The Best Approach to the Worst Conversation: "You're Fired"
Okay. So you’re growing fast. You started out with three people and now you have 47! You’ve picked up some funding. The product is coming together. You’re hiring so fast you don’t even know everyone you see in the halls anymore. People are jamming.
Then there’s Jeff.
If you’re at a growing startup, you probably already know him. He’s not a bad person. He’s not doing anything illegal or even anything blatantly wrong. But there’s something off. He’s not productive. He’s not a culture fit. He’s simply not right for your company right now. Whatever it is this person is doing — whether they’re named Jeff or not — it’s costing you, often much more than making a change. So what do you do about it?
Michael Lopp specializes in answering this question. Known to the tech blogosphere as Rands, Lopp managed engineers at Apple for over eight years and now works at Palantir, where he focuses almost exclusively on maintaining and retaining some of the world’s best technical talent (including how to make meetings suck less). Unsurprisingly, this job also entails identifying and managing people who don’t measure up — “the Jeffs of the world,” as Lopp puts it.
There’s no avoiding it — this is one of the worst, hardest things founders and managers ever confront. And while there’s no painless approach to performance interventions or letting people go, there is a way to go about it that dodges lawsuits, permanent scars, and — ideally — the need to fire anyone. Lopp presented this approach in an exclusive talk at First Round’s recent CTO Summit.
Rule #1: It’s Your Fault
If you’ve gotten to the point where your Jeff has become a real and recognizable problem, the bad news is it’s probably your fault.
"If you're lying awake at night thinking, ‘God, Jeff is pissing me off,’ your chance at a low-cost solution probably came about three months to a year ago," Lopp says. "You had the chance to improve things but you sat and waited until it actually started to hurt."
The longer someone works for you, the harder it is to change their behavior. The longer you don't deal with something, the less likely you can fix it.
More bad news: Your inaction is costing you more than you think.
“You’re probably focused on the work that isn’t getting done, or maybe how team morale is suffering because Jeff is kind of a jerk. Plus your sanity might be compromised because you’re waking up at 4 a.m. every night wondering what you’re going to do about the whole Jeff situation. You’re burning calories on this when you should be worrying about your product. This is the stuff you know,” Lopp says.
“But you’re not thinking about all the other people on your team who have problems with Jeff — Angela and Frank who know Jeff is a problem but every day see you not doing anything about it. I guarantee you, whatever you believe Jeff is costing you, it’s really three to five times that. The non-obvious cost is all the credibility you’re losing as a leader.”
When you’re a manager, no one is going to make the first move for you. People on your team may have hinted around it. They may have submitted subtle or even assertive feedback. But until you sit down with Jeff and try to turn things around, you’re the problem.
If you're leading a team, your job is your people.
Whether you're a founder, a CEO, or a manager, this should be your mantra. Your startup will only be as successful as the people who work there. And the key to keeping the best people doing their best work is to make sure they have the best leaders.
This means two things, Lopp says: Making sure managers get the training they need to be great at their jobs; and not passing responsibility for people and their performance to human resources. So, when Jeffs or prospective Jeffs pop up, it’s the team leads that should handle the situation, and they should be equipped to do a good job.
“Empathy is required for any conversation about performance or layoffs to go well,” Lopp says. “You can’t just ask HR to do it. Especially with engineers, they probably don’t even understand what the person does — and that’s kind of cruel, right? HR gives us the excuse to outsource things we don’t want to do, to avoid the hard conversations. And since most of us are engineers ourselves, we love this. But that’s a problem.”
To flip the script at Palantir, Lopp pushes team leads to take responsibility for everything related to performance. People Operations, Palantir’s brand of HR, is there to help, but the goal is that leads are empowered to handle performance issues.
“All leads get a boot camp on how to handle the things that will inevitably happen to them — they get toolkits, they get documentation,” says Lopp. “It’s worth spending the money to capture these best practices and make sure everyone who becomes a lead learns the same things.”
Particularly when it comes to letting people go, you’re stepping into a minefield. You have to be open to asking for help. Even when people are employed at-will there can be legal ramifications. There can be consequences for your brand. The list goes on. If you don’t have an HR person or agency, ask your investors, ask your advisors. If you even have an inkling you could get sued, ask a lawyer. Don’t wing it.
Ultimately, process will be your savior. “I know engineers hate the word process, but don’t underestimate its virtue,” Lopp says. “It’s the documentation of what works. Whether you have 40 or 400 people, it’s a handbook that says, ‘Here’s how to approach this very tricky situation, even if you’re an engineer who’s bad at this stuff.’”
Beyond 10 employees, it’s never too early to establish process. However, before that point, it doesn’t make sense to hang on to poor performers like Jeff. If someone like that makes up 10% or more of your company, you need to cut them loose asap — as cold as that sounds, Lopp says, there’s no doubt about it. It’s only when you get bigger that it makes sense to develop a protocol. After all, people can and do change.
“Eventually, someone will fall into this category where they need performance management, and you’ll want to have a process to refer to,” he says. “That way you can be transparent, and say ok, here’s exactly what we do. At a certain size, you’ll always have a set of people going through this. So figure it out now.”
Tackle Performance Head On
PIP is one of the worst and most-feared acronyms out there. Standing for “performance improvement plan,” a PIP usually takes the form of an official, written agreement drafted and overseen by HR that outlines exactly how an employee needs to very quickly get better at their job in order to keep it. They may be more common at large corporations than startups, but even new companies should be familiar with PIP principles to keep their staff on track, especially as they enter rapid growth. Lopp recognizes the need, but hates how they’re used: too often as a last-ditch, half-hearted effort to save someone’s job.
“There are two problems with how PIPs are used. First is that you should want to fix something as soon as you see it go wrong, not at the very end of a long, slow decline. And second, you can’t just throw a switch and fix everything. There isn’t just one or even a couple things you can do to make Jeff better. It’s not just one conversation. It’s a lot of little things that need to be addressed over months, every day, every hour. If you’re thinking about putting someone on a PIP, your first question should be what could you have done earlier?”
There’s a reason most people are surprised when their manager asks them to go on a performance improvement plan. Of course, people are biased toward denial. But even if they suspected something was wrong, it’s likely no one articulated it to them in a way that they understood and agreed to fix. People are also biased against confrontation.
Lopp recommends what he calls a pre-PIP — essentially an agreement made between a manager and employee to improve performance without signing anything with an unspoken “or else” at the end of it. This is even easier to implement at a startup that doesn’t have a formal PIP process. Here’s what the pre-PIP route looks like:
- Feedback needs to be immediate. As soon as someone steps off the path or veers into dangerous territory, let them know. Ideally during the first 90 days, give people “an exorbitant amount of feedback,” Lopp says. “Just think, you could have fixed it six or nine months earlier by pulling Jeff aside and saying, hey you really frustrated people in that last meeting because you weren’t listening.”
- Be very specific. Give granular examples of the mistake Jeff is making and how things would look different if he changed his behavior.
- When you tell Jeff that something is wrong, have him repeat it back to you until what he’s saying matches what you mean. Too often people think they know what is expected of them, but still fail to meet expectations.
- Take the threat out of it. One of the worst things about performance improvement plans is that they’re surrounded by an air of doom. This causes people to either push back and have a bad attitude, or feel hopeless and unable to put in their best effort.
- Still write things down. Even though this isn’t an official PIP filed with HR, it should be quantified and codified. “You should write a well-defined list of things that you can measure. Jeff should be able to see for himself that he is succeeding. You should be able to see the changes that result from this process.” Even if there’s something subjective Jeff should improve, try to put something quantifiable around it.
- Be patient. “Changing behavior is a lot of work. A lot of people assume it’s impossible. But by investing in feedback and giving your leads the ability to have hard conversations, you can do it, and it's often worth it.”
- Always transition the conversation from critical feedback to possible solutions. People naturally cling to the negative, so ending on a positive note could make the difference between motivating Jeff and convincing him to give up.
Pre-PIP training is a key component of Palantir’s bootcamp for team leads. The company even retains an external coach that comes in and trains people to give constructive feedback. There’s no time to be intimidated.
PIP for Real
Sometimes a pre-PIP just isn’t going to cut it. Sometimes you have to escalate to a full-fledged, documented PIP with the person’s job hanging in the balance. But there’s good and bad ways to go about this. Most importantly, it absolutely shouldn’t be a surprise.
“The conversation should never come out of the blue,” Lopp says. “It should always start with, ‘Hey look, we’ve been talking a lot about these things and measuring them, and we haven’t seen a whole lot of progress, so let’s look at our options.’ You can tell them that you’re starting to document things, but it still doesn’t have to be this big threat. You can be gentle about it.”
Lopp recommends a three-month plan with a checkpoint at the end of each month:
- Phase 1: “You need to be a bit of a jerk micromanager. Involve yourself as much as possible. You’re learning how they work and what they do. You’re being highly directive.”
- Phase 2: “Shift from tactical to strategic advice. Stop micromanaging. Give them the space to do things themselves, but keep watching. Make sure they know you’re engaged.”
- Phase 3: “In the last month, basically let the reins go and tell them you’re expecting them to do their job without any advice or help from you at all.”
The problem with this is that — especially if your Jeff is smart, which he probably is if he landed a job at your company — he’ll nail this plan. Maybe he’ll nail it all in one month. But there’s no guarantee that the problem is solved. This is one of the biggest liabilities with performance improvement plans.
50% of people who are put on performance improvement plans become repeat offenders.
When the same behavior surfaces six months later, you’re left with few options. A lot of managers and companies will repeat the PIP process, but Lopp isn’t optimistic. “The majority of the time, I just tell the person, ‘We’re not doing this twice. It’s great that you can rise to the occasion when you need to, but we’re going to have to let you go.’”
The Toughest Conversation
Most people who run companies have fired someone. But there’s a huge delta when it comes to experiences. You can let someone go, and they can leave feeling good about other opportunities. Or you can get sued. The reality is usually somewhere in the middle. Regardless, if you’re doing the firing, it’s going to hard, nerve-racking, negative.
“If you’re nervous about it, roleplay with someone you trust, someone you know well,” Lopp says. “At Palantir, we run this program called Mock Mystery Disasters where we actually walk team leads through situations like these so they can see how it’s done well.”
In addition to practice, he has several other tips to make letting someone go just a little easier.
For example, managers often overlook the rumor mill that spins up every time someone’s job is on the chopping block. Firing someone can be disastrous for morale, especially if the person getting fired has a tight inner circle.
“At Palantir, we make a point of crushing rumors. We not only talk to the person in question, but to the people who know them well. That said, remember that people love to gossip, so you need to be very thoughtful about what is the respectful amount of information to share. If people are talking about someone’s performance, you want to make it clear that you’re aware of the issue and investing in its resolution — that it’s being handled. If you’re actually letting someone go, then your team’s number one concern is going to be whether everyone is getting fired. You can allay those fears while respecting confidentiality. Develop a communications plan around it. That’s an important part of the healing process.”
On top of this, be sure to take comfort in the aftermath of your decision. Firing someone isn’t going to be pleasant. But once it’s done, recognize and appreciate the positives that come out of it. “It’s natural to be attached to people even if they're horrible. You might very well be unable to imagine a world without your Jeff there. But every time I’ve let someone go and it’s been awful, I’m always surprised by who comes into the vacuum left behind. Suddenly there’s David doing a great job at something you didn’t even know he could do.”
Fix It Before It Starts
“The number one learning if you’re a manager who needs to fire someone or put someone on a PIP is that you need to invest in fixing situations earlier,” Lopp says. “Think about the small things you could have done early on. Figure out how to nip those things in the bud next time.”
Most people consider their first 90 days on a job to be a “honeymoon period.” But you should bust this myth as soon as possible. “Those first three months are when you really get to understand people and what they’re about. This is the time when you should be paranoid and take the time to make those little adjustments when something feels weird. Don’t just let things slide.”
This might sound harsh. It might sound pessimistic. But it’s actually the opposite, in Lopp’s opinion. “Taking a tough stance from day one is all about believing in everyone’s ability to change,” he says. “It may not be drastic, but it may be enough to make a difference.”
Read These Next
80% of Your Culture is Your Founder
When Molly Graham joined Facebook in 2008, the company still felt scrappy. With 400 employees serving 80 million users, people were so busy “moving fast and breaking things” that the culture still needed to be defined. Graham was hired to help make this happen — to not only tell the company’s story externally, but to build a shared vision and identity as it grew from 400 to thousands of employees. She started by asking two questions: 1) Who do we want to be when we grow up? 2) How do we talk to people outside about what it’s like to work at Facebook? Since then, these two questions have formed the foundation of Facebook’s culture discussions. Most notably, they resulted in the “hacker” identity that has distinguished the company as a technology powerhouse that is always experimenting to bring the world closer together. At First Round’s recent CEO Summit, Graham, who managed Culture and Employment Branding at the company for two years — and who now runs business operations for slick, modern word processor Quip — talked about what startup founders have to gain from defining culture early and often, and how to do this when there are dozens of competing priorities.
Here’s Why You’re Not Hiring the Best and the Brightest
What’s the most consistent piece of advice you get as a startup? Always hire the best people. Never compromise in your hiring standards, no matter how big your company gets. And it’s true. A great team can take an okay idea and transform it into an incredible, world-beating product. But something has always bugged me about this advice. There’s an elephant in the room in the form of an implied clause: Always hire the best people… who are willing to live in San Francisco. Substitute Mountain View, New York, Boston, Chicago, or any other city. The problem is the same. We pay lip service to the idea of hiring the best people in the world — but in reality, we’re only hiring the best people who happen to be close by. Call me crazy, but I think if we’re going to talk about hiring the best talent available, we should actually try to do that. This means letting go of the idea, at least in IT work, that people need to be physically present for any meaningful work to occur. There’s a better way.
About 10 years ago, after joining Netscape as its youngest employee, Angus Davis learned how to fly airplanes. Along the way, he grew the company he co-founded — voice recognition startup Tellme — into an $800 million business bought by Microsoft, and became the founder and CEO of payment company Swipely. Recently, he sent an email to his company recounting some of his most important and relevant lessons from flying to help guide them in their approach to the business and their careers. After reading it, we thought it should be shared with even more people in the startup community. Davis generously agreed to adapt and re-publish it here.