When I started out in the app industry, I worked at a growth agency that partnered with a development agency. As a result, a lot of our projects involved launching entirely new websites or app redesigns.
It was amazing — a fresh slate to use all our customer insights and improve performance! …But also slightly terrifying. Not because of the work itself, but because of the waiting.
That horrible in-between phase after you’ve shipped something, where you’re refreshing dashboards, overanalyzing early feedback, and catastrophizing every dip in the numbers.
What helped me wasn’t positive thinking. It was the opposite: leaning into pessimism.
Deliberately imagining everything that could go wrong before launch. Thinking through every possible scenario. Having the trickiest, most uncomfortable conversations with clients and teams to explore all the what-ifs.
It’s called a pre-mortem. (A bit morbid, but it works.)
Since then, I’ve used it for everything from website redesigns to subscription app launches; especially in products where feedback loops are fast and the stakes compound quickly.
And it’s become one of my favorite planning tools.
What is a pre-mortem?
A regular post-mortem happens after something fails, the classic sprint retrospective. You gather the team, ask not only ‘what went right’ but also ‘what went wrong’ or ‘what could have gone better’, and try to learn from it.
It often comes with a few cringe-inducing “we could have predicted that” moments, where Captain Hindsight suddenly looks like a genius.
The problem is, it’s reactive rather than proactive. You’re analyzing what went wrong after the fact, but it doesn’t change the outcome.
A pre-mortem flips that. The concept is simple: before you launch, imagine it’s three months later, and the launch has completely flopped. Really put yourselves in that moment: the disappointment, the stress, and the why-is-this-happening panic. Bring it to life.
Then ask the full team: why did it fail?
I called it pessimistic, but honestly, it’s not. It only becomes pessimistic if you assume failure is inevitable and stop there. The real goal is to surface risks while you can still do something about them.
If anything, it’s more realistic — and even optimistic — because it assumes you can fix and prevent those risks.
And the interesting part? The things that come up in a pre-mortem are rarely surprising. They’re usually the things everyone already half-knows but hasn’t said out loud.
Who came up with such a morbid concept?
The credit goes to Gary Klein and his often-cited Harvard Business Review article from 2007. But Gary (and yes, given how much I love this concept, we’re absolutely on a first-name basis) had been using the method for a good 30 years before that. He first developed it in the early 90s, so no one needs to do confusing maths back from 2007.
A psychologist by trade, he was fascinated by how people make decisions under pressure and uncertainty. He studied this in one of the most high-stakes environments imaginable: the US Air Force. Later, he developed the idea of pre-mortems further within his own research company, drawing not only on that work but also on medical post-mortems.
What Gary understood — and what more of us need to recognize — is the flawed reality of group decision-making. We’re wired for harmony. We want to get along. No one wants to be the negative Nancy, pointing out risks or poking holes in a plan.
I recently emailed a client outlining all the risks in their current target-setting model and ended with, “sorry, this is a bit negative…”. I felt bad raising those risks, but I also knew, from experience, what happens when you don’t. When you let people believe targets are promises rather than stretch goals. Still, it sucks to be the one to say it. That’s exactly the tension Gary Klein was trying to solve. A pre-mortem removes the pressure to keep up a ‘rah-rah, everything is great!’ mindset.
Instead, it creates a space where calling out risks isn’t awkward; it’s expected.
Since then, the concept has taken off. You’ll find pre-mortems everywhere now: in startups, corporate boardrooms, and even back in the military (a full-circle moment, really).
Daniel Kahneman — who many people know from his book Thinking, Fast and Slow — even called it his single most valuable decision-making technique. Not bad for something that started as an internal fix at a small research company.
Gary later expanded on the idea with what he calls the ‘double-barreled pre-mortem’. After imagining failure, you run a second round where you daydream about wild success — and what could go wrong there, too.
It sounds more optimistic, but the goal is the same: surface risks. Even success can break things. If only Ticketmaster had done that before the Taylor Swift Eras Tour ticket sales…
So that’s how, and why, pre-mortems started. But beyond making it easier to speak up, why else is this tactic so useful?
Why do pre-mortems actually help?
There are five reasons why pre-mortems are so powerful:
1. They make you a better fortune teller
Pre-mortems use a technique known as prospective hindsight (a fancy way of saying imagining the future), which Deborah Mitchell at Wharton found can increase your ability to correctly identify reasons for future outcomes by 30%. All that by pretending it has happened. You basically become a better fortune teller, pretty cool, right?
So it helps you surface real potential risks more effectively than just thinking about them without a specific prompt.
2. It creates psychological safety to talk about risks
We’ve talked a bit about this already. By making the goal ‘identify risks’, you make it safe to say them out loud. People are often all thinking “the founder keeps changing priorities” or “pretty sure this plan was ChatGPT and hasn’t been checked”, but they might be too scared to say it out loud.
As you see others be more open, you’ll also end up being more open.
I think this is especially valuable when you have a room full of young, smart people. I remember at my first job, I assumed whatever senior leadership said was the right call. It felt intimidating to challenge someone with 20 years more experience than me.
Now, as a leader, I know you can tell people to speak up — but the reality is, not everyone will. A pre-mortem helps level that playing field. It gives people permission, and a clear prompt, to question assumptions, raise concerns, and say the thing they might otherwise keep to themselves.
3. It makes the fear concrete
Vague anxiety is paralyzing; we’ve all catastrophized over endless, nameless what-ifs. Specific risks, though, are manageable.
Once you’ve written “what if no one buys it?” on paper and mapped out a plan for it, it stops being a powerful fear. It becomes just another scenario to handle.
Even better, involving the whole team in a pre-mortem helps counteract the planning fallacy — that tendency to be overly optimistic, assume everything will work out, and underestimate time, costs, and risks. By making the risks concrete, you can create a clear action plan.
Think of it as a bit of growth group therapy.
4. It removes panic from decision-making
Have you ever stood in line for coffee, and suddenly it’s your turn way faster than you expected? The barista asks, “What would you like?” and you stammer out, “Um… errr, a vanilla matcha latte, wait, an iced vanilla matcha latte!” Only to take a sip and think, “What was I thinking?”
Okay, that might have just been my own therapy moment the other day, but the point stands: under pressure, we’re terrible at making decisions on the spot.
The same goes for post-launch problems — you’re stressed, stakeholders are asking questions; you’re tempted to make snap, reactive decisions. You blurt out the first thing that comes to mind, and hope for the best.
A pre-mortem changes that. If you’ve already decided, ‘If X happens, we do Y’, then you don’t have to think under pressure. You execute the plan. You’ve already discussed the right course of action as a team, ahead of time. I do this for every experiment I run because it makes ‘failed’ experiments easier to handle: we know what failure looks like, we can learn from it, and we already have a sense of what to test next.
5. It gives you permission to take action imperfectly
I once worked with a Head of Product who was basically a pre-mortem specialist, though he probably didn’t realize it, since we’d never talked about pre-mortems. He had an uncanny ability to spot every possible risk, think it through, and tackle it. Sometimes, it felt frustrating, as it could slow projects down. Until I realized that if I named the risk and laid out a plan, he was more than happy to roll with imperfection.
The extreme opposite of optimism in project planning isn’t pessimism, it’s perfectionism. Try to control everything, and you’ll never go live.
A pre-mortem shows you that even if things go wrong, you have a path forward. That makes it easier to ship. The goal isn’t to prevent every risk, it’s to be conscious of which risks you’re willing to accept and which you aren’t.
By now, hopefully, you’re convinced and starting to wonder how to run your own pre-mortem.
How to run a pre-mortem
While the concept itself is simple, a few small tweaks in setup and approach can make your pre-mortem far more powerful.
Step 1: Gather the relevant individuals
While you can do a pre-mortem alone, I can’t recommend doing it with your team enough. Different people spot different risks based on their expertise, and hearing others’ perspectives can reveal risks you might never have considered.
Even better: bring in someone outside the project; someone who won’t have rose-tinted glasses. Ideally, a classic naysayer who can look critically at what’s being assessed (kind of like that Head of Product I once worked with).
Step 2: Set the scene
Next, set the scene.
Say something like: “It’s [date 2–3 months from now]. This launch has failed. Not a minor disappointment, but a proper flop. What happened?”
Pick a timeframe that feels realistic for the project. Then lean into the drama — think back to high school theater class and really bring it to life. The goal is to make it feel real, to transport yourself to that future moment, and look back with hindsight.
You can even get creative: have AI mock up a founder email, a critical internal memo, or an article in a relevant newspaper or magazine. The more tangible and vivid the scenario, the easier it is to uncover the real risks.
Step 3: List every reason it might have failed
Have everyone list reasons separately. Write down everything, even if they seem unlikely. For example:
- We couldn’t reach enough of our target audience
- The trial-to-paid conversion rate is too low
- We saw very high initial churn
- We struggled to get downloads
- The price felt wrong (too high, too low, confusing)
- A competitor launched something similar at the same time
- We got bad press or reviews early on
- The team burned out before we could iterate
The more specific, the better. We aren’t aiming for perfection here; quantity over quality is the key at this stage. The goal is to surface as many potential risks as possible.
Step 4: Go through the risks and group them by theme
There will inevitably be some overlap in the concerns people raise. That’s okay, this is a no-judgment moment. At this stage, you’re not prioritizing; you’re organizing.
Give everyone a chance to voice their concerns. Whatever you do, don’t try to diminish them or explain why something won’t happen, especially if you’re the leader or founder.
Instead, focus on grouping the concerns by theme. This helps you see more clearly:
- Which concerns are the most common
- Which areas seem to carry the most potential risks
Optional additional step: When pre-mortems require a culture change
In some companies, getting into the pre-mortem mindset can be tricky. When things are busy or deadlines are tight, it’s easy to skip it. Or sometimes, people just feel uncomfortable bringing up the uncomfortable.
Here, the framework Shreyas Doshi suggests can be really useful. He recommends categorizing risks into three types:
- Tiger: a real threat — something that will hurt the company if left unaddressed.
- Paper tiger: a perceived risk/threat that, on closer inspection, is unlikely to cause real problems. These are the things the team might be stressing over, but are actually minor and manageable.
- Elephant: unsurprisingly, this is ‘the elephant in the room’ — a risk that is present but isn’t being talked about.
Using this playful, visual language helps build the psychological safety that’s so important for effective pre-mortems. It makes discussions lighter and more approachable, and it gives you a way to raise risks in other settings too. For example: “This might be a paper tiger, but have we considered…?”
Step 5: For each risk theme, ask two questions
Now that you’ve done some idea spring cleaning, it’s time to assess risk levels and impact:
- Certainty assessment: how likely is it to happen?
- Impact assessment: how bad is it if it happens?
This helps you map everything onto a simple 2×2 grid, so you don’t get stuck debating unlikely scenarios or low-impact risks. Instead, you can focus your energy on the ones that actually matter.
Depending on the size of the group and the scope of the project, this might even be a separate session. The goal here isn’t to solve everything yet; it’s to start identifying which risks deserve your attention first.
Step 6: Determine whether you need to plan or prevent those risks
Next, focus on those biggest risks and decide whether they’re preventable, plannable, or both:
- Prevent: can we reduce the likelihood of this happening?
- Plan: if it happens anyway, what will we do?
If you try to prevent every single risk, nothing will ever go live, so this is where you need to be critical. Ask yourself: Is this something we can realistically mitigate, or is it a risk we need to consciously accept?
The goal isn’t to eliminate all risk. It’s to understand it and feel prepared for what might happen.
Step 7: Work out potential solutions
Now, depending on the complexity of the risk, this might require an extra session — or a few — to properly think through. Don’t get caught trying to solve everything on the spot. Some risks need deeper brainstorming or input from others.
What does help is summarizing your plans into simple if/then statements. They’re clear, practical, and bring a sense of calm to the unknown.
For example:
- If activation is below 10%, we’ll simplify the first-time experience and retest
- If we attract the wrong customers, we’ll adjust our targeting and messaging before scaling spend
- If nobody hits the paywall, we’ll move it earlier and test different copy
- If day 1 churn is above X%, we’ll investigate whether the paywall is creating impulse conversions
Each of these should be tied to a specific metric you’re already tracking — trial conversion rate, activation rate, churn, and MRR. Set your trigger thresholds before launch, so you’re not debating what ‘bad’ looks like after the fact.
If you’re running paywall or pricing experiments, RevenueCat’s Experiments feature lets you set up A/B tests and define success metrics before you launch — so you’re not making decisions under pressure.
Step 8: Review and revisit
One last thing: don’t let your pre-mortem gather dust in a shared doc like so many other strategy decks. Depending on your timelines, set a few clear check-in points, for example, Day 7, Day 30, and Day 90 post-launch.
At each one, revisit your if/then statements and ask: are any of these scenarios actually playing out?
This is where the pre-mortem really earns its keep. The planning is valuable, but the follow-through is what helps you avoid those “we should have seen that coming” moments later on.
For longer projects, it can even be worth running a second pre-mortem closer to launch; a fresh look, with more context, often surfaces new risks you couldn’t see earlier.
Example: subscription app pre-mortem
Before launching a new subscription tier for a client last year, we ran a quick pre-mortem.
One of the risks we identified: ‘Existing customers might feel cheated if new customers get a better deal’.
We couldn’t fully prevent that. The new tier genuinely offered better value, and it needed to. It unlocked a segment of customers we were previously pricing out. But by naming the risk, we could prepare for it:
- We spent more time positioning the existing tier, not just promoting the new one
- We added a relevant feature to the higher tier to reduce cannibalization
- We communicated proactively with existing customers, explaining the new tier and who it was for
- We prepared responses for the support team
The result? Far fewer complaints than expected, and only a small increase in support tickets (mostly from customers who missed the updates).
More importantly, the new tier drove growth rather than cannibalizing the existing one, with overall MRR increasing significantly.
When to use pre-mortems
Pre-mortems work in a wide range of scenarios:
- Launching a new product or SKU
- Changes to subscription models
- Pricing adjustments
- Entering a new market or channel
- Rebrands or major website overhauls
- Paid campaign pushes
I even run them with new advisory clients to make sure I can support them effectively and drive real impact.
Basically, try a pre-mortem anytime failure would hurt, and you want to have thought it through in advance.
A pre-mortem won’t make launches less scary
You’ll probably still refresh dashboards too often and overanalyze early data — I certainly do.
But the quality of your anxiety changes. Instead of a vague dread that something might go wrong, you have a clear list of what could go wrong, and a plan for each scenario.
The best launches I’ve been part of weren’t the ones where nothing went wrong. They were the ones where we’d already talked through what we’d do if things did go wrong.
So before your next paywall test, pricing change, or new tier launch, take 30 minutes with your team. Imagine the worst, write it down, make a plan.
Then ship it anyway.