Format 1: Start, stop, continue
| Start doing | Stop doing | Continue doing |
|---|---|---|
| Write the current state of a case before handing it to another region. | Adding a second reviewer to low-risk changes on a Friday afternoon. | Naming an owner in the first line of every handover note. |
| Ask for the deadline before promising a date. | Discussing priorities in a channel where half the team is asleep. | Running the 15-minute check-in on Monday and Thursday. |
Format 2: The 4 Ls
| Liked | Learned | Lacked | Longed for |
|---|---|---|---|
| The weekly demo made progress visible without a status report. | Two teams had different definitions of done, which we found out in week three. | A shared place to record decisions that everyone could read. | One page that lists who decides what, so we stop asking in meetings. |
| The on-call rotation held when two people were away. | We can decide in a day when the decision owner is named up front. | Time to test a change before the next release. | A named person to review the experiment with us. |
Action tracker
| Action | Owner | First step | Review date | Status |
|---|---|---|---|---|
| Write the current state before every handover | Priya | Add the field to the handover template | 3 Oct | In progress |
| Publish the decision list | Tom | Draft the first ten decisions | 10 Oct | Not started |
What you'll leave with
The group selects one improvement experiment based on specific observations from a defined period of work.
Before the session
- Define the period or project under review and share it in the invitation. Ask for specific events rather than general assessments of people.
- Prepare a shared document with three columns: useful moments, friction, next experiment. Decide who can see the notes and how long they will be kept.
- Check whether the group is ready to discuss the work together. Handle sensitive individual concerns through an appropriate private route.
How to fill it in with your team
Create a common frame
State the time period and purpose. Say: “We are examining how the work happened and what we can change. Describe events and effects; avoid guessing other people’s motives.” Explain who will see the notes and offer a pass option.
Reflect independently
Ask each person for two useful moments and two points of friction. For each, include the event, its effect and a question. Example: “The acceptance criteria changed after development began; we repeated two tasks; how could we check them earlier?”
Share and make sense
Invite one observation at a time, then group related notes. Ask clarifying questions before discussing causes. Keep different experiences visible rather than forcing agreement. Separate a known fact from a possible explanation.
Choose a useful focus
Ask each person to mark one issue that is both important and within the team’s influence. Discuss the leading issue. If tied, choose the one the group can test most cheaply. Write a narrow question that can guide action.
Design the next experiment
Generate two or three options. Choose one and record the change, owner, duration, expected signal and review date. Example: “For the next five work items, the requester and developer will check acceptance criteria together before starting. Track clarification requests after work begins.”
Close with a commitment
Ask the owner to restate the first step and invite concerns. Each person writes one way they can support the experiment. Schedule the review before ending. Keep other ideas visible without committing to all of them.
Participant worksheet
Copy this table into a document, or print the worksheet from the bundle. Share only what you choose.
| Item | event | effect | what we want to keep |
|---|---|---|---|
| Period or project under review | |||
| Useful moment | event | effect | what we want to keep |
| Friction | event | effect | question to investigate |
| Our selected issue and why it matters | |||
| Experiment | owner | first step | duration |
| Starting situation and expected signal | |||
| How and when we will collect evidence | |||
| Review date | keep, change or stop | reason |
Debrief questions
- What did we observe, and what are we assuming?
- Is the experiment within our control?
- What result would make us change our approach?
Adapt it for your group
- For a quiet group, collect notes before the call and read them without attribution only if contributors agreed.
- For more than ten participants, split reflection into small groups and allow extra synthesis time; do not promise a deep 45-minute review with a large group.
- If the discussion becomes personal, pause and return to the event and its effect. A retrospective is not a substitute for resolving a serious complaint.
After the session
Open the next retrospective by reviewing the previous experiment. Share the evidence, including unexpected effects. Keep, revise or stop it before creating a new action.
Questions people ask
Which retrospective format should I use?
Use start stop continue when the team already knows each other and needs to decide one change quickly. Use the 4 Ls when the period was difficult or the team is new, because liked and longed for make it easier to say what was missing.
How long should a retrospective take?
Forty-five minutes for a small team and a two-week period. Half the time goes to understanding what happened and half to designing one change with an owner.
How many actions should come out of a retrospective?
One. A tracker with five actions from every retrospective tends to have none done by the next one. Carry an unfinished action forward before adding a second.
What if the same problem comes up every time?
Treat the repetition as evidence rather than a complaint. Write the obstacle as a condition to test, name who can approve the change, and review it at a fixed date.
Should the manager be in the room?
Yes, if they can hear the obstacle without defending the decision. If their presence stops people speaking, ask them to join for the action-setting part only.
How to use the files
Print the PDF, edit the Word file in Word or Google Docs, or open the CSV or XLSX in a spreadsheet. Open files need no email.
Can I use this with paying clients?
Yes. Use and adapt this original Superglue kit in your own training, including paid client sessions. Please share this page rather than redistributing the unmodified kit. Read the full usage terms.