Tool
Retro Topic Picker
Two independent draws — a retro format and a prompt — combined into one session plan.
Retro Topic Picker
Two independent draws: 8 formats and 12 prompts, giving 96 combinations. Drawing them separately is what stops a format becoming a ritual nobody hears any more.
Two independent draws — a retrospective format and an opening prompt — combined into one session plan. Eight formats and twelve prompts give ninety-six combinations.
Drawing them separately is the design. A format is a structure that shapes how the hour runs; a prompt is the specific question that opens it. Pairing them randomly means a familiar format arrives with an unfamiliar question, which is what stops a retro becoming a ritual nobody hears.
You can lock the format and redraw only the prompt, which is what most teams settle into after a few sprints.
How the Retro Topic Picker works — and why it's fair
Each draw is an independent unbiased uniform pick from its own list. Independence is the property that produces the combination count: eight formats times twelve prompts is ninety-six pairings, and a team running a retro every two weeks would take nearly four years to exhaust them.
That multiplication is worth noticing because it is why two small lists beat one large one. Ninety-six hand-written retro plans would be an enormous amount of content and would go stale as a set; twenty items drawn in pairs produce the same variety and stay individually maintainable.
The lock control makes the format draw conditional rather than removing it. With the format locked, only the prompt is redrawn, so the pool for that draw is twelve rather than ninety-six — repeats arrive roughly eight times faster, which is exactly what a team that has settled on one format usually wants.
There is no no-repeat memory here, deliberately. A retro format that comes up twice in a row is fine and sometimes preferable; a prompt that comes up twice is a reasonable prompt to revisit. Adding a cycle would create the impression that the tool is working through a curriculum, which it is not.
The format notes are the practical content. "Sailboat" is not self-explanatory, and a format name without a one-line explanation of when it works is a name rather than a suggestion — so each format carries a note about the situation it suits, which is the part a facilitator actually uses.
Neither list is filtered or weighted. Every format is as likely as every other, including the ones that suit a hard sprint and the ones that suit a good one, because the tool has no idea which kind of sprint you have had.
When the Retro Topic Picker is fair — and when it is not
What it does guarantee
- Format and prompt are drawn independently, so all ninety-six pairings are equally likely.
- Locking the format leaves the prompt draw uniform over its full twelve.
- No format or prompt is favoured or suppressed, including the ones that fit badly.
What it does not
- There is no memory, so the same format or prompt can come up in consecutive sprints.
- It has no idea how your sprint went. A format suited to a difficult sprint is as likely after a good one.
- The format notes are one person's judgement about when each works, not a measured finding.
Two worked examples
A full draw
- 8 formats × 12 prompts = 96 pairings, each with a 1-in-96 chance.
- A fortnightly team would take nearly four years to see them all.
- Consecutive identical pairings have a probability of about 1%.
Format locked, prompt redrawn
- The pool drops from 96 to 12, so repeats arrive about eight times faster.
- A fortnightly team locked to one format sees a repeat prompt within about six months.
- That is usually acceptable — a good prompt is worth revisiting.
Retros that have stopped working
The specific failure this addresses is the retro that has run Start-Stop-Continue for eleven sprints and now produces the same four cards every time. The format is not wrong; the familiarity is, and changing either half of the pairing breaks it.
New facilitators use it differently, as a source of formats they have not run before with a note about when each fits. Rotating the facilitator without also rotating the format tends to produce the same meeting with a different voice.
Teams that have inherited a retro they did not design use it to find out which formats their particular problems actually suit, by trying several. Eight formats is few enough to work through in a quarter.
For deciding who facilitates rather than what the format is, the standup order picker draws an order and the facilitator rotation use-case explains why a rotation beats a draw for that job.
Frequently asked questions
Why draw the format and prompt separately?
Because it multiplies the variety. Eight formats and twelve prompts give ninety-six combinations from twenty maintainable items.
What does locking the format do?
It stops the format being redrawn, so only the prompt changes. The pool drops from ninety-six to twelve, so repeats come around faster.
Does it avoid repeating last sprint's format?
No. There is no memory, and a repeated format is often fine — adding a cycle would imply the tool is working through a curriculum.
Does it know what kind of sprint we had?
No. A format suited to a difficult sprint is exactly as likely after a good one, because the tool has no information either way.
Are the format notes based on research?
No, they are practical judgement about when each format tends to work. Treat them as a starting point rather than a finding.
Can I add our own formats or prompts?
No, the two lists are fixed. What you can do is lock the format and ignore the drawn prompt, using the tool only for the half you want.
How many combinations are there?
Ninety-six with the format unlocked, twelve with it locked. Both numbers are shown on the page.
Should I use the drawn prompt as the whole retro?
No — it is an opening. The format is what structures the rest of the session; the prompt gets it started.
Is one format better than the others?
No, and the tool does not rank them. Each note says when that format tends to fit, which is a different claim from being better.