Most teams pick a retro template once and never change it. "Went Well / To Improve / Action Items" becomes the default forever, and after a few months the team is running on autopilot — writing the same generic cards without really thinking.
The template matters less than people think, but changing it matters more. Here's a practical framework for picking the right one.
The simple decision: what happened this sprint?
Don't overthink template selection. Match the template to the sprint's context:
| If your sprint was... | Use this template |
|---|---|
| Normal, nothing unusual | Went Well / To Improve / Action Items |
| Rough, people are frustrated | Mad / Sad / Glad |
| Great, team is energized | Start / Stop / Continue |
| After a big launch or incident | 4Ls (Liked, Learned, Lacked, Longed For) |
| The team feels stuck | Sailboat (Wind, Anchors, Rocks) |
| You've used the same template 5+ times in a row | Anything different |
That's the whole framework. Pick based on context, rotate periodically, move on.
The templates you should actually know
Went Well / To Improve / Action Items
Columns: What Went Well | What To Improve | Action Items
The default for a reason — it's simple, balanced, and works for any team. Start here if you're new to retros or facilitating for the first time.
When it stops working: After 6-8 sprints, people start writing the same cards on autopilot. That's your signal to switch.
Example cards:
- Went Well: "The new test fixtures cut our setup time in half"
- To Improve: "Requirements for the search feature changed mid-sprint, causing 2 days of rework"
- Action Items: "PM and lead dev will review acceptance criteria together before sprint starts"
Start / Stop / Continue
Columns: Start Doing | Stop Doing | Continue Doing
More action-oriented than Went Well / To Improve. Every card implies a behavior change, which makes the transition to action items smoother.
When to use it: When you want the team to think in terms of concrete changes, not just observations.
Watch out: The "Stop" column can feel confrontational. Frame it as "what's no longer serving us?" rather than "what's wrong?"
Example cards:
- Start: "Pairing on complex stories — the auth module went smoother when Alex and Jordan paired"
- Stop: "Scheduling meetings during 9-11am focus blocks"
- Continue: "Async standups in Slack on meeting-heavy days"
Mad / Sad / Glad
Columns: Mad (frustrations) | Sad (disappointments) | Glad (celebrations)
An explicitly emotional format. This is the right choice when the team has been through something hard — a production incident, a missed deadline, organizational changes — and needs to process feelings before they can think constructively.
When to use it: After rough sprints when morale is low.
Facilitator tip: Start with Glad to establish a positive foundation before opening the floor to frustrations. And use anonymous mode — people are more honest about anger and sadness when they're not attributed.
Example cards:
- Mad: "We were on-call for a system we didn't build and don't understand"
- Sad: "We missed the sprint goal despite working evenings"
- Glad: "The team supported each other without blame when things went wrong"
4Ls (Liked, Learned, Lacked, Longed For)
Columns: Liked | Learned | Lacked | Longed For
A learning-oriented format that works well after milestones. "Learned" captures knowledge gained (positive or negative). "Longed For" surfaces aspirations — things the team wishes they had.
When to use it: After launches, major features, or significant sprints. Every 8-10 retros as a change of pace.
Why it's good: The distinction between "Lacked" (concrete gaps) and "Longed For" (aspirations) produces different kinds of insight. "We lacked test environments" vs "We longed for dedicated time for technical exploration" — both are valid but lead to different conversations.
Example cards:
- Liked: "The RFC process gave everyone time to think before committing to a design"
- Learned: "Our API tests are too coupled to implementation details — they break on unrelated changes"
- Lacked: "A clear definition of done for the onboarding epic"
- Longed For: "Dedicated tech debt time built into sprint planning"
Sailboat
Elements: Wind (what's propelling us forward) | Anchors (what's holding us back) | Rocks (risks ahead)
A visual metaphor that works well for teams that feel stuck or need to think forward. The "Rocks" element is especially useful — it surfaces risks before they become problems.
When to use it: When the team needs a fresh perspective, or when you want to balance looking backward (anchors) with looking forward (rocks).
Example cards:
- Wind: "New CI pipeline cut deploy time from 20 minutes to 4"
- Anchors: "Tech debt in the billing service means every feature takes 2x longer"
- Rocks: "We're dependent on the platform team and they're already overcommitted"
Templates you probably don't need
There are dozens of retrospective templates out there — Speed Car, Oscar Awards, Superhero, Hot Air Balloon, Star Wars themed. Most of them are the same 3 questions (what's good, what's bad, what should change) dressed up in different metaphors.
The metaphor can help if your team responds to visual thinking. But if you find yourself spending 5 minutes explaining how a speed car parachute relates to sprint retrospectives, the metaphor is getting in the way of the actual conversation.
Stick to 4-5 templates you know well. Rotate between them. That's enough variety to prevent autopilot without the overhead of constantly learning new formats.
The rotation strategy
Change templates every 4-6 retros (roughly every 2-3 months). Here's a practical rotation:
- Sprints 1-5: Went Well / To Improve / Action Items (build the foundation)
- Sprint 6: Mad / Sad / Glad (emotional check-in)
- Sprints 7-10: Start / Stop / Continue (behavior-focused)
- Sprint 11: Sailboat (visual change of pace, forward-looking)
- Sprints 12-15: Went Well / To Improve (return to foundation)
- Sprint 16: 4Ls (deep reflection)
- Repeat with variations
Signs you need to switch templates now:
- Cards feel generic and repetitive
- Team says "same issues every time"
- Energy is low from the start
- People are clearly on autopilot
Let the team choose
Here's an engagement hack that costs nothing: at the end of each retro, let the team vote on next sprint's template. "Should we do Start/Stop/Continue or Sailboat next time?"
When people choose the format, they feel ownership over the ceremony. Ownership drives engagement.
Try NextRetro free — 17+ built-in templates with facilitation guidance. Switch templates with one click.
Last Updated: February 2026
Reading Time: 6 minutes
