# Interview synthesis sheet

A one-page sheet for turning customer interview notes into findings: coded observations, clusters, a findings table and a list of what you do not know.

> **How to use this.** Fill in one sheet per round of interviews, soon after the round ends. Write observations before interpretations, and keep the raw notes untouched in their own file.

## 1. The round

- **Question this round was meant to answer:**
- **Interviews in this round:**
- **Who you spoke to and how you found them:** say if most came from one source, such as friends or one introducer.
- **Who observed or took part in the analysis:**
- **Consent:** [ ] Each participant agreed to the interview. [ ] Each participant agreed to any recording.
- **Where the raw notes are stored, and who can open them:**

## 2. Observations

One observation per row, in plain words, with no interpretation. Code each one:

- **D** is something the person did.
- **S** is something the person said about the past.
- **P** is a prediction of what the person would do.

| ID | Interview | Code (D, S or P) | Observation |
|---|---|---|---|
| 1 | | | |
| 2 | | | |
| 3 | | | |
| 4 | | | |
| 5 | | | |

## 3. Clusters

Group observations that belong together and give each group a plain name.

| Cluster name | Observation IDs | Interviews covered | Single case or pattern? |
|---|---|---|---|
| | | | |
| | | | |
| | | | |

## 4. Findings

| Finding (one plain sentence) | Evidence (interviews, and how many are D, S or P) | Confidence (high, medium, low) and why | What it changes |
|---|---|---|---|
| | | | |
| | | | |
| | | | |

## 5. What you do not know

- **Interviews that disagreed with a finding, and how:**
- **Questions the interviews did not answer:**
- **People you have not yet spoken to, who could change a finding:**

## 6. Decisions this informs

| Decision | Which finding it rests on | Next test or step | Owner |
|---|---|---|---|
| | | | |

## 7. Example of a filled row

**Illustrative example, not a real company.** A fictional shift-scheduling tool for small cafes.

| Finding | Evidence | Confidence and why | What it changes |
|---|---|---|---|
| Owners fix schedule conflicts by text message on the day. | Six of eight interviews. Five D, one S. | High: most are things people did. | Test a same-day change feature before a planning feature. |

## 8. Checks before you share the sheet

- [ ] Every finding points to specific interviews.
- [ ] Predictions are not counted as behavior.
- [ ] Disagreeing interviews are listed.
- [ ] Names and contact details are not on the sheet.
- [ ] Contact details are deleted when you no longer need them.
- [ ] The sheet is saved with the date, next to your conversation log.

---

World Enterprise, https://worldenterprise.org/. Last updated 2026-10-10.