Ask participants for a photo during Live Vision interviews
Participants can now upload or take a photo mid-interview. Conveo checks that it matches, screens on it, and asks follow-up questions about it.
Dag MalstafAt a glance
- Add a picture question to any topic guide and describe the photo you want
- Participants take a photo or pick one from their phone or computer
- Photos that do not match are rejected with a reason and a retry
- Qualify or screen out participants based on what their photo shows
- The AI research assistant asks follow-up questions about the picture
Ask someone about the laptop they work on or the yogurt in their fridge, and you get their memory of it. Memory skips the brand on the lid, the half-empty pot at the back, the second fridge in the garage. Screeners have the same gap: people tell you what they think qualifies them. Until now, the only way to see the real thing in a Conveo interview was to hold it up to the camera, which fails when it is in another room or at the office.
How it works
Describe the photo you want. Add a picture question to your topic guide in plain words, for example "the computer you use".
Participants take or upload it. They snap a photo on the spot or pick one they already have. It works on desktop, in mobile browsers and in the Conveo app for iOS and Android.
Conveo checks the fit. The AI research assistant compares the picture with your description. If it does not match, the participant sees why and gets two more tries.
The interview builds on it. Follow-up questions pick up what is in the picture, in the direction you set in the guide.
Screen on the photo before the interview starts
A picture question also works as a screener. Define the groups a photo can fall into and which of them qualify. In the demo below, a MacBook qualifies and a Windows or Linux laptop does not. People who do not fit are screened out before the interview begins, and you can review every picture next to the reason for the decision.
Research teams asked for it in five situations
These come from teams we work with in consumer goods, food and drink, beauty and retail. We kept them anonymous.
Home-use tests. Participants show the product you shipped them before the conversation goes deeper, so you know they actually tried it.
Screeners. Only people who can show the brand and variant you need get through. That matters most when a claimed purchase is hard to check any other way.
What sits next to your brand. The bathroom cabinet, the pantry or the fridge shows what people really keep at home.
Routines over time. At each check-in, a photo of the setup gives the answers a fixed reference point.
Out in the store. Participants turn the camera to the shelf and talk you through what they see.
Ask for the product and keep faces out of frame. The photo only needs to show the thing you're studying.
What this means for your research
The check, the screen and the follow-up now happen in one video conversation, while the participant still has the picture in mind. There is no separate upload to chase and no callback to schedule. Every photo is checked against your brief and discussed while it is fresh.
One limit to know: a photo from someone's gallery could have come from anywhere. When you need proof the item is in front of the participant right now, use visual verification, which checks on the live camera.
Three minutes on picture upload, start to finish
How it works, the screening demo and the five situations research teams asked for, in under three minutes.
Watch Dag set it up
Dag Malstaf, who built the feature, walks through a picture question, the screening rules and what participants see. Picture upload is available on request from September 28: ask your Research Impact Lead to switch it on for your next study. Read the topic guide setup guide, or see the screener questions guide for the qualifying rules.