Hiring · Featured Debate

4 guests 7 episodes 3,058 words

The Firing Speed Paradox: Why the Most Empathetic Leaders Move Fastest

How quickly should you fire someone who is not working out?

Every manager has a story about the person they should have let go six months earlier. The underperformer whose mother had cancer. The loyal early employee who grew past their competence. The nice person who just couldn't cut it. In every case, the manager's empathy -- the very quality that makes them a good leader -- was the thing preventing them from doing the right thing.

Kim Scott calls it "ruinous empathy." Matt Mochary calls it fear. Elizabeth Stone calls it avoiding the keeper test. Uri Levine calls it the CEO's failure to make hard decisions. They all point to the same uncomfortable truth: most managers fire too slowly, and the person suffering most from that delay is often the underperformer themselves.

Seven voices from Lenny's Podcast -- spanning CEO coaches, Fortune 500 executives, serial founders, and leadership authors -- converge on a surprising consensus: the answer to "how fast should you fire" is actually "how fast should you give feedback." The firing timeline is almost always a symptom of the feedback timeline.

When someone on your team isn't performing, how quickly should you move to part ways? Is there a meaningful coaching window, or do leaders systematically wait too long and call it "giving someone a chance"?

The 7 Positions

1 / 7

Fire Quickly but With Grace -- Become Their Agent

Matt Mochary

Matt Mochary, CEO coach to leaders at Coinbase, Notion, Brex, and Clipboard Health, and author of The Great CEO Within, sees the same pattern in every company he coaches: leaders delay firing because they believe they are protecting the underperformer. They are not. The person stuck in a failing role is being denied the chance to find a position where their skills and passions actually fit.

Mochary's framework separates the decision from the implementation. The decision part is clear -- ask what the customer or the business needs and act accordingly. The implementation is where the humanity enters. He coaches managers to warn the person that a difficult conversation is coming, give them a few seconds to prepare so their amygdala is not hijacked by surprise, deliver the news, and then let them process emotions openly.

"The reason people are bad at this is because they think that they're hurting the person who they're letting go. I mean, how many times have I heard from someone, 'Yeah, this person's not performing, but gosh, they really need this job and their mother has cancer.'"
Matt Mochary
▶ 00:24:03

Then comes the critical step most leaders skip: become the person's agent. Not in the passive sense of "feel free to use me as a reference," but in the active, Michael Ovitz sense -- reaching out to your network, making introductions, spending one to two hours of your time helping them land somewhere better.

"I think for almost all managers that aren't good at letting go, it's because A, they've never done it, or B, they've done it badly. They didn't help the person, and that person then went off and had a very painful time and now hates that manager. But if you actively help that person, they will appreciate it."
Matt Mochary
▶ 00:27:25

Mochary's key insight is that this approach creates a positive feedback loop. Managers who have fired well -- who have seen the person land on their feet and even thank them later -- develop the capacity to fire at the right tempo in the future. Managers who have never fired, or who fired badly without supporting the transition, accumulate underperformers until the team drowns. Wei Deng of Clipboard Health, whom Mochary calls the best he has seen at this, separates stakeholder analysis from emotional reaction and acts on behalf of the customer every time.

When This Applies

when the underperformance is clear, feedback has been given, and the person is not improving. This is Mochary's, Levine's, Graham's, and Wilkinson's territory. At Waze, Levine expected this decision within 30 days of a new hire. At Tiny, Wilkinson acts at the first moment of doubt. At Netflix, Stone applies the keeper test as a regular mental exercise.

The 30-Day Rehire Test

Uri Levine

Uri Levine, co-founder of Waze (acquired by Google for $1.15B), nine-time company founder, 20-time board member, and author of Fall in Love with the Problem, Not the Solution, draws his conviction from a chilling pattern he observed across dozens of failed startups. When he asked founders why their company failed, about half said "the team was not right." When he asked the follow-up -- "When did you know?" -- the answer was terrifying in its consistency: within the first month. Every time.

"If you knew within the first month that the team is not right and you didn't do anything, the problem was not that the team was not right. The problem was that the CEO did not make hard decision."
Uri Levine
▶ 00:54:58

Levine's framework is brutally simple: every time you hire someone new, mark your calendar for 30 days. Ask yourself one question: "Knowing what I know today, would I hire this person?" If yes, go to them, tell them they are exceeding expectations, and give them more equity. If no, fire them immediately.

"Every time that you hire someone new, mark your calendars for 30 days down the road and ask yourself one question, knowing what I know today, would I hire this person? If the answer is no, fire them immediately."
Uri Levine
▶ 00:00:23

The hidden cost of inaction, Levine warns, is not just carrying deadweight. It is losing your best people.

"The top performing people would leave. They would leave because they don't want to be in a place that is unable to make hard decisions and they have a choice."
Uri Levine
▶ 00:56:02

He deliberately titled his book chapter "Firing and Hiring" rather than the conventional reverse, because firing is the hard decision and you must learn to make hard decisions first.

The Keeper Test -- But Build the Feedback Foundation First

Elizabeth Stone

Elizabeth Stone, CTO of Netflix and believed to be the first trained economist to serve as CTO of a Fortune 500 company, describes a system that sits between Mochary's urgency and the more cautious voices. Netflix's culture depends on high talent density as the foundation upon which candor, freedom, and responsibility are built.

"We can't really have any of the other aspects of the culture, including candor, learning, seeking excellence and improvement, freedom and responsibility if you don't start with high talent density."
Elizabeth Stone
▶ 00:00:00

The keeper test is the mechanism that maintains this density. It is a mental model every manager should apply regularly: "If this person told me they were leaving, would I fight to keep them?" If not, a tough conversation is overdue.

"The reason the keeper test and that question is useful is because no one wants to think that way. It's very hard to say to someone, 'I think this isn't the right fit.' So we have to introduce some of those reflections in order to encourage the behavior."
Elizabeth Stone
▶ 00:32:20

But Stone is adamant that the keeper test only works because Netflix has invested in continuous feedback. They do not have traditional performance reviews. Instead, they run an annual 360 feedback cycle purely for learning -- not tied to compensation or ratings. The real feedback flows in one-on-ones and project debriefs. By the time a manager concludes "I would not fight to keep this person," the person should already know they are struggling. Stone herself reports feeling less anxious in Netflix's transparent system than in previous roles where she was never sure where she stood.

When This Applies

when you realize you have been working around the problem without ever naming it. This is Kim Scott's and Alisa Cohn's territory. Before you can fire at the right speed, you must build the candor foundation. Have the conversation, state the deal breaker, set the timeline.

Radical Candor First -- 90% of Managers Never Even Get to the Feedback Step

Kim Scott

Kim Scott, author of Radical Candor (the most-referenced book on Lenny's Podcast, with over 1 million copies sold), former CEO coach at Dropbox and Twitter, and a faculty member at Apple University who led teams at Google, reframes the entire debate. Before you discuss firing speed, you have to address the fact that most managers never give honest feedback in the first place.

"By far and away the most common problem occurs when we do remember to show that we care personally, but we're so worried about not hurting someone's feelings that we fail to tell them something they'd be better off knowing in the long run. And that is what I call ruinous empathy."
Kim Scott
▶ 00:05:48

The scale of this problem is staggering:

"I would say 90% of us make 90% of our mistakes in that ruinous empathy bucket. It's by far and away the most common problem."
Kim Scott
▶ 00:07:11

Scott is not arguing that firing should be slow. She is arguing that the firing timeline is a symptom. A manager who has been practicing Radical Candor -- caring personally while challenging directly -- has already ensured the person knows where they stand. Improvement timelines are clear. The decision to part ways becomes a natural conclusion rather than a shock. The firing conversation becomes easier because it is not the first conversation; it is the last in a series.

Firing Is a Skill as Important as Hiring

Molly Graham

Molly Graham, who held early operational leadership roles at Facebook and Google and now runs Glue Club, a community for operational leaders, makes the organizational case. Companies invest enormous energy in hiring -- the right interview process, the right sourcing strategy, the right evaluation criteria -- but almost no energy in developing the complementary skill of recognizing when someone does not belong.

"Firing people is as important as hiring people. Getting good at identifying when someone does not belong or someone is not going to work out is actually a skill and being good at it as a company and as a leader is as important as identifying the right talent."
Molly Graham
▶ 01:03:11

The math alone demands this skill. Graham observes that the best hirers in the world are right about half the time. That means half your hires will not work out, and if you cannot fire effectively, you accumulate what she calls "barnacles on a ship" -- people who are not pushing the team forward, creating drag that slows the entire organization.

"Serve the business, not the people... everyone is better off if this company is wildly successful."
Molly Graham
▶ 01:05:05

Her mantra is not a dismissal of people's feelings. It is a recognition that the best way to serve everyone -- including the person who is not thriving -- is to make the decision that allows the business to succeed.

When This Applies

give feedback early, give it often, make consequences explicit, and if you do part ways, invest in the person's transition. Become their agent, not their adversary.

If You Have Doubted Once, Act Immediately

Andrew Wilkinson

Andrew Wilkinson, co-founder of Tiny (a holding company that has acquired and operated 75+ businesses), offers the most aggressive heuristic on the spectrum. Across dozens of companies, he has observed a simple diagnostic signal: when someone is a superstar, you cannot imagine firing them. When you start wondering whether someone is going to work out, the answer is almost always no.

"The kind of heuristic I have is if I ever think, should I fire this person even once, I should fire them immediately. It has always been a signal, at least for me, that when someone's a superstar, I can't imagine firing them."
Andrew Wilkinson
▶ 00:38:12

Wilkinson also challenges the common advice to hire for potential and coach people into roles:

"I've just found that I've never been able to change someone. You can never mentor someone out into being a good employee."
Andrew Wilkinson
▶ 00:38:12

He compares this to romantic relationships: it never works out if you marry someone thinking you can change them.

Check Whether the Performance Issue Is Actually a Leadership Failure

Drew Houston

Drew Houston, CEO and co-founder of Dropbox (which he built from a Y Combinator startup to a publicly traded company with over 700 million registered users), adds a diagnostic step that every other voice implicitly assumes has already been completed. Before pulling the trigger, examine whether the underperformance is actually a systems failure.

"Ben Horowitz calls it the Product-CEO Paradox, where the first way that companies die is from founders not letting go. And so you need to learn to scale yourself and hand off your responsibilities and operate at a higher level of abstraction. But then the second way that companies kill themselves is the founders get too far away."
Drew Houston
▶ 00:51:19

Sometimes the person is not underperforming -- they are under-supported. The founder got too far away, the context was unclear, the role was poorly defined. Firing in those cases does not solve the problem; it just resets the clock until the next person fails in the same broken system. Lenny's Newsletter reinforces this in "Five principles for successfully managing managers," where Saumil Mehta (former CPO of Square) argues that all manager challenges are either directly or indirectly about the skip lead -- the manager's manager who hires, reviews, and coaches them, or fails to.

When This Applies

when multiple people have struggled in the same role, when the person was recently a strong performer, or when you (as the leader) have been too distant from their work. This is Drew Houston's territory. At Dropbox, many performance issues traced back to founders who had over-delegated and lost context.

Evidence from the Archive

Mochary Method

Wei Deng of Clipboard Health uses stakeholder-first decision-making: identify what the customer wants, then address...

Wei Deng of Clipboard Health uses stakeholder-first decision-making: identify what the customer wants, then address each person who gets hurt in the implementation

Matt Mochary · CEO Coach ▶ 00:24:03

Netflix

Netflix does not have traditional performance reviews -- instead they run an annual 360 feedback cycle purely for...

Netflix does not have traditional performance reviews -- instead they run an annual 360 feedback cycle purely for learning, not tied to compensation or ratings

Elizabeth Stone · CTO ▶ 00:31:10

Radical Candor

Scott's boss at Google told her 'When you say um every third word, it makes you sound stupid' -- this direct...

Scott's boss at Google told her 'When you say um every third word, it makes you sound stupid' -- this direct feedback led Scott to work with a speech coach and changed her career trajectory

Kim Scott · Author and former CEO coach at Dropbox, Google, Apple ▶ 00:05:48

Waze

One of Levine's portfolio companies gave annual bonuses and four people received nothing -- the CEO had known for...

One of Levine's portfolio companies gave annual bonuses and four people received nothing -- the CEO had known for months those four were underperforming but had not acted

Uri Levine · Co-founder ▶ 00:00:23

Lambda School, Facebook, Google

Graham observes that most leaders in Glue Club are people-pleasers who get 'tangled in the people' -- the emotional...

Graham observes that most leaders in Glue Club are people-pleasers who get 'tangled in the people' -- the emotional difficulty is the main barrier, not analytical uncertainty

Molly Graham · COO / VP Operations ▶ 01:03:11

Independent

Cohn coached a leader who avoided giving feedback to a direct report for 15 years. When he finally did, the person...

Her termination script: 'We talked about this multiple times. You haven't made these changes. We're going to part ways.'

Alisa Cohn · Executive Coach ▶ 00:31:05

The Synthesis

Seven voices, ranging from "fire at 30 days" (Levine) to "check the system first" (Houston), converge on a framework that is more unified than it first appears. The disagreement is not really about speed. It is about sequencing.

Four-Step Sequence

Is the disagreement about speed or sequencing?

Ruinous Empathy Root

Why do 90% of managers never give direct enough feedback?

System Diagnostic

When is underperformance a systemic problem rather than a people problem?

Firing Muscle

How does helping someone leave well make you better at future decisions?

The disagreement is not about speed -- it is about sequencing. Step 1: give radically candid feedback at first signs. Step 2: have the explicit deal-breaker conversation. Step 3: verify the person has the right context, support, and role fit. Step 4: if no improvement, part ways quickly and generously.

90% of managers never reach the point of giving radically candid feedback. They work around the problem, assign lighter work, pick up slack, and hope it resolves itself. This ruinous empathy is the root cause of nearly every 'fired too late' story.

If multiple people have struggled in the same role, the problem is almost certainly structural. Fix the system before firing the person. Sometimes the underperformance is systemic -- wrong context, insufficient support, or poor role design.

Helping someone find their next role is not just ethical -- it makes you capable of firing at the right speed in the future. Managers who have fired well develop the muscle. Managers who have fired badly or never fired accumulate underperformers until the team drowns.

Which Approach Fits You?

Answer 3 questions about your situation. We'll match you to the right approach.

Question 1

Have you given this person clear, direct feedback about the performance issue?

Question 2

If this person told you they were leaving, would you fight to keep them?

Question 3

What is preventing you from acting?

Notable Absences

The Bottom Line

The non-obvious insight that ties everything together: **helping someone find their next role is not just the ethical thing to do -- it is the thing that makes you capable of firing at the right speed in the future.** Managers who have fired well develop the muscle. Managers who have fired badly or never fired at all accumulate underperformers until the team drowns.

**Step 4: If improvement does not happen, part ways quickly and generously.** Mochary's "become their agent" approach, Levine's 30-day test, Graham's "serve the business" mantra, and Wilkinson's "doubt is the signal" heuristic all converge here. Once you have done steps 1-3, further delay hurts everyone: the underperformer (who is stuck in a failing role), the team (who carries the slack), the top performers (who leave organizations that cannot make hard decisions), and the manager (who burns out compensating).

Sources

Matt Mochary — "How to fire people with grace, work through fear, and nurture innovation" — Nov 10, 2022
Uri Levine — "Lessons from a two-time unicorn builder, 50-time startup advisor, and 20-time company board member" — Jun 9, 2024
Elizabeth Stone — "How Netflix builds a culture of excellence" — Feb 22, 2024
Kim Scott — "Radical Candor: From theory to practice" — Dec 10, 2023
Molly Graham — "The high-growth handbook: Frameworks for leading through chaos, change, and scale" — Jan 4, 2026
Andrew Wilkinson — "I've run 75+ businesses. Here's why you're probably chasing the wrong idea." — Jul 3, 2025
Alisa Cohn — "Scripts for difficult conversations" — Jan 5, 2025
Drew Houston — "Behind the founder" — Jan 9, 2025