Retro Prompts
Fresh prompts for your next sprint retrospective. Pick a framework, shuffle the deck, run the retro.
335+ prompts across 10 retro frameworks. Free, no sign-up.
Pick a framework
The classic. Three columns, fast to run, hard to argue with.
When to use it: When you want a fast, action-oriented retro with no warm-up. Works for any team, any sprint length.
Start
What does this team need that we don't currently have?
10 prompts in this bucket
Stop
What's the thing that drains energy every week?
10 prompts in this bucket
Continue
What's working that we should protect?
10 prompts in this bucket
Retros are every two weeks. Connection is daily.
Halftime is a 2-minute game your team plays together every workday. It's the daily ritual that makes the retro a check-in, not a rescue.
Free for teams up to 8 · No credit card · 30-second setup
More free tools
How to run a great sprint retrospective
A sprint retrospective is a recurring meeting where the team reflects on the last sprint and decides what to change. The shape of the meeting matters less than the consistency. Teams that run a 30-minute retro every two weeks outperform teams that run a 90-minute retro once a quarter, every time.
The job of the facilitator is to create enough psychological safety that the team can name uncomfortable things, and enough structure that the meeting produces actionable change. Good prompts do both. They name the topic explicitly so it's easier to engage with, and they vary often enough that the team doesn't fall into scripted answers.
How long should a retro be?
For a two-week sprint, 45 minutes is a good default. For a one-week sprint, 25 minutes. Longer is rarely better. If your retros need more time, that usually means you're using the meeting to do work that should have happened earlier, like surfacing blockers in real time.
Who facilitates?
Rotate. Not just because it's fair, but because every facilitator brings a different style and the team gets to learn from all of them. Engineering managers shouldn't default to facilitating their own team's retros — it makes it harder for the team to bring up management-related issues.
What if people don't speak up?
The fastest fix is to make the first round silent and written. Give everyone three minutes to write notes against each prompt before anyone speaks. People who think before they talk get a fair shot, and you sidestep the loudest-voice problem entirely.
Should you use the same framework every time?
No. Even good frameworks go stale. Rotate every few sprints — even switching between Start/Stop/Continue and 4Ls is enough to keep answers fresh. When the team starts answering on autopilot, the framework has worn out.
When to use which framework
- Start, Stop, Continue. When you want a fast, action-oriented retro with no warm-up. Works for any team, any sprint length.
- 4Ls. When the team needs to surface feelings, not just facts. Good after a hard sprint or a big release.
- Mad, Sad, Glad. When energy is low, when there's been conflict, or when a team is forming. Use sparingly — it's high-trust.
- Rose, Bud, Thorn. When you want a balanced view — celebration plus problems plus emerging things — without it feeling heavy.
- Sailboat. Mid-project or quarterly retros. When you need to talk about momentum, drag, risk, and direction together.
- KALM. When the team is mostly happy and you want to fine-tune rather than overhaul. Less binary than start/stop.
- DAKI. When you want every prompt to produce a clear next action. Best when the team commits to one item per category.
- Three Little Pigs. When you want to talk about durability — what would survive turbulence, what wouldn't. Great for resilience and risk.
- Glad, Sad, Mad, Wondering. When the team has unanswered questions floating around and you want to name them. The 'Wondering' column is the magic.
- Lean Coffee. When you don't want to prescribe categories at all. Generates topic prompts the team can vote and dot-vote on.
The frameworks, briefly
Start, Stop, Continue
The classic. Three columns, fast to run, hard to argue with.
Categories: Start, Stop, Continue.
4Ls
Liked, Learned, Lacked, Longed for. More texture than 3-column.
Categories: Liked, Learned, Lacked, Longed for.
Mad, Sad, Glad
Emotional check-in. Surfaces the stuff data can't see.
Categories: Mad, Sad, Glad.
Rose, Bud, Thorn
What's blooming, what's growing, what's prickly. Garden metaphor, real insight.
Categories: Rose, Bud, Thorn.
Sailboat
Wind, anchors, rocks, island. Visual metaphor that travels well.
Categories: Wind, Anchors, Rocks, Island.
KALM
Keep, Add, Less, More. Subtler than Start/Stop/Continue.
Categories: Keep, Add, Less, More.
DAKI
Drop, Add, Keep, Improve. Action-first, easy to translate to a backlog.
Categories: Drop, Add, Keep, Improve.
Three Little Pigs
House of straw, sticks, bricks. Fragile, sturdy, unbreakable.
Categories: House of Straw, House of Sticks, House of Bricks.
Glad, Sad, Mad, Wondering
Mad/Sad/Glad with a curiosity dimension. Surfaces uncertainty, not just emotion.
Categories: Glad, Sad, Mad, Wondering.
Lean Coffee
Open-agenda format. The team picks the topics in real time.
Categories: Topics to surface.