Colectiv Mosaic is the platform we use to gather and analyse qualitative information at scale, so teams can act on what they learn.
We work with partners to define the questions they need to answer and design the best way to listen to the people closest to those answers. That might mean AI-led interviews, a few questions sent regularly to frontline staff, or open feedback gathered in person from communities.
We then analyse the data in Mosaic, turning large amounts of dispersed frontline and community feedback into patterns, themes and insights that teams can explore and use for themselves.
Easy for participants
The people sharing insights don’t need to use Mosaic itself. We design the easiest way for them to contribute in each context.
That might mean responding by WhatsApp or a web link, using voice or text, answering in their own language, or simply sharing an observation with a frontline worker in person.
Where we use AI interviews, people can respond in their own time and the interviewer follows up on what they say. Participant feedback has been overwhelmingly positive, with people valuing how simple, flexible and conversational the experience feels.
Read what participants told us about the experience.
Anonymous by design
We want to understand what people collectively know and experience, not who said what.
Participant contributions are collected anonymously, with access to study data limited to authorised users. Where demographic information is useful for understanding differences between groups, it is handled separately and securely.
Interview content is not used to train general AI models, and personal data is handled in line with GDPR and the requirements of each study.
Human-led analysis at scale
Our research team leads the analysis. We define the topics we need to understand, and Mosaic systematically codes the data against them.
Mosaic helps identify recurring patterns within each topic, grouping related evidence into themes and insights.
We combine AI with separate algorithmic checks to ensure that every quotation used as evidence appears exactly in the original transcript or text. This creates a transparent evidence trail from each insight back to what people actually said.
Explore the evidence
The analysis and underlying data sit together in Mosaic, so teams can see both the patterns and the evidence behind them.
This is how a single topic looks in Mosaic: a summary of what people said, followed by the themes behind it.
