Survey fraud
Compromised people or compromised answers entering a survey.
Includes bots, farms, duplicates, spoofed geo, professional takers, and AI-written responses. Detection that only looks at the person misses half of it.
Full pageGlossary
The language of sample integrity
Short definitions for the threats and controls research ops actually argue about. Respondent quality is half the dictionary. Response quality is the half most vendors skip.
Other platforms score the respondent. Maxna scores the respondent and the response.
Device IDs, bots, VPNs, and duplicates are necessary. They are not enough. A real device can still paste ChatGPT into every open-end. Maxna's proprietary AI grades what people type - so sample integrity is both who they are and what they say.
Respondent quality
Where most tools stop
- Device fingerprinting and environment risk
- Bots, scripts, and click farms
- VPN / proxy and geo mismatch
- Duplicates and professional respondents
Response quality
Maxna's proprietary AI layer
- AI-written and synthetic open-ends
- Low-effort and copy-paste answers
- Cross-question consistency
- Open-end scoring before the complete is paid
Library
Terms
Respondent quality
Is this a real, unique, in-market person?
Device, IP/VPN, bots, farms, duplicates, professional respondents. Necessary. Where most competing tools stop.
Full pageResponse quality
Are the answers usable and human?
Open-end effort, specificity, consistency, and AI-generation. Maxna’s proprietary layer. Identity tools do not do this job.
Full pageSample integrity
Respondent quality plus response quality, enforced at the gate.
The standard Maxna is built to operationalize. Broader than “fraud detection” and broader than a fingerprint.
Full pagePre-survey screening
A pass/fail before the questionnaire starts.
Prevention, not post-field cleaning. Maxna’s default architecture: redirect or API in front of Qualtrics, Forsta, or a router.
Full pageClick farm
Organized humans completing surveys at industrial scale.
They survive bot filters. They now use residential proxies and LLMs. Detection needs clusters plus content AI.
Full pageProfessional respondent
A person who lives in surveys for incentives.
Games screeners, hops panels, often pastes AI into open-ends. Not a bot. Still an integrity failure.
Full pageDuplicate respondent
The same person attempting a study more than once.
New browser or new panel ID does not make a new human. Cross-supplier duplicates wreck blended sample.
Full pageAI-written survey answer
Open-end or constructed text generated or heavily rewritten by a model.
Fluent, generic, low specificity. Respondent-quality tools cannot see it. Maxna scores it in-field.
Full pageOpen-end scoring
Grading verbatims for effort, specificity, and synthetic language.
Broader than AI detection. Lazy humans fail too. Should happen before the complete is paid.
Full pageDevice fingerprinting
Probabilistic environment matching across sessions.
A respondent-quality signal. Necessary and easy to over-trust. Not a substitute for behavior or response AI.
Full pageVPN / proxy spoofing
Hiding or faking network location.
Breaks in-market specs. A risk signal, not always an automatic ban - fuse with the rest of the stack.
Full pageSurvey bot detection
Catching automated or scripted completes.
Headless browsers, replay, non-human timing. Table stakes. Hybrid human+LLM traffic needs more.
Full pageCross-supplier linking
Seeing the same actor hop panels mid-field.
Each supplier’s own dedupe is blind to the others. A blended study needs one identity graph.
Full pageFalse positive
A good respondent terminated by the gate.
Every detector has them. Integrity software should expose reasons and let ops tune - not hide in a single opaque score.
Lumen, Cadence, Aether
Maxna protection tiers.
Same gate, different intensity. Lumen is baseline respondent quality. Cadence adds behavior and open-end signals. Aether is maximum integrity including deep AI grading and cross-supplier linking.
Full page