There is no single price for a survey response. What published rate cards and cost explainers agree on is what moves the number: how many completed responses you need, how long the survey takes, how hard the people are to find, and how fast and how much service you want. This article walks through each driver, shows what one published rate card looks like, works a sample budget two ways, and states what Chorus Research charges so you can compare like with like.
Sources: this article draws on two public pages, the pricing page of the Open Probability-Based Panel Alliance and a Drive Research explainer on online survey cost, both read on 30 September 2026. Figures from them are paraphrased, not quoted, and may have changed since.
The four cost drivers
1. Sample size. Cost scales with the number of completed responses, or with respondent-minutes where a panel bills by time. Some rate cards lower the unit rate after a volume threshold; others charge one fixed fee for the whole panel however many respondents you use. Drive Research, a market research agency, describes about 400 completes as the usual target for most general-market surveys.
2. Survey length. Two of the panels on the rate card discussed below bill per respondent-minute. On that basis a ten-minute survey costs exactly twice what a five-minute survey costs for the same sample. Cutting questions is the one cost lever you control entirely before you talk to anyone.
3. Audience. Agencies call this sample feasibility. A customer-provided list is the cheapest sample because nobody has to be found; specialised audiences such as C-level executives cost more. The underlying variable is incidence, the share of people screened who qualify. The lower it is, the more people have to be screened for every completed response, and the price rises with it.
4. Time and services. Drive Research recommends planning around a two-week fielding window and notes that rush delivery carries a premium. A full agency quote also carries lines for project management, survey design, programming and testing, incentives, and analysis and reporting.
What one published rate card looks like
The agency explainer we read describes the drivers without putting numbers on them. One page that does publish numbers is the pricing page of the Open Probability-Based Panel Alliance, a consortium of probability-based research panels whose rate card is set for academic and publicly funded research rather than commercial buyers. As read on 30 September 2026 (figures paraphrased from that page; check the current rate card before you budget):
- Understanding America Study (US): $3 per participant-minute for the first 500 participants, $2.50 per participant-minute after that.
- LISS Panel (Netherlands): about €1.30 per respondent per minute, with a minimum of about €4,250.
- Swedish Citizen Panel: roughly €7,600 for 1,000 responses and €11,800 for 2,000, 2026 pricing plus VAT.
- GESIS Panel.pop (Germany): roughly €25,000 plus VAT for a standard five-minute module, with a €12,500 plus VAT minimum for a two-minute survey; all of the roughly 5,000 panel members are included and there is no small-sample discount.
These are probability panels built for academic and public-sector research. They are not commercial opt-in sample, and they are not the price to expect from a consumer panel or a self-serve tool. They are useful here because they are public, dated, and show the arithmetic plainly.
A worked example at the academic rates
Take a five-minute survey and price it at the Understanding America Study rate:
- 200 respondents: 200 × 5 minutes × $3 = $3,000.
- The same survey at ten minutes: 200 × 10 × $3 = $6,000. Length alone doubled the bill.
- 1,000 respondents at five minutes: the first 500 cost 500 × 5 × $3 = $7,500, the next 500 cost 500 × 5 × $2.50 = $6,250, total $13,750.
Price the 200-respondent, five-minute survey at the LISS per-minute rate instead: 200 × 5 × €1.30 = €1,300. But the LISS minimum is about €4,250, so that is the bill. At GESIS, a 200-person study costs the same as a 5,000-person one, because the whole panel is included in the module fee.
Three lessons fall out of this:
- Length is a multiplier you control before you spend anything.
- For small orders, the minimum matters more than the unit rate. Ask about it first.
- Rates and minimums are dated and change. Re-check before you budget.
The same example at Chorus Research's public price
Chorus Research publishes its prices on the homepage. A US general-population audience is $1.00 per completed response, $5 minimum; custom audiences are from $1.00 per completed response, and the exact figure for a custom audience comes from the estimate for that audience (prices as published 30 September 2026; the estimate shown in the product before you pay is the source of truth).
- 200 completed responses, US general population: 200 × $1.00 = $200.
- 1,000 completed responses: $1,000.
- 5 completed responses: $5, which is also the minimum. A three-response launch is still $5.
Pricing is per completed response. For a custom audience, run the estimate first: through the hosted MCP server, the estimate_cost and estimate_launch tools return the figure, and that endpoint never charges on its own.
Why quotes vary so much between vendors
Two vendors can quote very different numbers for the same brief because they bundle different things. An agency quote includes design, programming, testing, project management and reporting. A panel rate card covers only the respondents, usually with a minimum order. Chorus prices the launch itself per completed response. Before comparing quotes, strip each one back to the cost per complete for your audience and length, then add back only the services you actually need.
Questions to ask before you buy
- What is the price per complete for fewer than 100 responses, and is there a minimum order?
- Is pricing per complete, per respondent-minute, or a project fee?
- How does incidence change the price, and who estimates incidence for my audience?
- What is the standard turnaround, and what does rush cost?
- Is there a self-serve or API path, or only a quote?
Neither of the two pages we read for this article answered the first or the last question directly.
How to get an estimate from Chorus
Chorus Research exposes the same operations through a REST API and through an MCP server for AI assistants. The hosted MCP endpoint is mcp.chorusresearch.io/mcp; its tools include list_audiences, create_survey, estimate_cost, estimate_launch and get_survey_status, and it never charges: a launch returns a browser checkout link instead. Refund terms are on the refunds page. Chorus Research is operated by Collective Measure, Inc.
See the public prices on the homepage, the REST API documentation, the MCP documentation, the getting-started guide, the FAQ, and refund terms.