Letting Go

“Our goal is to inspire every person at Netflix to be excellent at their job. If you would not fight hard to keep someone at Netflix, we want to make it safe and quick to let them go.”

— Reed Hastings, No Rules Rules

The early morning playbook

It’s 8:30am on a Monday; I’m never at the office this early, everybody knows I’m a night owl who rolls in dangerously close to lunch. A few employees are already here, and they keep looking at me. Shit, I think they know something is up, that there’s no reason for me to be here at this hour, unless…

Yeah, someone is getting fired.

It’s not their first rodeo. I show up early, my cofounder is here with me as my co-pilot with all of the paperwork ready. The conference room is booked well in advance. My anxiety is through the roof, my hands are frigid and sweaty, my heartbeat is pounding, face red from the adrenalin. It’s not my first firing, but that doesn’t make it all that much easier.

Saying the words isn’t hard, I’ve rehearsed them more times than I would like to admit. It’s the after I’m worried about. Will he blow up? Cry? Argue? How awkward will it be as we walk the guy out of the office right after? And how will the speech I will give after this to the team land? Will my employees lose faith in me for having been unable to make this relationship work? Or will they be relieved he’s finally gone, after having experienced the same friction themselves for months?

I just want to be done with this. Everything is set, we’re just missing Henry, the full-stack engineer who’s unfortunately about to have a bad day.

Then, I spot him.

“Hey Henry, do you mind popping into the conference room for a sec?” My cofounder is already there. This is so out of the norm; it’s got to be obvious to him and everybody else at the office what’s happening, no way he’s not starting to put the pieces together. People are staring. The conference rooms are all glass, it’s now a full spectacle. “Look, I know this isn’t the news you wanted to hear today, but unfortunately we’re letting you go, effective immediately.”

He turns pale. The earlier anxious attempt at a polite smile is quickly replaced with suppressed surprise, trying a last-minute poker face to hide the emotional impact of the news. 1000 thoughts go through Henry’s head. What is he going to do now? Does he get to keep his stuff? Will he be okay financially?

The reality of firing

There comes a time in the journey of every startup founder or CTO when it finally dawns on you that you’re going to have to do something you’ve been avoiding for weeks, possibly months. Not just fire somebody, but fire someone who you personally hired, who you did your best to attract to the company, whose joining you celebrated with the team, full of hope for this relationship enriching the collective and bringing you closer to realizing your ambitions, together. But who ultimately, either for one simple reason or for a myriad of different cuts, is going to have to go. And you’re the one who’s going to walk them out the door.

It blows. Firing is the shittiest part of the job. Every part of you that doesn’t want to make people feel unwelcome, humiliated, dishonored, discredited and ostracized screams at you to avoid this act. You’re a nice guy. You vastly prefer harmony to conflict. You want people to like you.

You spent countless hours trying to make it work, but at some point realized that the opportunity cost of fixing the fit of this employee combined with your bandwidth and the runway of the company meant this was simply not going to happen. You neglected your best performers for months trying to patch this hiring screwup, and you still have nothing to show for it. You are now faced with the inevitable.

Letting someone go sucks, a lot. Startups aren’t a clock-in-clock-out gig that you do for a few months and then replace with another one. They’re an all-in multi-year emotional dive into a cause, a tribe, a commitment to being in the trenches with a small group of people, doing superhuman feats under ludicrous deadlines. You weather losses together. You share epic wins. You break bread. They’re your tribe, people who share similar goals and values, comrades in arms in the technology industry race for survival. Having that bond suddenly severed against your will is a shock, not to mention the economic impact of needing to find income, a new healthcare arrangement, tell your spouse your plans might need to change, maybe you have to move somewhere to deal with the sudden employment change.

It’s a big life change, and you didn’t have a say in it.

But because of those bonds, you have to prioritize the team over any one individual. Your fiduciary duty is to shepherd the company you built yourself through the maze to a meaningful outcome for yourself, your employees, your customers and your investors. If a team member is holding the company back and there’s not much you can do to change that in the short term, they have to go. Being friends with everybody would be great, but you have responsibilities to your company and being everybody’s best bud is, unfortunately, not one of them.

To fire someone is one of the heaviest decisions a leader or manager will make. You may want to say it’s not personal, but it is. Or, it will feel that way to the person being fired. It has a real impact on people’s livelihoods, their well-being, and their sense of self-worth. Any leader worth their salt should not take that lightly. And yet the consequences of not firing someone are at least as serious and severe. Avoiding the painful, in the process harming the prospects of the larger collective, is a road to hell paved with the best of intentions. I want to emphasize the gravity of letting someone go. But the weight of that act cannot override a leader’s responsibility to keep the right people on board and see those who aren’t fitting out, regardless of how uncomfortable it feels in the moment.

Navigating this sensitive moment—both for the company and the soon-to-be-former-employee—is about more than managing a person’s emotions in the moment. Yes, treat the departing member with empathy and kindness, but also make sure the company understands why this was the right decision. You may even garner respect for holding the team to high standards.

Avoid letting a struggling member go and you weaken your momentum and build a reputation for preferring harmony over effectiveness. Let someone go with cruelty and callousness, and now others will expect the same when it’s their turn, and will not want to work for you. Screwing this up can make-or-break a team.

Not all departures are the same. Letting someone go for performance is a fundamentally different conversation than a layoff driven by runway, a cultural misfit, or a mutual recognition that the role isn’t right. Each has its own playbook, its own emotional texture, its own way of going wrong. In my experience, our teams were good at filtering for culture and role fit during hiring, and we were slow enough with headcount that runway-driven layoffs never came into play. Practically the entirety of my firing experience has been around performance, and that’s what the rest of this piece is about: parting ways with someone who isn’t performing, when you don’t have the luxury of time to wait it out.

Reluctance to fire

The harsh truth is, many managers out there don’t let people go often enough. Talk to anybody who’s worked at a poorly run startup, or in large orgs that have allowed standards to relax, and you’ll hear about plenty of contributors and managers who are a net negative to the team. But the managerial incentives to fire are often against you in these circumstances. Building a big team looks impressive on one’s resume. Number go up. It sounds stupid but it’s true. Firing someone is unpleasant and uncomfortable, an admission of managerial defeat. You can always give them a warning and postpone doing anything for a quarter. Another quarter rolls in, and maybe you delay it by one more. Maybe you put them on a PIP. Or maybe they get shuffled to some other team and the cycle continues. Either way they’re going nowhere, especially if the performance is “okay”. You didn’t have to fire them, looking bad, and your underperformer simply made a perfectly normal lateral career move; your hiring was flawless, you had no duds, clean track record.

According to Leadership IQ’s landmark study, 46% of newly-hired employees fail within 18 months, while only 19% achieve unequivocal success. That’s nearly half of all hires not working out. And yet many teams go years without letting anyone go, which suggests either they’ve cracked the code on hiring (unlikely), or they’re holding on to people they shouldn’t be.

Performance management as a dark art

Performance management is usually tricky. If someone is truly abysmal, letting them go is trivial. Look at their work for the last month or two, they didn’t ship a single thing, lots of good explanations of why everything was stopping them while their coworkers doing similar work were firing on all cylinders. Clearly poor performance, off they go. Any manager with a pulse can do that.

But firing someone whose performance is a 6 or 7 out of 10? Nightmare fuel. They’ll occasionally do good work and you’ll feel vindicated, confident; the right person is in the seat. Then suddenly they will struggle, throwing you off. The performance is inconsistent, it’s never terrible, but you’re never impressed either. The stars are obvious, they’re unstoppable machines that demand constant feeding of quality work. But these in-betweens, they’re contributing somewhat, but never consistently enough where you can place them in the “terrible” or “stellar” buckets.

You’re faced with a tricky dilemma: do I invest in them, see if I can bring their performance up? Maybe they can ship more. Or maybe the quality of their work will go up? Or maybe I can teach them to ask better questions so that their deliverables require less rework. How long do I run this experiment for? A month? Two? And what if it’s just a temporary boost under pressure, but as soon as you stop looking, they regress back?

In a larger, well-established org with less growth to go around and more need for stability and manpower on the less glamorous tasks, this type of performer might be fine. Do you really need the MIT PhD, with twelve Claude Codes running in parallel, a long tenure at OpenAnthropini and a stack of Zyns to maintain an internal employee portal? Probably not. It’s unlikely you’d be able to hire and retain someone truly hungry, craving bigger and bolder challenges, to stick to that kind of work.

But startups are different, everything is growing, everything is new. Your 1st, 10th or even 50th engineering hire all have a real impact on what your team is going to look like for the rest of the company’s life. Hire a bunch of “The Best” and give them ambitious projects to bite into, score a few public wins, and you’ll have other stellar candidates knocking on the door, seeking glory. But hire all people who are “good enough”, and nobody phenomenal will want to be associated with you. They want to be on a winning team, and this one ain’t it, no different from any other team sport. The best want to learn, be inspired, keep up with peers they respect and maybe even envy a little.

Having a high level of talent density actually makes your job much easier: anybody who’s not at the now customary high level is naturally visible to the eye of a manager looking for clear performance indicators.

The earlier you are in the company’s lifecycle, the more essential it is that everybody is actually great. More shots on goal, more validation cycles, more and better ideas about how to deliver what customers are looking for.

It’s generally a good idea to review your new hires around 60 to 90 days in, while they’re still fresh to the company. This is the best time to make a keep-or-part-ways decision, before everyone’s judgment gets clouded by familiarity with the person. Do some variant of Netflix’s Keeper Test: if a member of your team told you they were leaving for a similar role at a peer company, would you fight hard to keep them? If the answer is meh, you have your answer.

Timing

Early stage startups operate at a totally different speed compared to the rest of the industry, and it’s not just about shipping velocity. Your rate of managerial decisions has to keep up as well. You may only have 9 months in the bank left. Every individual contributor’s month’s worth of work is a terribly precious resource, and a 10x contributor will generate so much more valuable change than one operating at 70% capacity. On a team with 3 developers, that’s a massive hit to the company’s productivity. On a team of 20, it’s more tolerable, but the discrepancy will still be plenty visible to anybody with eyes. On a team of 100, one contributor lagging behind is insignificant, but now the question is how often this is occurring. With more overall capacity, the slack of low performers is more easily picked up, leaving them unnoticed. How many are struggling without you realizing it?

So, what now? You can’t wait a quarter. You can’t put them on a PIP. You have a few weeks, maybe a month to provide feedback and monitor for changes, and after that it’s time to go. There’s no time for training someone up, the company may be dead within a few quarters. Maybe a stern talking-to and pointing out all the ways in which they should be performing better will work within a month, but I’ve never seen it myself. If you’ve already gotten that blunt in your feedback, it’s unlikely more coaching and time will make much of a difference, but it may be worth burning a few weeks on it. Either way, it’s very likely you won’t be able to fix anything, and that’s ok. You’re not in the fixing business; you’re in the “solving paying customers’ problems before running out of money” business.

But no matter how certain you are, it will always suck to part with an average performer because every part of you will crave more time, more certainty, further confirmation that you’re making the right call. Your company’s bank account, however, disagrees. It demands efficiency yesterday, and keeping someone on the payroll who isn’t lifeboat-grade at this stage of the company’s life only makes you more likely to sink.

How much rope you give someone should be proportional to what they’ve already proven. As David Reis, a trusted authority within the Y Combinator CTO community, puts it, if someone is brand new, a few weeks of bad performance is enough signal. You don’t have history to draw on, and the 60-to-90-day window is exactly when you should be making this call. But if someone has been your right-hand person for years and hits a rough patch, they’ve earned months of patience. The track record matters. Then there’s the gut check underneath all of it: fire as soon as you stop believing they can improve. Not when the spreadsheet tells you to, not when HR’s timeline says so, but when you’ve lost faith that the trajectory will change.

Signs for engineering leaders

How do you know it’s time to pull the plug? Software unfortunately is not an easily measurable profession. Lines of code, numbers of tickets, code reviews, bugs, stories, complexity points… they’re all rough proxies for value delivered. Even if you find a couple of metrics to zero in on, sooner or later Campbell’s Law will catch up to you and they’ll cease to be useful. Ultimately, especially if you’re doing something new and unproven, you have to follow your judgment as a craftsperson and domain expert to evaluate another expert’s work. Many years ago when I was first wearing the hat of the Head of Product at Freckle due to a sudden departure of its previous holder, I had the privilege of being coached by Hope Gurion in all things leading PMs, and I never forgot this quote from her, paraphrasing: “good product people follow their hearts, but don’t forget to run the numbers”. I think the same can be said for great managers who are determining how well an employee is performing. You shouldn’t purely vibe your way through this decision, but the numbers alone also won’t tell the whole story. Especially if the performance is not truly atrocious, the analytics will not clearly tell you what to do. The secret sauce is in combining the spreadsheet with the gut.

Here are some questions to ask yourself to help you arrive at an answer: Look at what they shipped in the last couple of weeks. Are you impressed? Is it a large volume of valuable work, or perhaps low volume but high-sophistication work? Were they solving high-complexity problems? Did the work break anything, or did it seamlessly slot in place and immediately create value? Is the work usable as reference and foundation for future work? Were corners cut, relationships hurt in the process, or debt sneakily accumulated without the team’s buy-in, requiring someone else to clean up?

If the individual isn’t delivering what you would define as great work, ask yourself why. It’s entirely plausible it’s leadership’s fault. It’s possible you did a poor job at selecting the bets for them to take, you didn’t communicate the urgency or the context, you were absent and blocked them from progressing when they could have. But also, did others on the team experience the same, and yet were able to still deliver, working around your limitations as the CTO or manager? Startups are often a mess, and people need to know how to move forward even when the track hasn’t been built yet. They’re often the ones building it.

Assuming it’s on them, take note. Provide feedback. If a few weeks later you’re still unimpressed, go through the loop again, providing further feedback. Season to taste, but if this cycle continues again, that’s a signal it’s time to part ways. If your company had predictable cash flow and years of reserves in the bank, maybe you could have justified persevering further, but the business is just not there yet.

I remember an instance of a senior engineer reporting to me who was the world’s friendliest guy and was everybody on the team’s favorite person while being a reasonable, average performer. But he would continually bury tech debt landmines in the codebase that would take an excruciating amount of time to fix due to being highly contagious. Fixing things built on top of those design decisions was taking weeks of time from an already emaciated team. Code reviews would catch many of them, but some would still slip through and wreak havoc.

I delivered the feedback. He said it made total sense and he was perfectly in agreement that this wasn’t okay. I thought we were done. Nope. It kept happening. Okay. I even went out of my way to write a doc listing out exactly what good work looked like in our codebase, making it clear that not doing this was a cardinal sin. Once again he’s in vehement agreement. Feedback received, boss. It happens again, only this time he sneaks it in without a code review without anybody noticing. I lose all hope, it’s time to go, I’m out of ideas as far as how to get my point across.

A much better manager and coach may have figured out a way, the one judo move to get them to grok the practice I was trying to instill, but I simply wasn’t that manager, and I wasn’t going to become him overnight. As usual I felt countless feelings of regret after letting the person go, wishing “I could have fixed them”. The sentiment, however, quickly evaporated once I found someone to replace them who ended up being 10x more effective in the role. Experiencing that shocking contrast as a CTO is an important wake-up call about how much of a difference the right hire can make.

I learned over time that if I’m spending a good chunk of my day thinking about an employee, day after day, it’s a strong indicator that I have to let them go. I’ve never been able to “fix” someone who started being constantly top-of-mind for me due to underperforming or becoming a source of demotivation for those around them. I’m not sure that even great managers can take on “long term projects” and “fix” them consistently. At this point in my career I’m convinced that almost everything is decided at hiring time, and if I have to overhaul someone after the fact, I’m guaranteed to regret attempting.

The unpleasant reality of great hires

At an early-stage startup, great employees aren’t grown, they’re found. That’s not a universal claim about human potential; people absolutely grow into roles given the right environment and enough time. But time is exactly what you don’t have. If you have 9 months of runway left, you cannot spend 6 of them developing someone who isn’t delivering today. If someone is not working out, it’s almost always a mistake at hiring time, not a lack of training. Yes, there are always exceptions where someone mediocre will turn phenomenal after months or years, but a startup burning through its seed round doesn’t have months or years to find out. Great startup teams are built at hiring time, not through coaching. Coaching will tune things up, enhance teamwork and efficiency, but it will not conjure a maverick out of someone average on a startup’s timeline. As I mention in What “The Best” Looks Like, you can’t pep talk your way into making someone hungry, they either are in that phase of their lives or they aren’t, and it’s not up to their manager to convince them to be.

Where things went wrong

Whenever you end up having to let go of someone, it’s important to go back to the sequence of events that led you there. Most people don’t do this. After all, looking at your failures feels bad; it’s simpler to move on and try again. However, it’s likely that whatever led you here in the first place will repeat again if you don’t revisit the process. Where in the hiring funnel did things go wrong?

The permutations of failure here are practically endless. It’s possible you sourced the wrong person, but your employees helping you screen assumed there was a reason why they were interviewing the candidate, and maybe it was their fault in not seeing something great in them. They move the candidate forward. It’s possible your interview stages actually don’t give you much signal, and are mostly performative. It’s possible the interviewing team is getting a shallow signal, or testing the same quality over and over. It’s possible they’re verifying traits that don’t correlate with what makes people succeed on your team. Your grading system may be too liberal, with lukewarm scores being treated as a sign to move the individual forward. There’s a chance the candidate changes too many hands in the process and people lose track of how well they’re doing. Like I said, it’s complicated, but very worth spending time on and refining over time.

Hiring is like poker. You can do everything right, shifting probability in your favor, but 1 in every 5 hires still ends up being a dud. Those are likely still better numbers than most hiring managers are getting. Remember that 46% of new hires fail soon after. Sometimes you do everything right and still lose. Avoid resulting.

One pattern to watch for specifically: if you’re firing multiple people in the same role or on the same team, stop looking at the candidates and start looking at the role, the team, or yourself. A string of “unlucky hires” is almost never that. Something structural is off. Maybe the role is poorly defined, the team dynamics are toxic, or your expectations don’t match reality. One firing is a datapoint in isolation. Two in the same area? That’s a trend to look into.

Common failure modes post-hiring

One of the most common mistakes made by managers earlier in their journey is to not communicate clearly enough that someone is doing poorly. You try to be gentle, subtle, not startle the person or send them into an anxiety tailspin in the hopes that they get the hint. In practice, often you end up being opaque to the hire and they don’t realize you’re saying they’re in trouble. If you’re not directly telling them that, it must mean their deviation from the expected standards is minimal at best and there’s not much for them to worry about. That’s what they think right up until the dreaded conference room meeting.

I find that the best hires appreciate direct, unminced feedback, they accept the challenge, own up to their mistakes and try to do better so that they never get that feedback again.

Patrick Lencioni’s approach here is sensible: be crystal clear and constant with letting the underperforming hire know that they’re falling behind and it’s critical for their tenure that they change the trajectory. If you’re persistent enough with this, to the point of embarrassment, either the hire changes course and the problem goes away, or they get sick of hearing you badger them and quit. A win-win.

You also need to be crystal clear about what excellent looks like. Sure, you could argue that it should be obvious, but people come from all sorts of backgrounds, including teams that had a totally different bar. There’s little performance standardization in the industry. At Double Dusk I had a sizable standards doc explaining what great work looked like for our engineers, so that there was no need to mind-read me. Of course the challenge of such a doc is keeping it small enough that it’s readable, but not so short and generic that you can easily argue any point within it. You’ll want to update, refine and curate these rules over time with the team and ensure everybody’s buy-in and familiarity with them, until it becomes second nature in the DNA of the firm.

Ultimately, I find that the quality of onboarding and hand holding in the early days is irrelevant by the 60-90 day mark, AKA the break point where you consider letting a new hire go. It’s easy to blame the company for not having supported someone enough, but the real go-getters always figure it out and don’t need scaffolding. They roll up their sleeves and get to work. Startups can often be chaotic well into the mid-to-late stages, it’s just the reality of a high-growth environment. There’s not a lot of support to go around and the best hires end up figuring out how to be useful quickly without needing much hand-holding.

Scaling past yourself

Everything above assumes you’re the one doing the managing: watching the work, giving the feedback, making the call. That works when you’re running one or two teams directly. But as the company grows and you’re managing through tech leads or engineering managers, your personal standards don’t automatically stay the benchmark. What was intuitive to you, how you evaluated work, when you decided someone wasn’t cutting it, what “great” looked like, now needs to be explicit enough that others can execute the same formula.

This is a whole topic unto itself, but the short version is: codify the standards, the rituals, and the evaluation practices that worked when you were doing line management yourself. The leaders reporting to you need to be on the same page about what good looks like at every level of the engineering organization, and they need to be actively applying it. Otherwise your standards won’t scale past the first team or two. If the company is growing fast, there’s no substitute for being explicit about how performance evaluation is done; you want some degree of standardization, or you’ll end up with pockets of the org where underperformance goes unaddressed simply because the local manager has different instincts or habits.

Cover your bases

This is not legal advice, but, at a 10-person startup burning through its seed round, you’re probably not going to get sued over a firing. You barely have a product, let alone the kind of deep pockets that attract adversarial employment lawyers. Most of the time, things are understood to be at-will, the conversation is straightforward, and everybody moves on. In the early days you don’t need to be meticulous about documenting every piece of feedback or lawyering up before each termination.

But as the company grows, so does the target on your back. The bigger you get, the more thorough you need to be; more employees means more surface area for misunderstandings, more potential for someone to feel they were singled out, and more reason for a lawyer to take the case on contingency. What was a casual Slack message at 8 people becomes a critical piece of your paper trail at 80. Make sure that as the stakes grow, your documentation grows with them: keep records of the feedback you gave, the timelines you set, the conversations you had. Even if it’s just a quick summary email after a 1:1, that goes a long way if things ever get contested.

Regardless of stage, have a go-to person on the legal front: an employment attorney who knows your state’s rules and can help you navigate this the right way for the type of employer you are. You don’t need them on speed dial when you’re five people in a garage, but you should know who to call and build a relationship with them prior to needing their services. A single conversation early on can teach you the basics: what retaliation looks like, how verbal promises made during hiring can come back to haunt you, and what documentation actually matters. It’s cheap insurance against an expensive mistake.

Parting ways with humanity

Well, you tried everything within your power and timeframe to improve things and it didn’t work, it’s time to go your separate ways. As I discussed above, this is going to be a big hit for the employee, there’s no point in sugarcoating it. Some employers go as far as congratulating the fired member on their “graduation”, which you, a non-sociopath, should absolutely avoid. Do it quickly, respectfully, with humanity, acknowledging the severity of what you’re doing, empathizing, but also being firm.

Have the logistics ready before you walk in: severance terms, final paycheck timing, benefits continuation, equipment return, and when their access will be revoked. All in writing, so they can process it later rather than trying to absorb details while in shock. Keep the meeting short, fifteen minutes at most. Don’t open with small talk, don’t relitigate the feedback you gave along the way, and don’t leave room for ambiguity about what’s happening. When someone pushes back—some absolutely will, nobody’s really themselves in this situation—don’t get drawn into debating the merits. “I understand this is hard, and I respect that you see it differently, but the decision is final.” Say it as many times as you need to.

The actual moment of you letting your employee know that they’re being let go is not the time to negotiate why, debate the merits of the decision, reach a compromise, even though some employees will want to. You’ve made your decision already, now is the time to clarify what the employee is walking away with.

Be generous with the severance, especially in the US where the safety net is rather thin. Be especially mindful if someone is on a work visa, as the consequences of job loss extend well beyond income. You may not have much capital to spread around, but you don’t want to build a reputation for throwing your former staff overboard at the worst time and putting them in a precarious situation. The failure of the relationship is in big part yours as well, your process led to this, so you should share the financial responsibility. Be generous with stock options expiration windows as well, if someone has stuck around long enough to earn their chunk of the firm, it’s only fair that they get to keep it even if they don’t have cash on hand to acquire the shares.

I remember the story of Michael Dearing running into an employee he had to fire at eBay many years earlier, getting a big hug from them, and being grateful for having done the right thing for them at the right time, which ultimately led to a much more successful career elsewhere where they actually were the right fit. Yes, that’s survivorship bias, but the point is that firing should not be personal, it’s a reflection of a mismatch and a mistake, not a condemnation of the employee’s character, skills, or potential. When I have to let someone go, I do my best to own the decision and what led up to it, the onus is on me even more than on them, and I aspire for no hard feelings long term.

If I have to part ways with an employee and they still recommend their friends to join us down the line, then I know I’ve done something right. The departing member felt like they were treated fairly, things just didn’t line up the way both parties had hoped.

Informing the team

Next comes the team itself. You’ll want to inform them as soon as possible.

First, you tell them why the person was let go. I like the Netflix approach here, respectfully and tactfully being honest about the reason, and how that relates to the values and standards of the company. By referencing those standards, you reinforce that they’re a real barometer for someone belonging on the team, not just something hanging on a dusty Notion page nobody ever looks at. Trying to be vague about why someone was fired because of discomfort or embarrassment just leads to further problems down the line and leads to only more Grapevine. Employees will want to understand what happened, and if they’re at risk themselves because of something they may or may not be doing. With no clear, authoritative word from management, they will be left to speculate for themselves, creating the sort of anxiety that could poison your whole team.

It’s then good for the leader responsible for hiring the departing individual to publicly take responsibility. Even better if they let the team know that they’ll do a post-mortem and figure out how to make the process more bulletproof. It shows that someone is accountable, and that someone will use this as an opportunity for the company to improve.

It’s then great for the leader to explain if anything will need to be done differently in the short term now that the individual is no longer there. Who’s going to pick up their responsibilities? Will the team need to work longer hours to make up for it? Will the timelines of the roadmap change at all in light of it?

Your own psychology

Having had to go through this process many times throughout the years, I’m well familiar with the psychology of it, how you feel after the fact. You wonder if you just crushed people’s morale by removing a friend they had gotten close to, if your team is questioning your ability to lead them, if you have what it takes to bring the right people on board to succeed in the mission. Do you actually know what you’re doing, or are you just LARPing the role? Will your peers on the leadership team think less of you? Is the CEO starting to wonder if someone else should be running the technology show instead of you? How many more of these screwups can the company stomach before it becomes a real problem?

All I can say is that these doubts are normal and, especially if you’re early in your career, healthy. The important thing to walk away from the situation with is the drive to improve your skills, the process, your communication around these issues, and try again. Ultimately, you can only control your own inputs into the system, and as long as you did everything within reason to learn from this event, you shouldn’t feel too down, it’s an unavoidable part of any position that involves hiring.

A pretty big chunk of hires never work out. Of course if you’re firing half of them, that should raise eyebrows. Equally, if nobody ever gets let go, that’s a real statistical anomaly, and you need to start wondering if your competence bar has gotten too low. Nobody can tell you what the right sweet spot should be for this, every team is different, and what is considered exceptional at one company could be considered barely passable at another. We have no real math for this as an industry, and that’s unlikely to change anytime soon. All we can do is do our best, and keep iterating.

What’s perhaps surprising to first-time CTOs is that many of the more successful members of your team will be relieved that the poorly-fitting member got let go. They might have even tried to highlight the dysfunction to you in the past, or they were personally affected by it as they were trying to make progress at their usual pace. They may feel like they have less cleanup to do after their former colleague, and that now they might be replaced with someone who will feel like a real addition to the squad, not a source of friction.

Conclusion

Henry took it about as well as you can take news like that. No blowup, no tears, just a quiet nod and a question about his health insurance. We walked him out, I gave the team speech, and by mid-afternoon the office felt different.

Nobody reached out to say they were glad I did it. People were respectful of the dignity of someone they’d worked alongside. But I could see the difference. In the days and weeks that followed, the team found a new gear. Things got snappier. People moved faster, unburdened from the cleanup that had quietly been dragging them down. And I finally had time for my strongest people, time to empower them instead of spending it agonizing over someone I already knew wasn’t going to work out. I was no longer walking around the office with a dark cloud over my head, ruminating about my bad hiring decision.

From what I can tell from his LinkedIn, Henry has been nothing but a solid success at other firms since. The mismatch wasn’t about his talent or character; it was about what we needed at that specific moment in the company’s life, and our inability to close the gap fast enough.

After all these years and all these firings, I still dread it. I still show up early with cold hands and a racing pulse. I’ve never become comfortable with it, and I hope I never do. Comfort would mean I’d stopped taking it seriously. What’s changed is that I no longer let the dread delay the decision. That’s the only real growth: not learning to stop flinching, but learning to act despite the flinch. The discomfort is the price of the responsibility, and if you’re not willing to pay it, you shouldn’t be in the seat.