4Ls retrospective template
The four Ls of a retrospective — liked, learned, lacked, longed for — what each is really asking, and how to run it so answers turn into changes.
8 questions · Surveys
The template
Liked
What was good. Ask it first because it is the easiest to answer, and it establishes that this is not purely a post-mortem.
- What went well that we should deliberately do again?
- Who made this easier, and how?
Learned
New knowledge. The most valuable answers are the ones that cost something to find out.
- What do you know now that you did not know at the start?
- What surprised you?
Lacked
What was missing. This is where the actionable material usually is.
- What did we not have that we needed — information, people, time, a decision?
- What slowed you down more than once?
Longed for
The wish list. Deliberately unconstrained, because the gap between "lacked" and "longed for" is where the interesting ideas live.
- What would you want next time, even if it is not realistic?
- If we ran this again from the start, what would you change first?
What this one assumes
There is more than one way to run this. Here is what we assumed, so you can tell quickly whether it fits — and what to change if it does not.
- It runs at the end of something — a project, a quarter, a release — not on a standing schedule with nothing to look back on
- Everyone answers before the discussion, so the loudest voice does not set the frame
- Someone owns turning the output into two or three changes, with names against them
- Nothing said in it is carried into anyone’s performance review
Why 4Ls rather than went-well / went-badly
A two-column retrospective collapses into a list of complaints and a list of reassurances. The 4Ls split both halves in a way that produces more useful answers.
On the positive side, liked and learned separate what to repeat from what you now know — two different outputs. On the negative side, lacked and longed for separate the gap you can close from the thing you wish existed. That second split matters most: "lacked" produces a backlog, "longed for" produces ideas, and running them together produces a complaint.
How the template is built
Liked goes first because it is the easiest question in the set and it sets the tone. The second question — who made this easier — is the one that surfaces the quiet contributor who will not nominate themselves.
Learned asks what changed in people's understanding. "What surprised you" does more work than it looks: surprise marks the place where the team's model of the work was wrong, which is exactly what is worth writing down.
Lacked is where most of the actions come from. Asking what slowed you down more than once is deliberate — a one-off problem is bad luck, a repeated one is a process.
Longed for is explicitly allowed to be unrealistic. Half of these will be impossible and the other half will be the only genuinely new ideas in the session.
How to run it
Collect answers before the session, not during it. A retrospective where people think out loud becomes a retrospective where the first and loudest answer frames everything after it.
Group the responses before you discuss them. Three people independently saying the same thing is the strongest signal in the whole exercise, and it is invisible if you work through answers one person at a time.
Leave with two or three changes and a name against each. A retrospective with no owner produces the same answers next time, and the fastest way to stop people answering honestly is to show them it changes nothing.
What we would change, and when
You will also see this as Loved, Loathed, Learned and Longed for. The stronger emotional pair gets more honest answers from a team that already trusts each other, and more guarded ones from a team that does not. Start with the version above and swap in loathed once people are answering freely.
For a retrospective after something that went badly, add a fifth question before the others: what did we get right that we should not lose in the rewrite. Teams discard good practice along with the bad when they are reacting to a failure.
For a long project, run just lacked at the midpoint. It is the only one of the four that is worth answering while there is still time to act on it.
Run this one, then change it
Load it into Crewmojo as a starting point and reshape the questions, the timing and who is involved around how your organisation actually works.