Only real human responses are good enough for your insights

Conjointly delivers over 20 advanced checks to protect your data across digital fingerprinting, real-time monitoring, and response analysis. They are enabled by default for Conjointly sourced sample, and fully available when bringing your own respondents.

What you get is cleaner data and savings of up to 30% on costs otherwise lost to fraud.

Multi-layered quality assurance

How Conjointly protects your surveys from low-quality responses

ID Verification

Every respondent sourced through Self-serve sample C (and, where possible, Predefined panels) completes a bank-grade identity verification step before joining the panel, ensuring you are surveying real human respondents who meet your target definitions.

Respondent data

Matches with ID provided
First name Jamie
Last name Sample
Date of birth 01/01/1990
Gender Female
Document type Passport
Issuing country Australia
Bank-grade identity verification step

Digital fingerprinting

Block scripted bots, AI agents, VPN users, and duplicate entries by IP, cookie, or link before they reach your survey.

Participant IDTime of opening surveyStatus
23913804926 June 2026 06:10 AMDisqualified due to the same IP address as another respondent
23918975126 June 2026 07:07 AMCompleted, but marked as a low quality responseMalicious variables detected
23918987326 June 2026 07:12 AMCompleted, but marked as a low quality responseAnomalous network activity detected
23918989626 June 2026 07:12 AMCompleted, but marked as a low quality responseAvoidance of network detection
Advanced fingerprinting blocks bots and fraud

Real-time monitoring

Flag respondents who rush, provide gibberish or copy-pasted answers, or skip the scrolling and mouse movement a question requires.

ParticipantTime of opening surveyStatus
19895410226 June 2026 06:00 AMCompleted survey
19895410926 June 2026 06:02 AMScreened out on Q5
19895411526 June 2026 06:04 AMCompleted, but marked as a low quality responseAverage duration on conjoint sets was too short
19895413126 June 2026 06:06 AMCompleted, but marked as a low quality responseRespondent switched tabs too much
Real-time monitoring flags low-quality responses

Response filtering

Review and manage the remaining responses within the Quality Checker, ensuring only genuine human respondents remain in your dataset.

Survey answers

Q3:

  • Positive: Nothing
  • Negative: It looks fancy and like it would be stolen

Q4:

  • Dark Black
  • Metallic Blue
  • Pearl Red
  • Green Neon
  • Walnut Brown

Status of 220467839

Exclude

Assessment of quality

Response started at 26 June 2026 01:23:54 PM and took 2 minutes 34 seconds.

Participant status: complete.

Desktop.

Location is deemed to be Indianapolis, USA based on the connection.

Response filtering keeps genuine respondents

Feedback looping

Every response and every researcher decision feeds back into the system to continually monitor panel health and sharpen your quality benchmarks over time.

PanelAll entriesBad entriesRatio of badMean quality
Supplier A28414449%71%
Supplier B1035647%69%
Supplier C612845%67%
Supplier D8450%77%
Supplier E5367%42%
Total46123546%70%
Feedback looping improves detection over time

Examples of quality checks in action

A closer look at specific checks working on every study. This is a selection, not the full list.

Flagged-source blocking

Blocks bots and repeat entries whose devices or connections are already flagged.

Network activity detection

Flags VPNs, proxies, and virtual machines used to fake eligibility.

Device & location filtering

Restricts entry by device type and geographic region.

Survey speed tracking

Flags respondents who finish suspiciously fast.

Selection-pace monitoring

Flags respondents who click through choices too quickly.

Scroll and mouse tracking

Reads scroll depth and mouse activity for signs of disengagement.

Tab-switching detection

Flags respondents who keep leaving the survey to look elsewhere.

Copy-paste detection

Catches answers pasted in from search engines or other sites.

AI open-end analysis

A language model flags gibberish and answers that mirror other respondents.

Certified quality, from source to data

Conjointly's sampling services hold ISO 20252:2019 certification (Annex A), confirming that the quality processes behind both Predefined panels and Self-serve sample A and B meet international research standards.

Certified quality, from source to data

Advanced tools and question types that keep respondents engaged

In addition to blocking bad survey actors, thoughtful design keeps genuine respondents engaged and focused, driving relevant, reliable insights you need. Conjointly provides a sophisticated suite of research methods and question types designed to hold participant attention from start to finish.

Advanced tools and question types that keep respondents engaged

Experts on hand at every stage

Conjointly's project management team is highly trained in securing all survey and respondent data while adhering to industry regulations, ensuring peace of mind for you and your research participants.

With a team based worldwide, you will always find an attentive research expert ready to discuss your research needs.

Conjointly support team

Learn more about topics of response quality management

Multi-layered response quality management

See every quality check available and how each layer protects your data.

Including or excluding responses

Review the quality scores and signals for each response, then decide which to include in your analysis.

Quality assurance for Predefined panels

See how Conjointly's fieldwork team hand-checks responses at every stage.