ResearchDart is the routing layer between sample suppliers and buyer surveys. It checks each respondent before the first question, records how every session ended, and gives both sides the same reasons when a complete is thrown out.
| Supplier | Signals | Decision |
|---|---|---|
| 1188 | None | Passed |
| 1263 | vpn datacenter | Blocked |
| 1042 | None | Passed |
| 1188 | quota_full | Over quota |
| 1042 | None | Passed |
| 1263 | multiple_accounts | Flagged |
Works with any survey platform that can redirect
Most bad data is cheaper to stop at the door than to clean afterwards. ResearchDart covers the whole session, from the entry link to the invoice.
Before the survey
The respondent's browser is checked while a short holding page loads. Automated browsers, anonymised networks and people already in the study are stopped before they cost a buyer anything.
While fielding
Each supplier works the project through its own allocation, entry link and quota. Closed cells return respondents immediately, and the survey reports outcomes on signed endlinks.
After fieldwork
Completes that fail cleaning are reversed with a Global Data Quality code, so suppliers see exactly why and can act on the source instead of arguing about the invoice.
Survey-taking agents now pass attention checks and write fluent open ends. The signals that still give them away are in the browser and the network, which is why ResearchDart checks there, before the survey loads.
01k8zq4r2mxv6c3eIndependent research from the last year explains why screening moved in front of the questionnaire.
99.8%
of attention checks passed by an autonomous AI survey respondent across 43,000 tests.
Westwood, PNAS, November 202513.7%
of respondents removed before or during the survey by sample suppliers worldwide.
Insights Association and GDQ, Data Quality Benchmarking Wave 175.2%
of supplier records used encrypted or server-to-server links, against 91.5% for research agencies.
Insights Association and GDQ, Data Quality Benchmarking Wave 1Buyers and suppliers use the same session records, the same outcome codes and the same removal reasons.
Agencies, brands and research teams fielding online studies.
Panels, rewards apps and communities with respondents to route.
Projects, allocations, sandbox tests, quality reports and reversals without writing code.
Create projects, pull live surveys, report outcomes and reverse completes programmatically.
We configure redirects and screening with your team and run the first sandbox tests together.
Suppliers pull live surveys with payout and remaining completes. Buyers report outcomes server-to-server and reverse completes with standard codes. Every call is scoped to one account key.
# Supplier: surveys open to your panel right now
curl https://researchdart.com/api/v1/supplier/surveys \
-H "Authorization: Bearer rd_…"
{
"project_code": "RDK7M2QX",
"country": "US",
"loi_minutes": 12,
"payout": "1.75",
"completes_remaining": 588,
"entry_link": "…/s/entry/k3m9q2xa7hpd?rid={RID}"
}# Buyer: record a complete without exposing a redirect
curl -X POST https://researchdart.com/api/v1/buyer/transactions/01k8zq4r2m…/status \
-H "Authorization: Bearer rd_…" \
-d status=complete
{ "recorded": true, "data": { "status_code": 10 } }# Buyer: reverse a complete after cleaning
curl -X POST https://researchdart.com/api/v1/buyer/transactions/01k8zq4r2m…/reversal \
-H "Authorization: Bearer rd_…" \
-d gdq_code=15 -d note="OE2 copied from a chatbot"
{ "reversed": true, "data": { "reversal": { "gdq_label": "Open end: AI completed" } } }Most sample websites tell you what you want to hear. These are the things we would rather you knew up front.
Screening respondents before the survey opens is the cheapest place to remove fraud. It is not free, it is not complete, and it creates false positive...
Data QualityA research team built an AI respondent that passed 99.8% of attention checks. That does not mean online sample is finished. It means the checks most s...
Research OperationsWe run a router, so you would expect us to say exchanges win. For some studies a single good panel is the better choice, and we say which ones.
Tell us whether you buy sample or supply it. We set up your account, share sandbox links and walk through the first integration with your team.