How to own a mistake at work
Who knew making a mistake could lead to more trust and responsibility?
At some point you’re going to mess up, and at some point you’re going to mess up badly. What happens next is going to determine whether your mistake shakes peoples’ confidence in you — or actually makes them trust you more.
If you’re in a healthy work environment, your boss (and anyone else involved) wants to get the mistake fixed ASAP before pointing fingers, playing the blame game, or really anything else. It will be “us against the problem” instead of “people against each other.”
(If you aren’t in an environment like that, remember you can leave if you want to).
I’ve made mistakes, or mistakes have happened on my team that I’m ultimately responsible for, that include:
Sending “your price is going up” emails to people whose price was not going up
Accidentally deleting 10,000+ contacts from a system
Sending out a press release with the wrong messaging
Not realizing huge groups of people weren’t getting sent onboarding messages
Broken links in emails to hundreds of thousands of people
And those are just the ones that involve external comms. Lots of mistakes come from not communicating well with other teams or not running things past an exec at the right time.
Mistakes happen. If you never make mistakes it probably means you’re making the biggest mistake (only doing things that are easy/you know well, instead of the highest impact things). It’s treating the mistakes as mistakes and following this process that lets you learn from them.
Your immediate reaction to finding out you made a mistake is key, and everything you do in the aftermath needs to feed the “us against the problem”-ness of it all.
Step 0: Be generally good at your job and have the results to prove it
Ok this is cheating. You can’t change how good you’ve been at your job in the aftermath of your mistake.
What you can be doing always — and what these steps are part of — is great work.
As a manager, if someone with a history of questionable work, who’s been on and off PIPs, who is stubborn and difficult to work with makes a huge mistake, I’m likely to think of the error as another slip in a long line of slips.
If the person on my team who’s absolutely crushing it, is great to work with, and who I would do anything to keep makes a mistake, it’s probably fine. Especially if they follow these steps.
In her book Reality-Based Leadership, Cy Wakeman describes “emotional expensiveness” as a key factor of your performance, and it’s especially important to think about in the context of mistakes.
“Emotionally Expensive employees are those that blame their circumstances or others. They resist change and insist on having time to adjust to change. They argue with the direction of the organization and withhold their buy-in. They believe that buy-in is optional. They waste energy and talent by creating drama rather than results. One source of emotional expensiveness that is often overlooked is the judgmental high performer. This is the person who is really good at his or her role; however s/he judges the shortcomings of everyone else and does little to help.”
Being skilled at your job is not the only thing that matters, but having a history of great work and not being emotionally expensive is going to help with the next steps here.
Step 1: Take 100% ownership without hedging
Shit went bad and it was your fault.
I don’t care if shit went bad because another team said they would do something and didn’t. I don’t care if some system you use is arcane as hell and caused the mistake. I don’t care if the timeline was unreasonable or the scope was unclear or the feedback came late or about anything else that went wrong.
I’ll care about all of that in step 2. Step 1 is to say that everything was your fault.
I feel like this is easy to gloss over with a “yeah yeah but…,” so I’m going to double down. You have to take the stance that everything was your fault, and doing so is in your best interests for these reasons:
If you frame everything as being your fault, you’re going to focus your own attention on solving the problem. If you start to hedge EVEN A LITTLE you are pointing your mental energy at something that is not solving the problem.
Nothing is ever actually 100% your fault, and anyone else who had a role to play in the mistake will appreciate you stepping up.
Nothing is ever actually 100% your fault, and your boss knows this too
You volunteering to take 100% of the responsibility is, paradoxically, going to make other people more likely to want to help you fix the problem
You volunteering to take 100% responsibility is one way of showing people (and your boss) that you’re taking this fucking seriously — that you care as much as they do
Plus, one more reason — hedging never works.
Playing the blame game is exhausting. Shit went bad and it was his fault, no her fault, no that team’s fault, no the technology’s fault, I can’t work in situations like this my team needs more resources and that other team doesn’t respect my team and it’s about trust and the timeline was ridiculous and this and that and OHMYGODCANWEJUSTSTOP.
All of that, and you’re still going to take the blame in your boss’ mind. Even if you manage to successfully point the finger at someone else, that person is not going to be your new best friend. They’ll remember this the next time you have to work with them, which will probably be in literally (figuratively literally) 30 seconds when you still have to fix the problem.
Trying to shift blame makes you seem like a guilty weasel. Taking blame makes you a martyr but without the dying part.
Step 2: Investigate the real picture of what went wrong. Understand what actually happened.
In step 1 everything was your fault. But we both know that’s probably not really what happened, and step 2 is about what actually happened.
“What happened” is probably useful information when it comes to solving the problem. It’s definitely useful information for preventing the problem from happening again.
In Managing the Unexpected, a very business-y textbook kind of book, Karl Weick and Kathleen Sutcliffe argue that a key defining feature of “high reliability organizations” (HROs) is an unwillingness to accept simple explanations.
Another way HROs manage the unexpected is by being reluctant to accept simplifications. It is certainly true that success in any coordinated activity requires that people simplify to stay focused on a handful of key issues and indicators. But it is also true that less simplification allows you to see in more detail what might be causing the unexpected. HROs take deliberate steps to create more complete and nuanced pictures of what they face and who they are as they face it.
This extends also to not fully addressing problems at their root cause. “The problem stopped and we don’t know why, but it’s probably fine” is the type of thing that can get you into trouble later.
So step 2 is about the root cause. It’s about all of the different dynamics — the teams, the timelines, the scope, the concurrent projects, the clarity, the technology, the data, the code, the weather — anything that could conceivably have affected the situation.
I made a mistake not too long ago in which I accidentally sent an email to ~16,000 users. Thankfully, the email was innocuous (linked to some helpful resources) and didn’t cause any issues. The mistake was because of how our system worked — when I added a way to track in-product behavior, it applied that tracking retroactively. Every user who met the criteria to get sent the email got the email.
That’s a technology problem caused by a tool not working the way I expected it to. The fix is easy, because now I know how the tool works and I can avoid that mistake in the future. This was still my fault for not checking in advance.
More complicated mistakes can have a lot of component parts, and this is where you list them out. The timeline changed last minute? Write it down. Communication between teams was messy? Write it down. A contractor didn’t get back to you quickly enough? Write it down.
The goal is understanding, which will help you manage the communication (the next step) and create a plan to avoid this mistake in the future.
Step 3: Update everyone who needs to be updated earlier than they need to be updated
Depending on the context, a mistake can be urgent and super messy. You probably won’t have all the details you need at the beginning of discovering a mistake, but you still need to keep everyone informed.
Here’s what I do:
As soon as I discover that a mistake has occurred, I’ll message the relevant person (usually my boss) and say something to the effect of: “There was an incident related to [mistake]. Right now it looks like [content of the mistake] happened, but we’re still digging in. [Person on team] and I are looking for more information on what happened, and I’ll keep you updated as we learn things.”
As details come in, I’ll follow up with the additional context and the continuing steps. “We’ve identified that there’s some problem with [system] and we’re working to understand it and implement a fix. Right now we’re trying [first thing], and the other things on our mind are [second possibility] and [third possibility].”
Once there’s a plan of action, I’ll communicate the plan of action and the timeline, as well as a request for approval (if needed). “Here’s what we’ve discovered, here’s our step by step plan, here’s who else is involved, and here’s the timeline.”
After the acute problem is solved, I’ll send over materials on specific practices we can put in place to avoid making similar mistakes (or generally fix the systems that broke) in the future.
Take a look at the first two responses:
“There was an incident related to [mistake]. Right now it looks like [content of the mistake] happened, but we’re still digging in. [Person on team] and I are looking for more information on what happened, and I’ll keep you updated as we learn things.”
I state clearly which area of the business is affected, and the current understanding of what happened. I communicate who else is involved, and make it clear that I’ll keep updating them as there are developments.
“We’ve identified that there’s some problem with [system] and we’re working to understand it and implement a fix. Right now we’re trying [first thing], and the other things on our mind are [second possibility] and [third possibility].”
I’ve narrowed down which thing broke. I also explicitly state what we’re trying to do (or investigating), and what we’ll do after that if it fails.
Every response reduces uncertainty.
It can be scary when shit hits the fan, and the scariest part is not knowing how much shit has landed on what sized fan, or what the fan is pointed at. The shit could strike at any moment. Columbia University, if you’re listening, I would like to submit that analogy (devoid of context) for Pulitzer consideration.
The responses here communicate:
The extent of what’s currently known
What steps are currently being taken
That the boss will be kept updated the moment new information comes in
They also share enough info that the boss can weigh in with context if they have it.
When you communicate like this during a crisis, you add some much needed stability. The boss doesn’t have to wonder what’s happening or if more has gone wrong, because they know what’s happening and they know that nothing has changed until they’ve heard from you next.
That clarity breeds confidence.
Step 4: Personally reach out to your boss and your boss’ boss (if appropriate) to indicate your concern and show that you share their frustration
In Why your least favorite manager got promoted, I wrote:
Imagine that you missed your targets this quarter by something small — let’s say 5% is small in the context of your org. The CEO asks you what’s up. Which of these responses is going to build more trust long term?
“We just don’t have the resources to hit this target. Sales keeps asking us for things and the devs said they would release this feature but didn’t, which threw us off. Either we need to not get all these other asks or we’re going to need more headcount.”
“I’m pissed we missed the target! I think we should be doing double that number. We’ve got to coordinate things to make sure we’re serving sales and devs in a way that lets us hit this target too — can I work with you on that?”
Show how much you care. If your boss feels like you care as much (or more) than they do, they’ll be calmer and more confident in what you do next.
I won’t always explicitly reach out like this — it depends on the size of the mistake — but if I do, I’ll reach out to either my boss or my boss’s boss (or both) and add to the messages from step 3.
Something to the effect of “I feel terrible that this happened, and it’s my fault” will do.
I send this type of message because that’s genuinely how I feel when I make a mistake, and because sharing that makes it clearer that I’m on top of things.
If you have a “difficult” boss (which I thankfully do not right now), this can also undercut the yelling and blaming and any other gross shit they might typically throw at you. If you say you feel terrible, it sort of takes the wind out of their sails and helps you focus everyone on the problem.
Step 5: Create a plan to show how you will avoid this mistake next time
I’ve omitted step 4.5: “actually fix the mistake,” because I think it goes without saying.
When the danger has passed, get to work on the long-term problem. People leave this step behind or delegate it to some sort of “post-mortem” that itself doesn’t have clear outputs. There are good post-mortems, but often they turn into people complaining, agreeing on things that don’t get fixed, and then repeating the process next time something goes wrong.
You can have a post-mortem and you can use your influence on the process to try to make it a good post-mortem and not a dumb one. But you should also make a plan yourself and talk to your boss about it.
In your plan:
Identify the components of your system that went wrong or failed
List steps that you personally can take — without anyone else’s buy-in — to avoid mistakes in the future (but be wary of “quick wins”).
Call attention to points of stress (weaknesses, but call them points of stress or tension) that exist between teams or that multiple people are involved in
Invite a discussion about those points of tension with your boss, out of which will come more next steps
Your plan should not be “try harder” or “double check next time.” You have to look for underlying system causes for mistakes. There’s a good chance you won’t make the exact same mistake just because you’re looking for it, so you need to find the weaknesses that contribute to this class of mistakes in general and solve that.
Oh also, your plan should be written out. With nice formatting. Subheads. Full sentences. Not just a list of bullets. If you’re in Google docs, hit “insert” and then “table of contents” to make it look really slick. You put work into this, and writing it up/formatting it shows that + makes people more likely to read it.
You do this because no one does it, and because you want to avoid making mistakes again. And if you do this, take ownership, and follow these steps, you’re going to come out on the other side of this mistake with more trust and responsibility than you started.