Rating scales
Rate each risk twice, from 1 to 5. Score = likelihood × impact.
| Rating | Likelihood: before the failure date | Impact: on the goal in the plan brief |
|---|---|---|
| 1 | Hard to imagine before the failure date. | A nuisance. We would absorb it and the goal holds. |
| 2 | Could happen, but nothing points to it yet. | Costs time or goodwill. The goal still holds. |
| 3 | As likely as not. | Part of the goal is missed or late. |
| 4 | Something already points to it. | Most of the goal is missed or late. |
| 5 | It is happening now, or it happened in a similar plan. | The plan fails or has to stop. |
Score bands and what each band needs
The bands order the group’s attention. They are not a forecast or a formal risk assessment.
| Score | Band | What the risk needs before the session ends |
|---|---|---|
| 15–25 | High | A prevention action, an early warning sign, one owner who said yes and a due date before the plan starts or before the first milestone. |
| 8–14 | Medium | A prevention action, an early warning sign, one owner who said yes and a due date. |
| 1–7 | Low | An early warning sign and one person who watches for it. No action unless the sign appears. |
Failure lens cards
Eight prompts for the silent writing step. Use them when people get stuck, not as boxes to fill.
| Card | Lens | Two questions about the failed plan |
|---|---|---|
| A | People | Who left, was absent or was overloaded at the worst moment? Whose agreement did the plan need that it never had? |
| B | Time | Which date clashed with something we already knew about? Which step took twice as long as planned? |
| C | Money and resources | What ran out: budget, rooms, licences or equipment? What did we assume someone else would pay for? |
| D | The people it was for | Why did they not take part, or take part and change nothing? What had they heard about it before it started? |
| E | Managers and sponsors | Who stopped supporting it, and when? Which decision was never made, or made too late? |
| F | Tools and places | What did not work on the day: a system, a room, a link or an access right? What did we test once, or never? |
| G | Handovers | What fell between two people or two teams? Who thought that someone else was doing it? |
| H | Outside events | What changed outside the team: a peak period, a reorganisation or a new rule? What did we see coming and decide to ignore? |
Words to say when the session drifts
Keep these near you. The pre-mortem works only when reasons are written before they are discussed.
| What happens | Words to say |
|---|---|
| The plan owner defends the plan | “Thank you. Hold that for the actions step. Right now we only collect reasons.” |
| Someone names a person to blame | “Tell me what happened rather than who did it. What did the plan miss that let it happen?” |
| Solutions arrive during writing or reading | “Write the fix in the margin and keep it for the actions step. What is the reason behind it?” |
| “That would never happen” | “It has already happened on the failure date. What would have had to be true for it to happen?” |
| Silence, or one person does the talking | “Let’s go round once more, one reason each, and passing is fine. Written reasons count the same.” |
| A reason lies outside the plan, such as pay or a reorganisation | “Is there something the plan could do about it? If not, I will park it and pass it to the plan owner.” |
Pre-mortem, kickoff risk log or after action review: which to use
The three kits make a series: the pre-mortem looks forward, the kickoff starts the work and the after action review looks back.
| Pre-mortem (this kit) | Kickoff risk log | After action review | |
|---|---|---|---|
| When | Before the plan starts, once it exists | At the first full meeting of a project | After an event, while details are fresh |
| Question | It failed: why? | What could get in the way? | What did we plan, and what happened? |
| How reasons are found | Each person writes a failure story in silence, with lens cards | The group lists risks and assumptions aloud | The group compares the plan with a timeline of what happened |
| Output | Rated risks with a prevention action, a warning sign, an owner and a revisit | A short risk and decision log beside scope and roles | Actions to keep or change, with owners |
| Time | 60 minutes | 12 minutes inside a 90-minute kickoff | 45 minutes |
Filled example: plan brief
A fictional learning team. The organisation, the people and the dates are invented for this example.
| Question | Our answer |
|---|---|
| The plan in one sentence | Run the team leader feedback programme for 60 team leaders at three sites: six cohorts of 10, each with two half-day workshops two weeks apart and a practice task in between, from Tuesday 12 January to Thursday 25 February 2027. |
| What success looks like, with a number and a date | By Wednesday 31 March 2027, at least 54 of the 60 team leaders have attended both workshops and returned the practice task, and each of their managers has logged one observed feedback conversation. |
| The failure date we imagine | Wednesday 31 March 2027. |
| Who is in the room, and who facilitates | Eight people: the learning lead (plan owner), two trainers, the programme coordinator, the operations manager, the HR business partner and two pilot team leaders. A colleague from finance facilitates and does not rate. |
| Fixed: what is not up for debate today | The budget, the two-workshop design and the start date of Tuesday 12 January 2027. |
| Already decided, and by whom | The venue at each site (operations manager) and which trainer runs which cohort (learning lead). |
| What happens to the risk table, and the revisit date | The learning lead adds the risk table to the programme plan and brings it to the revisit on Tuesday 8 December 2026. |
Filled example: failure story sheet
One person’s sheet. The two top reasons, 2 and 3, became part of R1 and R2.
| Prompt | Answer |
|---|---|
| Headline: the plan failed | Feedback programme ends with a third of team leaders never finishing. |
| What happened, in three sentences | January was our busiest month, so we were pulled out of workshops at short notice. Most of us never got to the second workshop. Our managers did not ask about the practice task, so it slipped. |
| Reasons, with the lens letter | 1 (A) Team leaders were called back to the floor mid-workshop. 2 (G, top two) Nobody covered our shifts, so we chose the floor. 3 (E, top two) My manager never asked about the practice task. 4 (D) The invitation read as if we were bad at feedback. 5 (B) The second workshop clashed with the stock count at our site. 6 (E) Nobody checked whether we gave feedback afterwards. |
| Hardest to say (on a separate slip, no name) | Some managers do not give feedback themselves, so they will not ask about ours. |
| First sign we would have seen, and when | By mid-December, fewer than half of the January places have shift cover agreed. |
Filled example: rating tally
Seven risks from 31 reasons. L and I are the middle of eight ratings. Scores: 20, 16, 12, 10, 9, 4 and 12. R3 and R7 share rank 3, so there is no rank 4.
| # | Risk: what happens, so what | Reasons | Likelihood ratings | L | Impact ratings | I | Discuss? | Score | Band | Rank |
|---|---|---|---|---|---|---|---|---|---|---|
| R1 | Team leaders are called back to the floor during workshops, so many miss part or all of a workshop. | 7 | 4 5 4 3 4 5 4 4 | 4 | 5 5 4 5 5 5 4 5 | 5 | 20 | High | 1 | |
| R2 | Managers do not ask about the practice task, so it is not done between the two workshops. | 6 | 4 3 4 5 4 4 3 4 | 4 | 4 4 3 4 5 4 4 4 | 4 | 16 | High | 2 | |
| R3 | Cohort dates clash with the January stock count at two sites, so places are cancelled late. | 4 | 3 2 4 3 3 4 3 2 | 3 | 4 4 3 5 4 4 4 3 | 4 | 12 | Medium | 3 | |
| R4 | Only one trainer can run the role-play section, so a cohort is cancelled if that trainer is away. | 3 | 2 1 4 2 2 1 4 2 | 2 | 5 5 5 4 5 5 5 5 | 5 | Discuss | 10 | Medium | 5 |
| R5 | The invitation reads like a correction, so team leaders arrive defensive or ask to defer. | 4 | 3 3 2 4 3 3 2 3 | 3 | 3 3 3 2 3 4 3 3 | 3 | 9 | Medium | 6 | |
| R6 | Site 3 has no quiet room for the role-play, so practice happens in an open office. | 2 | 2 2 1 2 3 2 2 1 | 2 | 2 2 2 3 2 1 2 2 | 2 | 4 | Low | 7 | |
| R7 | Nobody checks whether feedback conversations happen afterwards, so we cannot show that the goal was met. | 5 | 4 4 5 3 4 4 4 2 | 4 | 3 3 3 4 3 2 3 3 | 3 | Discuss | 12 | Medium | 3 |
Filled example: scored risk table
Two high, four medium and one low risk. As at Tuesday 3 November 2026 the six due dates are 17, 24, 31, 45, 66 and 87 days away.
| # | Risk | Score | Band | Prevention action | Early warning sign | Owner (said yes) | Due |
|---|---|---|---|---|---|---|---|
| R1 | Team leaders are called back to the floor during workshops, so many miss part or all of a workshop. | 20 | High | Agree written shift cover for every cohort date with the three site managers. | Fewer than 6 of 10 places in a cohort have cover agreed two weeks before it starts. | Programme coordinator | Fri 27 Nov 2026 |
| R2 | Managers do not ask about the practice task, so it is not done between the two workshops. | 16 | High | Brief each team leader’s manager in a 15-minute call on the practice task and the one question to ask about it. | Fewer than 7 of 10 practice tasks from the first cohort are back by its second workshop. | Lead trainer | Fri 8 Jan 2027 |
| R3 | Cohort dates clash with the January stock count at two sites, so places are cancelled late. | 12 | Medium | Check the January and February calendars of each site and move any cohort that clashes. | A site asks to move a cohort date. | Programme coordinator | Fri 20 Nov 2026 |
| R4 | Only one trainer can run the role-play section, so a cohort is cancelled if that trainer is away. | 10 | Medium | Have the second trainer run the role-play section at the December pilot. | The second trainer has not run the role-play section by Fri 18 Dec 2026. | Learning lead | Fri 18 Dec 2026 |
| R5 | The invitation reads like a correction, so team leaders arrive defensive or ask to defer. | 9 | Medium | Rewrite the invitation with the two pilot team leaders and send it from the operations manager. | More than 2 team leaders in a cohort ask to defer. | Operations manager | Fri 4 Dec 2026 |
| R6 | Site 3 has no quiet room for the role-play, so practice happens in an open office. | 4 | Low | None: watch only. | The quiet room at site 3 is not booked by Tue 1 Dec 2026. | Programme coordinator (watches) | |
| R7 | Nobody checks whether feedback conversations happen afterwards, so we cannot show that the goal was met. | 12 | Medium | Add one question to the managers’ monthly one-to-one notes: “Which feedback conversation did you observe?” | No observed conversation is logged for the first cohort by the end of February. | HR business partner | Fri 29 Jan 2027 |
Filled example: revisit
Tuesday 8 December 2026. Closed: R3, R5 and R6. Kept: R1, R2 and R4. Escalated: R7.
| # | Risk | Score then | Action done? | Warning sign seen? | New L | New I | New score | Change | Decision | Note |
|---|---|---|---|---|---|---|---|---|---|---|
| R1 | Team leaders called back to the floor during workshops. | 20 | Partly | No | 2 | 5 | 10 | -10 | Keep | Cover agreed for 5 of 6 cohorts. The February cohort at site 2 is still open. |
| R2 | Managers do not ask about the practice task. | 16 | Partly | No | 4 | 4 | 16 | 0 | Keep | 9 of 18 manager calls booked. Due Fri 8 Jan 2027. |
| R3 | Cohort dates clash with the January stock count. | 12 | Yes | Yes | 1 | 4 | 4 | -8 | Close | Site 2 asked to move its first cohort. It now starts on Tue 19 Jan 2027. |
| R4 | Only one trainer can run the role-play section. | 10 | No | No | 2 | 5 | 10 | 0 | Keep | The second trainer runs the role-play section at the pilot on Tue 15 Dec. |
| R5 | The invitation reads like a correction. | 9 | Yes | No | 2 | 3 | 6 | -3 | Close | New invitation sent on Fri 4 Dec. No requests to defer so far. |
| R6 | No quiet room for the role-play at site 3. | 4 | Watch only | No | 1 | 2 | 2 | -2 | Close | Quiet room at site 3 booked for every cohort date. |
| R7 | Nobody checks whether feedback conversations happen. | 12 | Partly | No | 4 | 3 | 12 | 0 | Escalate | HR agreed the question. The operations director must approve the change to the one-to-one notes. |
What you'll leave with
After 60 minutes the group has up to eight risks from everyone’s failure reasons, each rated 1–5 for likelihood and impact, an action, a warning sign and one owner for each risk scoring 8 or more, and a revisit date.
Before the session
- For trainers and project leads who run a pre-mortem with 4–12 adults before a plan is locked in.
- Fill in the plan brief with the plan owner a few days before. The plan owner listens and responds only in the actions step. Someone else facilitates, and you name a scribe.
- Print one failure story sheet per person, lens cards for every two or three people, and the tally, risk table and revisit sheet for the scribe. Have blank slips for unsigned reasons.
- Send the brief, the lens cards and the scales a day early for screen reader users. Every step works in writing, ratings can be typed, cameras are optional and anyone may pass.
The run of show
Open and read the plan brief
One minute: say “Today we assume the plan has already failed and we work out why, while there is still time to change it. This is not a vote on the plan.” Three minutes: the plan owner reads the plan brief aloud: the plan in one sentence, what success looks like, and the failure date. Two minutes: read the four rules. Write before you speak. Describe what happened, not who is to blame. Anyone may pass, and anyone may contribute in writing or on an unsigned slip. The plan owner listens and does not defend the plan until the actions step.
Set the failure scene
One minute: read the scene slowly, with the date from the brief: “It is Wednesday 31 March 2027. The plan failed. Not a small slip: it failed in the way that would hurt most. We are not asking whether it could fail. It did. Your job is to write why.” Two minutes: hand out the failure story sheet and the lens cards, or share them on screen. Answer questions about the task only, not about the plan.
Silent writing: failure stories
One minute: each person writes the headline and the three-sentence story. Five minutes: write at least four reasons in the table on the sheet, one per row, each as something that happened. Say: “When you are stuck, pick a lens card and answer its two questions.” One minute: anyone who wants to writes the reason that is hardest to say on a separate slip, without a name, and folds it. Online, it goes to the facilitator in a private message. One minute: each person ticks their top two reasons. Keep the room silent for all eight minutes.
Read reasons round by round
Eight minutes: go round the group. Each person reads one reason per turn, top two first, or passes. No discussion and no answers from the plan owner. The scribe writes each reason as a numbered note. When someone has the same reason, they say “same” and the scribe adds a tick. Keep going round until the eight minutes are up. One minute: the facilitator reads out the unsigned slips without comment. One minute: last call. Reasons not read aloud are handed to the scribe in writing.
Group the reasons into named risks
Four minutes: the scribe and the group put the notes into themes. Say: “Put two notes together when one fix would deal with both.” Two minutes: write each theme as one risk sentence: what happens, then “so” and the effect on the goal. Example: “Managers do not ask about the practice task, so it is not done between the two workshops.” Merge until there are no more than eight risks. One minute: number the risks R1, R2 and so on, and write the number of reasons in each one on the rating tally.
Rate likelihood and impact
One minute: read the two rating scales aloud and keep them on view. Six minutes: for each risk in turn, say “Likelihood, on three” and everyone shows a number from 1 to 5 at the same time: fingers, a number written on a card, or a number typed in the chat but not sent until the count. Then do the same for impact. The scribe writes the lowest, the middle and the highest number. When the highest and the lowest are 3 or more apart, ask each of those two people for one sentence, then move on without a second vote. One minute: multiply likelihood by impact, write the band, and read the risks out from the highest score down. Say: “The score orders our attention. It is not a forecast.”
Prevention actions with one owner
Two minutes: split into groups of two or three. Give each group one or two of the risks that scored 8 or more, the highest first. Six minutes: for each risk, the group writes a prevention action that starts with a verb, an early warning sign with a number or a date, a proposed owner and a due date. Four minutes: each group reads its lines aloud. The proposed owner says yes, or names who should own it instead. Only a name that has said yes goes on the risk table. The plan owner may now respond, in one sentence per risk. Risks that scored 7 or less get a warning sign and a person who watches, no action.
Read back, set the revisit date and close
Two minutes: read the risk table from the highest score down, with each owner and due date, then the low risks that are being watched. Two minutes: agree the revisit date, four to six weeks ahead or before the first big milestone, and who brings the table. Say what happens to the unsigned slips: their reasons are on the risk table without names, and the slips are shredded today. Two minutes: each person says or writes one sentence: “The warning sign I will watch for is…”, or passes.
Participant worksheet
Copy this table into a document, or print the worksheet from the bundle. Share only what you choose.
| Item | Detail |
|---|---|
| Plan brief | The plan in one sentence, what success looks like with a number and a date, and the failure date |
| Failure story sheet | One per person: a headline, a three-sentence story, at least four reasons and an early sign |
| Failure lens cards | Eight lenses with two questions each, for when people get stuck |
| Rating tally | Each risk with its reasons, the lowest, middle and highest rating, the score and the band |
| Scored risk table | Likelihood × impact, a prevention action, an early warning sign, one owner and a due date |
| Revisit sheet | Four to six weeks later: done or not, warning sign seen, new score, and keep, close or escalate |
Debrief questions
- Which reason surprised the plan owner most, and what does the plan now do about it?
- Which risk did only one or two people see, and what do they know that the rest of us did not?
- Where were the ratings furthest apart, and what would settle the difference?
- Which early warning sign will we actually look at, and who will notice if it appears?
Adapt it for your group
- For four or five people, keep the timings, read reasons for the full eight minutes, and write the prevention actions together as one group instead of splitting up.
- For a 40-minute version, ask people to fill in the failure story sheet before the session and send it to the facilitator, who groups the reasons into risks in advance. Open (4 minutes), set the scene (2), read out the risks and add missing ones (9), rate (8), actions for the top risks only (12) and close (5).
- Online, share the plan brief and the workbook on screen. People write in a shared document or send reasons to the facilitator in a private message, ratings are typed in the chat on the count of three, and the action groups use breakout rooms. Cameras are optional.
- For quiet participants and people who use a screen reader, send the sheets as plain text a day early and accept reasons in writing before or during the session. Every reason is read aloud with its number, unsigned slips count the same as spoken reasons, and anyone may pass in any round.
- With a large plan or more than twelve people, run one pre-mortem per part of the plan with the people who know that part, then merge the risk tables. Two parallel groups of six need two scribes and one shared rating step.
After the session
Type up the risk table within one working day, without names from the unsigned slips, and send it to everyone who took part. The plan owner adds each prevention action to the project plan with its due date. Each owner says in writing when an action is done. At the revisit, rate every risk again with the same scales, close the risks that are low and done, escalate the ones the owner cannot act on alone, and set the next revisit. After the work ends, run an after action review and check which risks happened, which warning signs appeared and which were never seen.
Questions people ask
What should a pre mortem template include?
At the minimum, the plan and the date on which you imagine it has failed, a place for each person to write their own reasons before anyone talks, and a way to turn the reasons into a short list of risks with actions. A template that a team can act on also needs rating scales with plain words, a rule for which risks get an action, one owner per action and a date to look again. This kit has a sheet for each.
How long does a pre-mortem take, and who should attend?
This script takes 60 minutes for 4 to 12 people. Invite the people who will deliver the plan and at least one or two who will feel its effects, such as the people it is for or a site manager. The plan owner attends and listens. Someone else facilitates, so the plan owner is not also running the room.
How is a pre-mortem different from an after action review or a kickoff risk log?
The after action review template looks back: it compares what was planned with what happened once the work is done. A pre-mortem looks forward and imagines that failure has already happened. The project kickoff meeting agenda has a short risk log inside a 90-minute kickoff. This kit goes further on how reasons are found: silent failure stories, lens cards and rating by the whole group, then a revisit.
How do you score likelihood and impact in a pre-mortem?
Everyone shows a number from 1 to 5 at the same time, so that nobody follows the first voice. The scribe records the lowest, the middle and the highest. The middle values are multiplied: a score from 1 to 25. In this kit 15 to 25 is high, 8 to 14 medium and 1 to 7 low. When ratings are 3 or more apart, ask the two people at either end for one sentence each. The score orders the discussion. It is not a forecast.
Where does the pre-mortem come from, and what are its limits?
The method was described by Gary Klein in “Performing a Project Premortem”, Harvard Business Review, September 2007. The script, sheets, lens cards, scales and example in this kit are written for Superglue. A pre-mortem finds reasons people can already imagine. It is not a formal risk register for compliance, health and safety or finance, and it does not replace one.
What is included in the free and editable files?
The open PDF has the plan brief, the failure story sheet, the lens cards, the rating scales, the rating tally, the scored risk table, the revisit sheet, a filled example of each and the 60-minute facilitator script. Subscribers also receive the editable Word file, an Excel workbook with a rating tally that calculates the middle value, spread, score, band and rank, a risk table with days to each due date, a revisit sheet that shows each change of score and filled examples, and a CSV of the session plan.
How to use the files
Print the PDF, edit the Word file in Word or Google Docs, or open the CSV or XLSX in a spreadsheet. Open files need no email.
Can I use this with paying clients?
Yes. Use and adapt this original Superglue kit in your own training, including paid client sessions. Please share this page rather than redistributing the unmodified kit. Read the full usage terms.