We live in a world of increased risk from AI-driven fraud, and concerns around panel quality are growing across the market research industry.
Automated bots, duplicate registrations, and fraudulent participants are increasingly able to enter panels and generate data that appears legitimate but is fundamentally unreliable.
When choosing a fieldwork partner and deciding which panel to recruit from, it is essential that protections are in place at every stage of the fieldwork lifecycle. In this article, we outline the advanced technologies required to prevent fraud and identify bots, duplicate respondents, and anomalous behaviour throughout the research process.
At QQFS, this starts with one of the most robust registration and verification processes for our healthcare professionals, built on triple verification including validating evidence of professional status, matching this information to personal photo ID documents, and cross-referencing it with a live image identity check.
1. Panel Quality at Recruitment
The recruitment stage is the first opportunity to prevent fraud and protect downstream data. Without robust controls, automated bots, duplicate registrations, and fraudulent individuals can create accounts that appear genuine, enter the panel, and later participate in surveys. Once this happens, poor-quality data becomes harder to detect and remove.
At QQFS, effective panel recruitment therefore combines multiple layers of verification, including:
- Bot, fraud, and duplicate detection, which we deploy at registration to stop bad actors from creating accounts that could later be invited to surveys. This includes analysing behavioural patterns, device signals, and location data to identify automated or repeated registrations.
- Risk-based registration checks, where high-risk registrations are flagged for additional verification, while low-risk registrations can be verified more efficiently.
- Photo ID and identity verification, ensuring the person registering is a real individual and that their identity matches their claimed professional or patient status.
2. Panel Quality Before and During Surveys
However strong recruitment processes are, panel quality can still be compromised once respondents reach the survey stage. Bots, duplicate participants, and inattentive respondents may attempt to enter surveys, sometimes bypassing basic entry checks. In other cases, genuine respondents may still deliver low-quality data through speeding, straight-lining, or disengaged behaviour.
To address these risks, our quality controls operate in real time, before and during survey participation:
- Multi-layer bot, fraud, and duplicate detection at survey entry, designed to block automated traffic and repeat participation, even if a respondent has previously passed panel registration checks.
- Live monitoring of respondent behaviour, using behavioural patterns, timing data, and location validation to identify suspicious or anomalous activity as it occurs.
- Automated QualityScore™ checks, used across our healthcare studies to assess response quality using a combination of passive data and survey answers, including speeding, straight-lining, poor open-ended responses, and inconsistent answers.
3. Panel Quality Over Time
Panel quality does not end when a survey closes. Without active review and follow-up, the same quality issues can persist across multiple projects. Patterns of inattentive behaviour, repeated low-quality participation, or emerging fraud trends may go unnoticed if panel performance is not monitored at a broader level.
Closing the loop at QQFS involves reviewing panel metrics to understand how respondents and recruitment sources perform over time. This includes:
- Panel-level quality metrics, which we continuously review to track trends in engagement, attentiveness, and risk across the panel.
- Identification of repeat quality issues, allowing patterns of poor behaviour or elevated fraud risk to be detected rather than treated as isolated incidents.
- Re-education and removal processes, where respondents are reminded of quality expectations or removed when standards are repeatedly not met.
In an environment of increasing AI-driven technology, panel quality can no longer be addressed through isolated checks or post-fieldwork data cleaning. Protecting data integrity requires an end-to-end approach – the model we apply at QQFS – covering verification at registration, recruitment, survey participation, and ongoing panel management.
For market research agencies delivering healthcare fieldwork to pharmaceutical clients, this level of rigour is essential. If you would like to discuss how panel quality is protected across your next healthcare research project, share your project details with us.
