterview | First 50 respondents free; plans from CHF 45/month for 150 respondents | Your customers via a reusable link, plus a recruited audience in Switzerland and Europe, tagged by source | AI follow-ups, open answers coded into themes, sentiment, cross-tabs, Excel/SPSS | Brands that need customers and non-customers in one study |
Own list (CRM, email, website) | Free, plus incentives | Email, in-app or website invitation | Whatever your survey tool offers | Satisfaction, churn, feature feedback |
Social media and communities | Free or ad spend | Posts, groups, paid ads | Whatever your survey tool offers | Early exploration, niche hobbies |
Pollfish | From $0.95 per response; screening, quotas and demographics cost extra | Panel of 250M+ consumers (Prodege) | Dashboard; full service from $5k | Cheap, fast consumer pulses |
Centiment | From about $1 per response, more for longer surveys or niche targets | Managed panel, works with your own survey tool | Results and export | Managed sample with support |
Prolific | Participant pay (min. $8/hour) plus a 42.8% platform fee for companies | 300,000+ verified participants, 38+ countries | None, bring your own tool | Academic and behavioural research |
User Interviews | $49 per session, $98 for B2B (pay as you go) | 7.5M participants, screened per project | None, bring your own tool | Interviews and B2B recruits |
Cint / Dynata | Quote-based | Large global sample marketplaces | Not the focus | Multi-country, high-volume studies |
Prices are the public entry points at the time of writing; niche audiences cost more everywhere. For a wider view of the platforms, see our market research tools roundup and the list of Pollfish alternatives.
Before You Pick a Source: Who, How Many, and How Clean
The channel is the last decision. Three questions come first. If panels are new to you, read what panel research is first.
Define who counts as a respondent
Write your target in one sentence that you could turn into screening questions: "French-speaking Swiss adults who bought a food supplement in the last six months" is usable. "Health-conscious consumers" is not. Each criterion becomes a screener question, and each one makes a paid sample more expensive.
Decide how many you need, and in which groups
For a simple yes or no across a population, 300 to 400 completes gives you a margin of error of roughly plus or minus 5 points. The catch is subgroups: if you want to compare customers with non-customers, or three regions, each group needs enough people on its own, usually 100 or more. Set quotas so that one group does not fill the whole sample. Our guide to sampling methods covers the trade-offs.
Plan your quality checks before launch
Every source attracts bots, speeders, duplicates and professional survey takers. Build in checks: a minimum completion time, one attention check, an open question you actually read, and a duplicate check on device or email. Also watch survey bias: an engaged customer list over-represents fans, and social media over-represents people who like answering surveys.
Expect low response rates and plan for them
Email surveys to your own customers often get single-digit response rates, and the trend is down, as we explain in why survey response rates are crashing. A short survey, one reminder and a modest incentive help more than any channel trick.
Survey Respondent Sources in Detail
1. SmartInterview — panel + your own customers in one study
Most studies need two audiences: people who know you and people who do not. SmartInterview handles both in one questionnaire. You share a reusable link with your customer list, newsletter or website, and in the same study you add an audience recruited in Switzerland and Europe. Each response is tagged by source, so you can read the groups separately or together. The AI drafts and scores the questionnaire, asks a follow-up when an open answer is vague, accepts text or voice in 99+ languages, and codes open answers into themes. Results typically arrive in about 48 hours. Pharmalp, a Swiss natural health brand, used it to validate the pricing and positioning of four new supplement concepts; see the Pharmalp case study.
Choose it if: you want customers and a fresh audience compared in one study, in Europe, with the analysis done for you.
2. Your customer list and CRM — the most relevant people you have
Your buyers know the product and have opinions. Segment the list (recent buyers, lapsed users, trial users) instead of mailing everyone, keep the survey under ten minutes, and send one reminder after three or four days. Under GDPR you need a lawful basis to email people for research. This source is ideal for customer feedback surveys and Net Promoter Score, but useless for sizing a market you do not yet serve.
Choose it if: the question is about current customers.
3. Website and in-app intercepts — catch people in context
A short pop-up or embedded survey reaches visitors while they are doing the thing you want to understand: comparing plans, abandoning a cart, trying a feature. Keep it to two or three questions.
Choose it if: you have steady traffic and a question about a specific moment.
4. Social media, forums and communities — cheap, but skewed
LinkedIn groups, Reddit, Facebook groups, Slack or Discord communities and niche forums can deliver respondents for free, especially for hobbies and professional niches. Ask the moderators first and share the results back. Expect self-selection: fine for exploration, risky for numbers you plan to quote.
Choose it if: you are exploring and the budget is zero.
5. Pollfish — low-cost consumer reach
Pollfish lists DIY surveys from $0.95 per response, drawing on more than 250 million consumers through its parent company Prodege. Screening questions, advanced demographics and quotas each add a published surcharge, so a targeted study costs several times the headline price. Full-service projects start at $5k. Our Centiment vs Pollfish comparison goes deeper.
Choose it if: you need a quick, broad consumer pulse and will do the analysis yourself.
6. Centiment — a managed panel that works with your tool
Centiment prices its panel per response, starting around $1 for short surveys to broad audiences, with its own examples rising to several dollars for IT decision makers or longer questionnaires. You can run the survey in your own tool.
Choose it if: you already have a programmed survey and only need sample, with human support.
7. Prolific — vetted participants for research
Prolific runs its own pool of 300,000+ verified participants in 38+ countries, with 300+ filters. You pay participants by time (the minimum is $8 or £6 per hour, the recommended rate $12 or £9) plus a platform fee of 42.8% for companies and 33.3% for academics. You bring your own survey tool and analysis.
Choose it if: you run academic or behavioural research and want a well-documented pool.
8. User Interviews and B2B recruiters — professionals and niches
Finding 50 procurement managers or hospital pharmacists is a different job. User Interviews publishes pay-as-you-go recruiting at $49 per session for general audiences and $98 for B2B, with incentives on top, from a pool of 7.5 million participants in 34 countries. To validate a product idea with customers in a B2B niche, this is often the fastest path.
Choose it if: you need a small number of specific professionals.
9. Cint, Dynata and large sample suppliers — scale and many markets
Cint runs a sample marketplace covering 130 countries and reports more than 200 million completed surveys a year; Dynata operates a global panel that includes B2B respondents. Both price by quote.
Choose it if: you run high-volume, multi-country tracking, usually through an agency.
Where I draw the line
I stop buying sample when the answers I get back are too thin to act on. Doubling a panel order narrows the margin of error a little, but does nothing when half the open answers say "good" or "too expensive". The fix is better probing, not more people.
The other mistake I see is surveying only your own customers and calling it market research. Customers tell you why they stayed. To learn why others never bought, you need people who do not know you, asked the same questions. That is the setup I would choose for almost any brand study.
Matthieu Saussaye, Founder of SmartInterview.
A Practical Checklist for Recruiting Survey Respondents
Write the decision first. Name the choice the survey will inform. It tells you who must answer.
Turn the target into 2 to 4 screener questions. Neutral wording, with plausible wrong options so people cannot guess their way in.
Set the sample size per group, not just in total. At least 100 per subgroup, with quotas.
Mix sources on purpose. Your customers for depth, a recruited audience for the market view, each tagged so you can compare.
Choose a fair incentive. A small voucher, a prize draw (check local rules) or a donation for customers; panels handle incentives for you.
Keep it short and ask at least one open question. See our survey questions guide and how to write open-ended questions people actually answer.
Soft-launch to 20 to 30 people. Check timing and drop-off before paying for the rest.
Clean before you analyse. Remove speeders, failed attention checks, duplicates and nonsense open answers.
Reading the open answers takes longest. Our review of survey analysis tools compares the options, or try it on SmartInterview with your first 50 respondents free.
When another tool is the better call
Dozens of countries outside Europe in one wave: a global marketplace such as Cint or Dynata.
An academic experiment with strict participant rules: Prolific.
Ten one-hour interviews with niche professionals: a recruiter such as User Interviews.
Frequently Asked Questions
Where can I find survey respondents for free?
Start with people who already know you: your customer list, newsletter, website visitors and social followers. Relevant communities, forums and LinkedIn groups also work if you ask the moderators and share the results. Free sources skew toward engaged people, so use them for exploration.
How much does it cost to buy survey respondents?
Self-serve consumer panels start around $1 per response (Pollfish lists $0.95) before targeting surcharges. B2B and niche audiences cost much more, for example $98 per session on User Interviews pay as you go. On SmartInterview, the first 50 respondents are free and plans start at CHF 45 per month for 150 respondents.
How many survey respondents do I need?
For one overall percentage, 300 to 400 completes gives roughly a 5-point margin of error. If you want to compare groups, plan about 100 or more per group. For exploratory open questions, 30 to 50 good answers often reveal most of the main themes.
How do I get more people to answer my survey?
Keep it short, explain why it matters, send one reminder, and offer a small incentive. A well-timed, targeted invitation, for example right after a purchase, beats a mass mailing.
How do I keep fake or low-quality respondents out?
Use neutral screener questions, a minimum completion time, one attention check, duplicate detection and at least one open question that you read. Short, copied or off-topic open answers are the clearest warning sign.
Can I combine my customers and a panel in one survey?
Yes. On SmartInterview, a reusable link for your own customers and a recruited audience in Switzerland and Europe run in the same study, and every response is tagged by source. That lets you compare customers and non-customers with the same questions.
What is the best way to find B2B survey respondents?
Start with your own customer and prospect lists and LinkedIn communities, then use a specialist recruiter or a B2B panel for volume. Expect higher prices and keep the survey short.
Ready to put your customers and a fresh audience side by side? Start your first study on SmartInterview, free for your first 50 respondents.


