BoardSnap is an iOS app that reads an affinity mapping whiteboard — clusters of sticky notes with header labels — and produces a structured insights summary organized by theme.
Affinity mapping is the step after a flood of data. You've done the user interviews, run the brainstorm, or collected the post-retro feedback. Now you have a hundred sticky notes that need to become a coherent set of themes. You sort them on a whiteboard — moving notes around, grouping similar ones together, finding the name for each cluster.
This process is valuable precisely because it's physical and collaborative. Multiple people cluster simultaneously. Disagreements surface in real time. The clusters that emerge from a good affinity mapping session are more honest than categories imposed before the data was collected.
But the board at the end of the session is a wall of colored sticky notes. Making it into a usable artifact — a synthesis doc, a design brief, a research report — requires reading every note, naming each cluster correctly, and writing the summary from scratch. A hundred-note affinity map takes an hour to synthesize even for the person who ran the session.
One observation, quote, or idea per note. Don't editorialize yet — write what was said or observed, not what you think it means. Direct quotes from user interviews are particularly valuable. Use one color per data source or participant if you want to track provenance.
Put every note on the board in a random arrangement first. No clusters yet. The visual mass of notes is the starting point. This step takes five to ten minutes and is important — seeing the full dataset before clustering prevents premature categorization.
Move notes that seem to belong together into proximity. Work silently for the first ten minutes. Everyone moves notes simultaneously. If a note could go in two places, duplicate it or put it at the border between clusters.
After silent clustering, discuss the groupings. Rename notes that are ambiguous. Merge clusters that overlap. Split clusters that are too broad. A good affinity map has five to twelve clusters — more than twelve is usually over-segmented.
Write a header label for each cluster in a different color or larger text. The cluster name is the insight — not a description of the content, but an interpretation. 'Users feel judged by the system' is a better cluster name than 'Feedback issues.'
Put a star or circle around the two to three clusters that represent the strongest themes — highest note density, clearest signal, most actionable. These are the findings that lead the synthesis.
Open BoardSnap. The board has clusters of sticky notes with header labels above each cluster. Step back to get the full map in frame. BoardSnap AI reads the headers, counts notes in each cluster, and summarizes the content of each cluster.
A structured insights summary organized by cluster. Each cluster section includes: the cluster name (from the header label), a count of notes in the cluster, and a summary of the key ideas in the cluster. Starred clusters are flagged as primary findings. The output reads like the insights section of a UX research report.
The research team had 120 sticky notes from ten user interviews. After a 75-minute affinity mapping session, they had eight clusters. BoardSnap read the cluster headers and summarized each one. The researcher used the output as the first draft of the research report — editing took thirty minutes instead of four hours.
A product team generated 80 feature ideas in a brainstorm. After clustering on the whiteboard, they had six themes. BoardSnap read the themes and the notes in each. The product team used the cluster summary to identify which themes had the most ideas — a proxy for where the team's energy was.
A consultant ran a large-format retro with a 20-person team. Too many voices for a simple retro board. Instead, they ran an affinity mapping session on all the feedback cards. BoardSnap read the eight clusters that emerged and produced a synthesis the consultant used in the executive debrief.
Yes, though legibility depends on handwriting clarity. For dense clusters where notes overlap significantly, step close enough to the board that the notes in each cluster are visible. Or snap cluster-by-cluster if the board is too dense for a single shot to capture all notes.
Yes — cluster membership is determined by proximity to the cluster header label and proximity of notes to each other. Clear spatial separation between clusters improves accuracy. Leave visible gaps between clusters when finalizing the map.
Yes. Snap the board at different stages — after initial clustering, after refinement, after labeling. Each snap is a board in your project. Comparing the stages shows how the thinking evolved during the session.
Snap the board, ship the action items in ten seconds.