When you’re conducting user research for your business, can you really trust the feedback of remote participants? The rise of fake identities, automated bots, and deepfakes makes it harder to separate useful data from manufactured responses. Learn more here.
Good Products Start With Research Integrity
From usability testing to surveys and focus groups, user research underpins any good business. Reaching consumers directly reveals potential friction before you dedicate significant resources to a new product or service. According to the Interaction Design Foundation, companies that invest in user experience at the start of a project manage to reduce product development timelines by 33 to 50%.
User research is also much more accessible now thanks to the ever-evolving digital environment. Online recruitment and data collection are popular because they reduce travel and logistical costs while reaching people worldwide.
However, remote research depends on one important assumption: participants are who they claim to be.
When Your “Customer” Isn’t a Customer at All
Deepfakes are a form of digital content made or altered by AI that impersonates real people. They’re becoming increasingly widespread in phishing scams because they can mimic someone’s appearance and voice convincingly.
This problem also affects user research. Sophisticated deepfakes have made it easier for “professional participants” to exploit digital anonymity and lax identity-verification practices for financial gain.
Fraudulent participation may involve the following:
- AI-generated or altered identities
- Multiple participant accounts
- Fabricated personal histories
- Misrepresented product experience
Make Your Research Harder To Fake
Why waste resources on in-depth scientific research if the risk of deepfake scams can make your findings unreliable? Consider these preventive measures.
Strengthen the Screening Stage
Businesses should tighten recruitment before a session even begins. Look for clusters of unusual applications, suspiciously similar contact information, repeat participants, or answers that seem unusually uniform.
Ask Questions That Require Real Experience
Move away from questions participants can answer with a simple yes or no. Someone who claims to use a product every day but cannot explain its basic features or recount a recent experience deserves closer scrutiny.
Watch for Patterns During the Session
Maintain a cautious mindset during the interview itself. Researchers, for example, can watch for repeated scripted language or knowledge that doesn’t match the participant’s claimed experience. Responses that seem “off,” combined with poor audio or a reluctance to use video, might indicate deception.
Know When To Stop
Create a simple response process for when something no longer adds up. This practical framework might include:
- Clarifying the discrepancy
- Ending the session if you can’t establish eligibility
- Documenting what happened
- Removing unreliable information from the dataset
- Reviewing how the participant passed screening
Let Technology Fight Technology
Why not use the same technology that enables sophisticated fraud to help detect it? AI-powered pattern recognition tools help flag repeat participants, suspicious response clusters, and unusual recruitment behavior at a scale human reviewers might miss.
Trust Is Becoming Part of the Research Process
Treat participant authenticity as part of the research methodology itself. As deepfakes become more convincing, businesses need to preserve the quality of remote insights through robust, proactive measures.