It’s never just “something went wrong”. It’s “something went wrong at 2.07 a.m., the night before launch, while you’re already running on four hours’ sleep and five shots of caffeine.” Everything feels louder, sharper, and more personal.
These aren’t just emergencies; they’re full-on crises, and you’re expected to deal with them right away, no matter what else is happening in your life or business.
Here’s the part your nervous system doesn’t believe yet: you don’t have to fix everything instantly, and the outcome doesn’t have to be quick or perfect – it just has to be handled. With that in mind, here’s a practical, human guide to getting through crisis moments without losing your mind (or your company).
If you’re reading this in a crisis, let’s keep it simple:
If you’re not currently on fire, great. Let’s walk through this slowly so your future self has a plan.
When you own or run a startup, very few things feel “just business”. A bug isn’t just a bug; it’s “we’re going to lose all our users.” One angry tweet isn’t just feedback; it’s “this will destroy our brand.”
Your brain loves a good worst-case scenario, especially when you’re exhausted. So in a crisis, you’re not only dealing with what’s actually happening – you’re also carrying everything your imagination is piling on top of it.
The goal here isn’t to become some unshakeable robot who never panics. It’s to build just enough calm and structure that you can act deliberately instead of reacting from pure fear.
Something blows up. Your heart rate spikes. Your brain screams, “FIX IT NOW”. That’s usually when people fire off emails or messages they regret, promise timelines they can’t possibly meet, and press buttons that should never, ever be pressed.
So, weirdly, the best thing you can do in a crisis is to pause. For you, that might mean taking three slow breaths, walking to the kitchen for a glass of water, or just pacing to the other side of the room and back. You’re not wasting time; you’re creating a tiny gap between “oh no” and “do something”.
In most cases, this pause won’t delay the solution in any meaningful way. You’ll lose maybe 3–5 minutes. But it will dramatically reduce the chances that you make the situation worse because you’re operating in full panic mode.
You’re not avoiding action. You’re buying just enough time to show up as the person who runs the company, not the person who’s about to torch it by accident.
We need to walk it back. A simple rule:
If you can’t summarise the problem in one sentence, you’re reacting to the fear of the thing, not the thing itself.
Try this: write down a single sentence that describes what’s going on. That might look like: “Our payment processor is failing for around 30% of transactions,” or “A customer has publicly accused us of mishandling their data,” or “A team member has reported harassment by a manager.” If you can’t do that, pause again. Strip out all the “what ifs” and focus only on what you know is true right now.
Once you can define the problem in one sentence, it stops being a vague cloud of doom and becomes something specific you can respond to.
If the situation involves anything legal, safety-related, or injury-related, that’s your cue to take some extra steps and recognise where your DIY limits are – for example, moments where you really do need to understand when to get a personal injury lawyer involved.
Depending on your country and situation, that might also mean speaking with an employment lawyer, data/privacy specialist, or relevant regulator. When people are involved, getting proper advice early gives you a much better chance of keeping a bad situation from turning into an unmanageable one.
Putting a label on the crisis helps you decide who needs to be in the room. You can think in five broad types:
A quick triage checklist can keep you grounded:
It’s also worth remembering that crises have very real financial costs. Industry studies suggest an hour of downtime now often costs over $300,000 for mid-sized and large enterprises, and can hit $1–5 million per hour at the top end – even smaller businesses report losses of hundreds of dollars per minute when key systems go down. You don’t need to be at enterprise scale for “this bug” to mean “this is burning real cash”.
The label doesn’t fix anything, but it does tell you who must be involved and what decisions can’t wait until tomorrow.
One of the quickest ways to make a crisis worse is to have six people all half-leading it and nobody clearly in charge.
You need three things: first, one incident owner. This doesn’t have to be the CEO. It should be someone close enough to the problem to make good calls. Their job is to coordinate, set priorities, and keep everyone updated – not to do all the work themselves. Second, a single “war room”: a dedicated Slack channel, a Zoom call that stays open for a while, or an actual physical room if you’re in the same place.
The point is one shared space where the live conversation happens, instead of scattered DMs and side chats. Third, a simple incident log: a Google Doc, Notion page, or anything that lets you jot down timestamped notes (“10:42 – outage confirmed in EU region”), who’s doing what, and what’s already been tried. Future-you will thank you for this, and so will anyone who needs help later.
It doesn’t need to be fancy. A named owner, one shared channel, one running log. That alone can take a crisis from “uncontrolled chaos” to “messy but manageable”.
Silence makes crises bigger.
When people don’t know what’s going on – whether that’s your team, your customers, or your investors – they fill in the gaps themselves. And their imagined version is usually worse than reality.
You don’t need a perfectly crafted statement to start communicating. You just need to be honest, specific where you can, and clear about when you’ll be back with more information.
There’s data behind this, too. Utilities, telcos and infrastructure companies that send proactive outage updates consistently see better satisfaction scores, and in survey after survey, a large majority of customers say they want proactive communication during service interruptions.
In service and support research, we see the same pattern: owning the problem early and setting expectations clearly can soften the impact of a failure and improve how people judge the whole experience – even if the underlying issue is pretty serious.
In practice, that might look like:
Internal (Slack / Teams) message
“Hey all – we’ve identified an issue with [brief description].
[Name] is the incident owner, and we’re working on it now.
For the moment, please direct any related questions to [channel/person] so we don’t lose track.
We’ll share an update by [timeframe], even if the only update is ‘still working on it.’”
Customer status/email
“We’re currently experiencing an issue affecting [who/what is impacted].
Our team is actively working on a fix, and we’ll update this page again by [timeframe].
We’re sorry for the disruption – we know you rely on us for [what you help them do], and we’re treating this as our top priority.”
Investor note (if relevant)
“Quick heads up – we’re dealing with [short description of crisis].
Right now, we believe the impact is [summary of impact].
The team is focused on [current actions].
We’ll send a fuller update once we’ve resolved the immediate issue and have clearer data, but wanted to keep you in the loop.”
You don’t need to share every tiny detail or speculate. Just show that you’re aware, you’re on it, and you’ll be back with more.
Nothing in this article is legal advice. If you’re in a high-risk situation – especially where people, data or regulations are involved – you should talk to a qualified professional in your jurisdiction.
As a founder or operator, you’re probably used to figuring things out on your own. That scrappy “we’ll just learn it” mentality is great… until it isn’t.
There are certain lines you really don’t want to cross without expert help. For example:
A useful rule of thumb: if people, data or laws are involved, don’t guess. Get help.
Yes, that might mean spending money you didn’t plan to spend. But in many cases it’s still cheaper than handling things badly. Once you factor in downtime, churn, fines, refunds, lost deals and the time you’ll spend trying to repair relationships, professional help starts to look like the bargain option.
Tempting, but expensive.
A short, honest debrief is one of the best crisis tools you’ll ever invest in.
You don’t need a big formal process. Just gather the people who were involved and walk through a simple outline:
You don’t need a 30-page post-mortem. You just need to make life a bit easier for the next version of you who has to deal with something similar.
Over time, every incident you handle, document, and learn from becomes a real, lived experience you can point to when investors, partners or future hires ask, “What happens when things go wrong here?”
In the middle of a crisis, it genuinely feels like every move you make is life-or-death for the company. Zoomed out, most crises are survivable – and some of them actually end up building trust with your team, customers and investors when they see how you show up under pressure.
You don’t have to fix everything instantly. You don’t have to craft the perfect response. You just have to slow yourself down, define the real problem, bring the right people into the room, communicate honestly, bring in experts when it’s bigger than you, and learn enough from it that you don’t have to fight the same fire twice.
That’s what “handling” a crisis really looks like in startup land. Not magic. Not heroics. Just messy, human, deliberate action when everything feels like too much – and the choice to come out a bit wiser on the other side.