Stop 03 · UX ResearchTopics · 10/11

Research Synthesis (Affinity Mapping)

In short

After a week of interviews you'll have pages of notes and no clear answer. Affinity mapping turns that pile into a few strong insights the whole team can act on.

Clustering raw research notes (sticky notes, quotes, observations) into recurring themes — turning a pile of interview transcripts into a small number of actionable design insights.

The jaadu sorts a big heap of clothes into neat piles of shirts, socks and towels.

Think of it like this

Sorting the hostel laundry

After a trip, everyone's clothes come back in one heap. You sort into piles by owner, then by type: shirts, socks, towels. Suddenly you notice that half the socks are missing. Sorting shows the pattern.

Watch it happen

From sticky notes to an insight

Insight: trust

Hidden fees

Unclear status

"Total jumped"

"Surprise fee"

"Did it go?"

"Paid twice"

1/4Write one observation or quote per sticky note. Dozens of them from all interviews go on the board.

The big ideas

What you need to know

One idea per note

Each sticky holds a single observation or quote, so it can move freely into any group.

Bottom up

Let groups emerge from the notes. Starting with fixed categories makes you see what you expected.

Insight is the 'so what'

A theme says what you heard. An insight says why it matters and what it means for design.

Save yourself the pain

Common mistakes

Oops

Summarising each interview separately and stopping there.

Do this instead

Mix all notes together. Patterns appear across people, not inside one interview.

Oops

Ending with theme labels like 'Payments'.

Do this instead

Write full insight sentences: 'Users pay twice because they can't tell if a payment worked.'

Your mission

Affinity map real notes

~40 min

Quick check

Which is an insight rather than just a theme?

Go deeper

1Take notes from 3-5 interviews (real or from public case studies) and cluster them into themes.
2Practice writing one clear insight statement per cluster.
3Learn to resist forcing data into themes you expected before you started.