Why Remote Testing Is Now the Default
Remote usability testing - conducted over video call or through asynchronous testing platforms - now handles most UX research because it removes the constraints of a physical lab: no travel, no facility booking, and no geographic limit on who you can recruit. It also produces a kind of data lab testing can't: users on their own devices, in their own homes, with their own distractions.
Two Types of Remote Testing
- Remote moderated: researcher and participant connect over video with screen share. Same think-aloud protocol as an in-person session. Nielsen Norman Group considers this the strongest remote format - findings comparable to in-person testing, at unmoderated-level cost and convenience.
- Remote unmoderated: participants complete tasks independently on a platform like Maze, Lyssna, or UserTesting, with no researcher watching live. Studies can launch and return results within hours, with participants working simultaneously instead of queuing for individual time slots.
What You Gain
- Access: recruit from any time zone, not just whoever can reach your office.
- Natural context: participants use their own devices and networks, surfacing device-specific and connectivity problems a lab session would never catch.
- Speed: unmoderated studies routinely return results the same day; remote moderated cuts out travel and room-booking overhead entirely.
- Cost: no lab rental, no travel reimbursement, no in-person incentive premium.
What You Lose - and How to Compensate
- Technical friction: run a pre-session tech check for moderated sessions, and always have a phone-call fallback if video drops.
- Drop-off on unmoderated studies: keep tasks under 15 minutes, pilot-test your instructions on one person before launch, and set incentives appropriately for the time asked.
- Weaker rapport: spend the first few minutes on small talk before the first task, and repeat that you're testing the design, not the person.
- Early prototypes don't unmoderate well: NN/g flags that without a moderator to explain gaps and help participants past broken states, unmoderated sessions on fragile prototypes produce noise, not signal. Keep early-stage testing moderated.
Tools Worth Knowing in 2026
- Maze: unmoderated testing with direct Figma prototype import - the fastest path from design file to results, from roughly $99/month.
- Lyssna: self-serve unmoderated testing with built-in panel access, removing the recruitment step entirely - from roughly $89/month.
- UserTesting: the largest pre-recruited panel (1M+ contributors), covering both moderated and unmoderated studies at enterprise scale and pricing.
- Lookback: purpose-built for live moderated sessions, with native web and mobile recording.
- dscout: mobile-first diary studies and in-the-moment "missions" for longitudinal, in-the-wild behaviour over days or weeks - enterprise-oriented and priced accordingly.
Most distributed teams land on one moderated tool (Lookback or UserTesting) paired with one unmoderated tool (Maze or Lyssna), rather than trying to cover both needs with a single platform.
Key Takeaways
- Remote moderated testing now delivers findings comparable to in-person testing - treat it as the default, not a compromise.
- Unmoderated remote testing is for validated, self-explanatory tasks at scale - not for fragile early prototypes.
- A pre-session tech check and a piloted instruction set prevent most of what goes wrong in remote studies.
- Pair one moderated tool with one unmoderated tool rather than forcing a single platform to do both jobs.
