The Company

weseethrough is a technology-driven market research company. We use wearable technology to capture a first-person view of people’s lives and a video curation platform to sift through huge amounts of video footage. By providing an unfiltered and unaffected observation of consumers, we are able to uncover truly powerful insights and answers without relying heavily on interview responses. We’re less than two years old, we’ve already won ~£400k of work from big-name clients like Unilever and we won the 2015 MRS award for innovation in research methodology. Business is booming.

Our Platform

Haystack is our proprietary web app that we use to analyse the videos we collect. Our crowd team annotates hundreds of hours of videos with ‘behavioural codes’ and transcriptions. These codes get added every time they occur. Once this coding process is complete, our researchers can search through the video and easily find every time someone uses Heinz ketchup, or every time someone crushes a stock cube, etc.

The role

We built our video analysis web app around the needs of our projects, and it works – we’ve used it to deliver 100-hour video analysis projects for big-name clients. But we’re more ambitious than that. We want to analyse much more video, much faster: we want to be able to analyse 10,000 hours of video for valuable insightful moments… per week, without breaking a sweat. To do so, we need a radical rethink of the UX and UI and a robust, maintainable implementation. It needs to intelligently coordinate human labour to achieve consistently high-quality analysis. It needs to make good use of machine vision tech.

We’ll be incrementally redesigning our entire platform, one use case at a time, with both running in tandem until the new has completely eclipsed the old.