United We Transform | Facilitator run sheet

The 4Ls QA Retrospective

A cross-functional software team reviews a release or sprint through a quality assurance lens using the 4Ls framework: Liked, Lacked, Learned, and Longed For. Participants write individual observations, cluster them into themes, and select concrete quality improvements.

45 minutes | 4 to 16 people

Choose this exercise

Purpose: Collect feedback across engineering, product, and QA roles on software quality, defect handling, and testing workflows to agree on process improvements.

Use it when: Use at the end of a sprint, release cycle, or major testing phase when team members need to examine how defects were caught, prevented, or missed.

Skip it when: Skip if the team has not shipped or tested code recently, or if testing and deployment workflows are not part of the scope being reviewed.

Group arrangement: Individual writing followed by whole-group clustering and paired action planning

Timing: UWT plan allocates 45 minutes to allow sufficient time for silent generation, thematic clustering, and action items.

Materials and tools

Before participants arrive

  1. Draw four columns on the board labeled: Liked, Lacked, Learned, and Longed For.
  2. Place sticky note pads and markers at each participant seat.
  3. Ensure everyone understands the scope (for example, the last release or sprint).

Say this to open

Today we are running a 4Ls retrospective with a specific focus on software quality. Quality is a shared responsibility across QA, development, and product. Take the next few minutes to write your observations under Liked, Lacked, Learned, and Longed For, specifically looking at how we tested, prevented defects, and handled releases. Write one clear thought per sticky note.

How to run it

  1. Introduce and frame quality scope 5 minutes

    State the specific sprint, release, or feature milestone under review. Read the definitions of the four columns: Liked (what went well in testing and quality), Lacked (what gaps, missing tools, or test data slowed us down), Learned (what technical or workflow discoveries we made), and Longed For (what testing tools, environments, or practices we wish we had). Answer any clarifying questions.

  2. Silent generation 8 minutes

    Participants work individually and silently. Everyone writes brief, concrete statements on sticky notes, one idea per note, categorizing each into one of the four Ls. Example neutral note: 'Liked: Automated regression caught bug #412 before staging.' Participants may write as many notes as they need or write none.

  3. Post and cluster notes 12 minutes

    Invite participants to post their notes in the matching columns on the board. The facilitator or a designated participant reads notes aloud briefly and merges duplicate or closely related items into visible clusters. If a note is unclear, ask the author for a single-sentence clarification without debating the topic.

  4. Prioritize themes with dot voting 5 minutes

    Give each participant 3 dot stickers. Participants place dots on the clusters or individual cards they believe are most urgent to address for team quality. Voting is individual and silent.

  5. Action planning and debrief 15 minutes

    Focus on the 2 or 3 clusters that received the most votes. Form pairs or talk as a whole group to draft concrete action steps. Each action must name a clear task, an owner, and a target date. Record these actions on the board.

Debrief questions

Finished output: A grouped 4Ls board with prioritized quality themes and an agreed list of 2 to 3 documented action items with designated owners.

Remote

Create an online whiteboard with four columns: Liked, Lacked, Learned, Longed For. Set a timer for 8 minutes of silent card writing. Group cards collaboratively on screen, then use digital voting pins (3 per person) to identify the top discussion topics. Run action planning via screen share.

Hybrid

Remote participants enter cards into a shared digital board while on-site members either type into the same board via laptops or have a co-facilitator mirror physical cards onto the shared screen. Project the digital board in the physical room so both groups view the exact same clusters and votes.

Access and participation choices

If the session gets stuck

The session turns into developer-versus-QA finger pointing over bugs that slipped into production.
Intervene immediately. Remind the group that quality is an organizational process, not an individual fault. Reframe specific blame statements into systemic process gaps under 'Lacked' or 'Longed For.'
Cards are vague, such as writing only 'bad testing' or 'communication.'
Ask the writer or room for a specific neutral example: 'Can you describe the specific moment that happened so we understand the process gap?'

Terms used here

Sources and adaptation

Observed in EasyRetro template directory, attributed to Quality Logic.

Retained the core 4Ls QA-focused template structure from Quality Logic / EasyRetro (Liked, Lacked, Learned, Longed For with a spotlight on QA and product quality). Added concrete stage timings, silent generation, clustering rules, dot voting, and action planning steps suited for cross-functional software teams.

See our editorial policy for AI use, sourcing and corrections.