Premortem

ℹ️ Skip to key parts of the session by clicking on the chapter dots along the timeline.

Premortem - a straightforward opposite of a ‘post mortem’. This tactic helps you to work out why your project might die; that is, working out what could go wrong, so that you can take steps to prevent it from happening.

We tend to make the same mistakes repeatedly - as evidenced by history! So this tactic helps product teams feel more comfortable about what’s to come, and reduces the anxiety around the unknowns.

This is one of Charles Burdett’s ‘Desert Island’ cards - he can’t live without it, and has seen it work in every instance. So who better to run a workshop tutorial on the pre-mortem card? Oh, and he’s the author of the Workshop Tactics deck, too. And the Founder of Pip Decks. We’re in safe hands!

See the full instructions on the Premortem card.

👉 View the Miro board for this session

How to set up your Premortem workshop

Charles uses Butter (video conferencing tool) and Miro (virtual visual collaboration platform) to run the workshop, so you’ll get a good overview of how to set your Miro board up for this particular session. You’ll see how Charles has brought in some imagery from the cards themselves, and you’ll get an idea of how to use features like the timer, post-its and ‘bring to me’ and how to run a Q&A.

Introduction to the Premortem workshop

This tactic is one of the ‘Project Kick-off’ Session cards - it’s the starting point for that set. You can use it on its own as a project kick-off, too. This is a high-level overview of how to use the tactic, whether you have a detailed project plan or not.

The example used here is a new garden-centre franchise that needs a website. The team on the workshop are acting as the digital marketing team.

Project aim: launch a new website for the garden centre.

Creating a project plan

Before you start, you need the outline of a plan. You can create that in the session, or prepare it in advance in a separate workshop. The Sticky Steps tactic is great for mapping out a simple path from X - Y, starting with the end site. In this case:

Launch site (end result)
Test code
Content finalised
Design brand
Plan architecture
Map scope/features (start point)

While creating the list, you can use the ‘Bring everyone to me’ on Miro to make sure they don’t get lost in the workspace.

Running a pre-mortem session

Kick things off with a vision of a catastrophic failure

Charles talks through the first part of the tactic: tell your workshop participants that you’re six months into the launch of the new website, and it’s all gone horribly wrong. Dead in the water. Everyone could get fired.

Gather and sort ideas

Ask the participants to create post-it notes (ctrl + n) on the Miro board, each with one reason why the project might have failed. You’ll need to watch the video to see all the ideas put forward, but here are a few:

Copywriter downs tools half way through the process
No one thinks to run any tests with users
Stakeholders don’t sign off

You’ll also get to see the timer setting in Butter, which plays some fancy hold music as people get to work.

So this list of post-it notes should hopefully reflect the range of roles in the workshop - marketers, designers, web developers etc. It’s only as successful as the people in the group, especially for collaborative projects. For agencies, getting the client involved is also a smart step.

The next step is to Theme sort those post-it notes into themes to make things more manageable. This is done by asking participants to move their post-its into groups based on their ‘theme’; for example: content, technical, budget, testing and domain.

Assumption map the themes

Charles then talks about the benefits of using Assumption Mapping to prioritise the themes so that you can choose which issues to tackle first. Which one is the most high-risk, unknown theme? And which are relatively low risk, and we understand how to mitigate easily?

This can take a while - minimum of an hour, up to three hours! Remember, you could be working through hundreds of post-its depending on the size of the project, how many people are in the session - and how many things might go wrong.

Then you’ll take the highest risk, highly unknown items first and ask the group (which should include experts in all areas) to investigate! How might it be mitigated against?

You can then use Who, What, When to produce a list of actions, owners and deadlines.

For example:

Risk: the website can’t handle the amount of traffic it gets
Mitigation: run a traffic spike test
Action from session: named developer schedules a test by X date

Remember to leave enough time to fix the issue if it arises. For example, if the website fails the traffic spike test, you’ll need time to resolve that issue before the launch date.

The workshop provides you with a list of clear, named and dated actions so that the whole project is more robust. And that’s the pre-mortem!

Q&A using Miro tools

Charles then asks the participants for feedback in the form of:

One question
One thing that surprised you
One thing you learned

Questions (and answers!)

How long would a premortem take to prepare?

How much do you want to be seen to know what you’re doing? You can give the card to someone half an hour before running this session, and they’d probably manage to run it.

If you’re a meticulous planner then you’ll want to start planning in advance - as in, days in advance. And if it’s a high-stakes meeting? Do a dry run with some help from others.

But you can over plan! So it depends on you. The cards are mostly designed to be picked up and used on the fly - although some have got pre-work that needs to be done. You can also work through the steps on the card with the people in the room, so you’re all helping to facilitate.

You’ve mentioned convergent and divergent tasks before - how do you spot which is which?

Divergent tasks open the discussion out (e.g., brainstorming). Convergent tasks try to bring the threads of discussion back together and sort them out (e.g., assumption mapping).

Anything where you ask everyone to fill in post-it notes is probably a divergent task. When you start moving those post-its around or narrowing them down (voting), that's convergent. The ‘Decision’ category of cards are all good examples of convergent tactics.

Do you set expectations or get people to do any work up front for the premortem session? And is there a back-up you can prepare in case the tactic doesn’t work?

Sometimes, you’ll let them know what the plan is and get them to think about the plan beforehand. Other times, you might keep it under wraps. If you keep it mysterious - “We’re going to predict how the project might fail” you’ll get more attendees!

In terms of it not working - if you have the wrong people in the room, for example - the session is planned to (hopefully) avoid any major blind spots. The assumption map helps you look at the issues raised through different lenses.

Surprises

Planning can be creative!
It’s possible to run the session cards without using all the steps if that’s what would suit your project.

Learnings

Learning an organised way to plan for and tackle issues
Pre-mortem be used with all stakeholders - not just the delivery team

Thanks to everyone that joined the session! Remember, you can download a Miro board template for this workshop.