Worked Well, Kinda Worked, Didn't Work Retrospective Template

Also known as: Worked Well Kinda Didn't · Went Well, Kinda, Didn't · Good, Meh, Bad

Most of a sprint isn't great or terrible. This format gives the 'kinda' its own column.

  • ⏱ 30–45 min
  • 👥 3–12 people
  • 🧩 3 columns
  • Free · no sign-up

Worked well

What clearly worked?

Things that clearly worked and should continue as they are.

Kinda worked

What half worked?

Things that half worked: promising, but with friction or mixed results.

Didn't work

What didn’t work?

Things that failed, got in the way or should be dropped.

What is the Worked Well, Kinda Worked, Didn't Work Retrospective?

Worked Well, Kinda Worked, Didn't Work is a three-column retrospective that sorts the sprint by how well things went. It looks like a classic 'went well / didn't go well' retro, with one important difference: a middle column for the things that kinda worked.

That middle column is where most improvement lives. New practices rarely work perfectly the first time: a new standup format, a review rotation, a testing approach. In a two-column retro they get forced into 'good' or 'bad' and are either kept unchanged or dropped. Here the team can say 'this is promising, let's tune it'.

The format is simple enough to need no explanation, which makes it a good alternative to Start Stop Continue for regular sprint retros, especially for teams that experiment a lot and want to review those experiments honestly.

When to use Worked Well, Kinda Worked, Didn't Work (and when not to)

✅ Use it when

  • ▸ After a sprint in which the team tried new practices or tools.
  • ▸ For regular sprint retros when you want something as simple as Start Stop Continue but more nuanced.
  • ▸ To review the action items from previous retros: which ones worked, kinda worked or didn't?
  • ▸ When the team tends to see things in black and white.
  • ▸ With new teams, since the columns are self-explanatory.

⚠️ Avoid it when

  • ▸ When the team needs to talk about feelings or conflict (use Mad Sad Glad).
  • ▸ When you need to plan ahead or review risks; the format only looks back.
  • ▸ If almost nothing changed during the sprint, as the middle column will stay empty.

How to run a Worked Well, Kinda Worked, Didn't Work Retrospective step by step

Total time: about 40 minutes for a team of 3–12 people. Adjust the timeboxes to the size of your team and sprint.

  1. 1

    Set the stage · 3 min

    Review last retro's action items. They make good first cards: did each one work, kinda work or not work?

  2. 2

    Silent writing · 7 min

    Everyone writes in the three columns. Encourage people to use 'Kinda worked' generously, since that's where the interesting conversations are.

  3. 3

    Read Worked well · 5 min

    Read and group the positive cards quickly. Confirm the team wants to keep these as they are.

  4. 4

    Dig into Kinda worked · 10 min

    For each group ask: what part worked, what part didn't, and what small change would move it to 'worked well'?

  5. 5

    Read Didn't work and vote · 7 min

    Read the failures and group them. Then three votes each across Kinda and Didn't work.

  6. 6

    Decide: tune, drop or fix · 8 min

    For each top item decide one of three things: tune it, drop it or fix it. Write the decision as an action with an owner.

Worked Well, Kinda Worked, Didn't Work questions and prompts for each column

Read a couple of these aloud if the board stays empty after the first minutes of writing.

Worked well

  • • What went exactly as planned?
  • • What would you recommend to another team?
  • • Which change from last sprint paid off?

Kinda worked

  • • What helped, but took more effort than it should have?
  • • Which new practice is good in theory but awkward in reality?
  • • What worked for some of us but not for others?

Didn't work

  • • What would you not repeat?
  • • Which experiment should we stop?
  • • What caused rework or wasted time?

Worked Well, Kinda Worked, Didn't Work retrospective example

A mobile team after a sprint in which they tried async standups and a new testing approach. Here is what their board looked like at the end of the writing phase.

Example Worked Well, Kinda Worked, Didn't Work Retrospective board in Giftro with the columns Worked well, Kinda worked, Didn't work filled with sticky notes
Example Worked Well, Kinda Worked, Didn't Work board

Worked well

  • Snapshot tests caught two UI regressions
  • Releasing behind a feature flag
  • The new on-call handbook

Kinda worked

  • Async standups: less time, but blockers get noticed late
  • Pairing on bugs, great but hard across time zones
  • Release notes are written, but late

Didn't work

  • Estimating in hours
  • Testing on only one Android device
  • Big PRs at the end of the sprint

Action items they agreed on

  • ☑️ Keep async standups but add a 'blocked' emoji that pings the lead (owner: Mia)
  • ☑️ Switch back to story points for estimation (owner: team)
  • ☑️ Add two more Android devices to the test matrix (owner: Kai)

Common mistakes and facilitation tips

Everything goes in 'kinda'

If the middle column becomes a safe place for everything, push back: 'is this closer to worked or didn't?'.

Dropping experiments too early

Something in 'Kinda worked' after one sprint deserves a tweak, not a funeral. Give promising practices another sprint.

No decision per item

Each discussed item should end with tune, drop or fix. Otherwise the same cards come back next time.

Forgetting what worked

Spend at least a minute on 'Worked well' so the team knows what not to change.

Variations and related templates

Worked Well, Kinda Worked, Didn't Work retrospective FAQ

What is the Worked Well, Kinda Worked, Didn't Work retrospective?

It is a three-column format in which the team sorts the sprint into what worked well, what kinda worked and what didn't work. The middle column captures promising but imperfect practices.

Why add a 'kinda worked' column?

Most new practices half work at first. A middle column lets the team tune them instead of keeping or dropping them outright.

How is it different from Start Stop Continue?

Start Stop Continue asks what to change. Worked Well, Kinda, Didn't asks how well things went, and then decides what to tune, drop or fix.

How long does it take?

Around 30 to 45 minutes for a team of up to ten people.

Can I run it online for free?

Yes. Use the template on this page to open a Giftro board with the three columns. Share the link and your team joins without an account.

Run your Worked Well, Kinda Worked, Didn't Work retro now

Free, real-time and no account needed. Share the link and your team joins in seconds.

Related retrospective templates

Start Stop Continue retrospective template

The classic three-column retro: what to start doing, stop doing and keep doing. Fast, clear, action-oriented.

KALM retrospective template

Keep, Add, Less, More: adjust the volume of your practices instead of just starting or stopping them.

Starfish retrospective template

Five-column format: Keep, Start, Stop, More Of, Less Of. The most precise retro format for scrum teams that want nuanced action items.

I Like, I Wish, What If retrospective template

A constructive feedback format from design thinking: what you like, what you wish were different and what if we tried…

4Ls retrospective template

Liked, Learned, Lacked, Longed for. A balanced retro that captures lessons and what's missing, not just problems.