Mobile ethnography is a qualitative research method where participants document their own lives with their phone. They record videos, take photos and leave voice notes in the place and at the moment something happens: in the kitchen at breakfast, in the aisle at the shop, on the sofa in the evening. The researcher plans the tasks, watches what comes in, asks follow-up questions and analyses the result.
It grew out of classic ethnography, where a researcher spends time with people in their own setting. The phone makes that setting reachable at scale. Instead of one researcher in one home for a day, you can follow thirty people in three countries for two weeks.
Why agencies use it
The main reason is recall bias. When you ask people in a group or a survey what they usually do, you hear what they remember and what sounds reasonable. When they show you, you see what they actually do. The difference is often the insight: the second coffee nobody mentions, the product stored behind the cleaning things, the label read three times before choosing.
Other reasons come up again and again:
- Context. The room, the objects and the people around tell you as much as the answer.
- Time. A study over days or weeks shows habits and change, not one moment.
- Reach. People in small towns, busy parents and night-shift workers take part when it suits them.
- Evidence. A ten-second clip of a real customer is more persuasive in a debrief than any chart.
Mobile ethnography and diary studies
The two terms are often used together. A diary study asks people to report regularly over time, often with the same short questions each day. Mobile ethnography puts more weight on showing context on camera. In practice most projects mix both: a daily check-in plus a few richer tasks, such as a tour of the fridge or a video of a routine.
How to plan a mobile ethnography study
1. Start from the decision
Write down what the client will decide with the results. "Which pack size to launch in Poland" leads to different tasks than "understand breakfast". Every task should help with that decision.
2. Choose the length
Five to seven days suits one routine. Two to three weeks suits a new product, a change in life or a habit that varies by day. Longer studies need lighter tasks.
3. Write tasks people can do in a few minutes
A good task asks for one thing and says what to show. Compare:
- "Tell us about your snacking." (vague, people talk in general)
- "Open the cupboard where you keep snacks. Film it and tell us what you would buy again." (specific, in context)
Mix formats. A photo, a short voice note and one video a day are easier than three videos.
4. Plan the rhythm
Decide on which day each task opens, at what time in the participant's own time zone, and how long it stays open. Use repeatable tasks for things that happen at random, such as "every time you snack".
5. Recruit for the method
People who are happy to film at home are not a random sample. Recruit carefully, tell people up front what the study asks, and include a light first task so they can practise.
6. Pilot it
Try every task yourself on a real phone, in every language, before the study goes live. On Zolensa, test participants are free for this.
What to watch out for
- Privacy at home. People film their homes and families. Ask them not to film anyone who has not agreed, and keep consent clear and in their language.
- Performance. Some participants act for the camera in the first days. It fades. Do not judge a study on day one.
- Fatigue. Too many long tasks and people drop out or give short answers. Reply to them and keep the effort fair.
- Data overload. Thirty people over two weeks can send hundreds of videos. Plan how you will review: tag as you watch, mark highlights, and use transcripts and search to find moments.
Where AI helps, and where it does not
AI is now good at the slow parts: transcribing videos, translating them into the team's language, describing photos and suggesting themes. That leaves the researcher more time to watch and think. It does not replace judgement. Treat AI themes as a first draft, check that every quote is real and in context, and decide yourself what matters for the client. We wrote more about this in AI in qualitative analysis.
How Zolensa supports mobile ethnography
Zolensa was built for this method. Participants use a browser app on their own phone, in one of eight languages, with nothing to install. Tasks open day by day in each person's time zone, with one reminder by email, WhatsApp or text message. Your team watches answers arrive with transcripts, asks follow-up questions, tags and highlights, and exports everything for the report. See the mobile ethnography use case for how a study runs.
