How to Run a 'What's Working' Retro

The traditional retrospective format goes something like this: what went well, what didn’t, what should we change. On paper, it sounds balanced. In practice, “what didn’t go well” almost always takes over the room, and “what should we change” becomes a triage list of complaints.

The “What’s Working” retro is a different animal. It’s not a sanitized version where problems are swept under the rug — it’s a deliberately appreciative format that produces more actionable insight precisely because it starts from strength.

Here’s how to run one.

Why Start With Strength?

Before the mechanics: a brief case for why this approach works.

When teams begin with what went wrong, they anchor to problems. The energy in the room tilts toward frustration, blame-adjacent thinking, and defensiveness. Even well-intentioned teams can find themselves in a loop of surface-level complaints rather than root-cause learning.

When teams begin with what went well, something different happens. People relax slightly. They remember that the team is capable of good work. And — crucially — the conversation about what enabled the good work often surfaces more actionable information than post-mortems of failures.

High-performing teams tend to analyze their successes as rigorously as their failures. This retro format builds that habit.

The Format (90 Minutes)

Part 1: Open (10 minutes)

Start with a single check-in question, one per person. Keep it brief.

“In one sentence: what’s one thing you’re genuinely proud of from this sprint/quarter/project?”

This is a warm-up, not a full reflection. The goal is to activate appreciative attention before the main discussion.

Part 2: Discover — What Worked (25 minutes)

Distribute sticky notes. Ask each person to silently write responses to two prompts:

A moment of peak performance: Describe a specific moment in this sprint/project where you felt the team was operating at its best. What was happening? What made it possible?
A quiet win: Describe something that worked that didn’t get enough recognition — a decision, a behavior, a process, a person’s contribution.

After 8–10 minutes of silent writing, post notes on a shared board and read them aloud. Cluster related themes. Notice patterns.

Part 3: Analyze — What Enabled It (20 minutes)

This is the analytical heart of the retro. For each cluster of “what worked,” ask:

Why did this work? What conditions made it possible?
Is this reproducible? What would need to be true for this to happen consistently?
Who deserves credit that hasn’t been named?

Assign a notetaker to capture the conditions, not just the outcomes. These conditions become your team’s operating theory — the things worth protecting and amplifying.

Part 4: Address — What Needs Attention (20 minutes)

This is not an elimination of the “what didn’t work” conversation — it’s a reframe. Ask:

“Given what we know is working, what one or two things most need our attention to protect and improve that?”

Framing problems as threats to what’s working changes the energy. Instead of venting about what’s broken, the team is protecting something valuable. The action items that emerge from this framing tend to be more targeted and better owned.

Part 5: Commit (15 minutes)

For each action item, agree on:

Who will own it
What “done” looks like
When it will be reviewed

End the retro by going around the room: one word or phrase that captures how each person is leaving the conversation.

Facilitation Tips

Protect the silence. The individual reflection time in Part 2 is critical. Don’t rush it. Eight to ten minutes of quiet thinking produces much richer data than a free-for-all discussion.

Name the namer. When a sticky note recognizes a team member’s contribution, ask who wrote it and have them say it directly to that person. This moment of named appreciation is often the most memorable part of the retro.

Resist the urge to solve. In Part 3, the goal is analysis, not action planning. Keep the group in inquiry mode — asking what enabled success — before jumping to solutions.

Keep the ratio. Plan for roughly equal time on Parts 2–3 (strength analysis) and Part 4 (problem framing). If the problem conversation starts to dominate, gently redirect: “Let’s keep that on the parking lot and come back to it in Part 4.”

Adapting for Async or Remote Teams

For remote or distributed teams, run Parts 1–2 asynchronously in a shared document or Miro board 24 hours before the live session. Ask team members to post their sticky notes before the call so the facilitator can pre-cluster themes.

The live session then starts at Part 3 — with richer data and a warmer room.

What You’ll Notice

After running a few of these, most teams report two lasting changes.

First, they get better at naming what’s working as it happens — not just retrospectively. The retro format trains a muscle of noticing that carries into daily work.

Second, the action items are better. They’re more specific, more owned, and more connected to the team’s actual identity and capabilities. Fixing a problem that threatens something you’re proud of is a very different motivation than just eliminating a complaint.

Try it once. The format is in the Toolkit.