Respondent tech support means helping the people in your study when something technical goes wrong, such as a camera that will not open, a login that fails, or a video that will not upload. It sounds like an operations detail, but in a multi-day bulletin board it directly affects your data. When help arrives late, people drop out and the study loses the exact moments you wanted to capture.
Why it is a methodology issue
When camera permissions fail, insight dies quietly. Emailing a project manager two days later does not resurrect Day 2. In a diary study the moment itself is the data: breakfast, the shelf at the store, the first use of a product. If a respondent cannot capture it and cannot get help right then, that moment is gone.
Typical problems that cause dropouts
- Camera or microphone permission denied on the phone or browser.
- Upload failing on a slow connection.
- Forgotten link or login problems on a returning day.
- Confusion about what a task is asking for.
- Not knowing who to ask, so doing nothing.
Why respondents rarely ask for help
Most people do not email a stranger about a technical problem for a small incentive. They give up quietly, or they assume they did something wrong. That means the support tickets you see are a fraction of the problems that exist. Good support is therefore visible and easy: a button where the problem happens, in plain language, with an answer in seconds.
What good support looks like
- Available from board home through every day
- AI that actually troubleshoots common device issues
- Clean escalation to humans with the transcript preserved
- Notification to researchers when a human is needed
- Plain language that does not assume technical skill
Old way versus in-board help
Swipe to see all columns
| Situation | Email-based support | Help inside the board |
|---|---|---|
| Camera blocked | Respondent emails, waits for a reply | Gets step-by-step help immediately |
| Researcher awareness | Finds out later, if at all | Alerted when escalation is needed |
| History | Scattered across inboxes | Conversation kept with the project |
| Impact on completes | Often lost | More people recover and continue |
Measuring whether support works
- 1Track the share of respondents who start a media task and complete it.
- 2Count support conversations per ten respondents.
- 3Measure how many issues are resolved without a human.
- 4Review escalations weekly to find recurring causes and fix the prompts or instructions.
Example: a respondent cannot open the camera on Day 2 of a home-organization study. In the board she asks for help, gets quick guidance for her browser settings, and uploads the photo within minutes. Without that, the researcher would have seen an empty slot and a no-show.
FAQ