In a Q3 talent review, the room did not care that the manager liked the employee. The judgment was simple: define the output gap, say it plainly, and give one short runway with dates. If the miss persists after 30 to 45 days of clear feedback, move from coaching to consequence, because ambiguity is what burns the bridge.
First-Time Manager: How to Handle an Underperformer on Your Google Team Without Burning Bridges
TL;DR
In a Q3 talent review, the room did not care that the manager liked the employee. The judgment was simple: define the output gap, say it plainly, and give one short runway with dates. If the miss persists after 30 to 45 days of clear feedback, move from coaching to consequence, because ambiguity is what burns the bridge.
Running effective 1:1s is a system, not a talent. The 0→1 PM Interview Playbook (2026 Edition) includes agenda templates and question banks for every scenario.
Who This Is For
This is for the first-time Google manager who inherited a direct report with missed handoffs, weak follow-through, or uneven quality, and now has to correct the problem without turning the relationship into a feud. It is also for the manager who still confuses being liked with being effective. If you need a performance conversation, not a pep talk, this is for you.
How do I know this is underperformance and not a bad fit?
Treat it as underperformance only when you can name the expected output, the actual output, and the gap for the last 30 days. If you cannot do that, you do not have a performance case yet, you have a vague discomfort.
The problem is not your answer. It is your judgment signal. In a calibration-style debrief I sat through, the manager kept saying the engineer was "not strategic." The room pushed back immediately, because "not strategic" was not evidence, it was a mood.
Not a motivation problem, but a standards problem. That is the first filter. If the person is trying but still missing deadlines, producing buggy work, or needing repeated rescue, the issue is usually clarity, capability, or capacity.
Not a personality flaw, but a role mismatch. That is the second filter. If you gave them the wrong scope, the wrong ambiguity tolerance, or the wrong level of independence, the bridge is not burned by the employee, it is burned by the manager who pretended the assignment fit.
On a Google team, the politics are quieter than at smaller companies, but they are not softer. The team will tolerate hard feedback. It will not tolerate a manager who cannot distinguish one bad week from a pattern.
A real pattern has repetition. One missed deliverable is a data point. Two missed commitments with the same explanation is a signal. Three is a management failure if you have still not named the gap.
What should I say in the first conversation?
Say the gap, the impact, and the date. Do not start with comfort, do not start with a story, and do not start with "help me understand." Start with reality.
The first conversation is not a therapy session. It is a specification. In a manager conversation I heard in a product org, the lead opened with "I know you're overloaded." The employee heard an excuse to continue. The team heard indecision.
Not empathy first, but clarity first. Empathy still matters, but after the facts. If you lead with softness, many people hear permission to stay vague.
Use one sentence for the gap. Use one sentence for the impact. Use one sentence for the next checkpoint. That is enough. Anything longer usually means you are trying to cushion your own discomfort.
The direct version sounds like this: "Your last two launches missed the committed date, and the status updates did not surface the risk early enough. That is affecting the team because dependencies are slipping. I need a different pattern by next Friday."
That is not cruelty. It is professionalism. The bridge stays intact when the employee can tell exactly what changed and exactly what success looks like.
If they explain context, listen once. If they drift into excuses, redirect. If they take responsibility, stay measured. The first conversation should end with a concrete follow-up, not a shared feeling.
The worst first conversation is the one that feels kind and leaves nothing on paper. That is not kindness. That is delayed conflict.
When do I coach, and when do I move to a formal plan?
Move to a formal plan when the same miss survives a clear correction and a short runway. Thirty to 45 days is usually enough to see whether the person can close the gap when the expectations are unambiguous.
A PIP is not the first move. It is the formalization of a conversation that should already have happened. If you jump straight to process, you look punitive. If you never move to process, you look evasive.
In a Q3 debrief I watched, the hiring manager kept asking for "one more quarter." The VP asked a simpler question: which behavior will look different in 30 days? The room went quiet, because that was the real test and nobody had an answer.
Not a rescue mission, but a management obligation. That is the shift. The point is not to save the relationship from discomfort. The point is to stop pretending the problem will dissolve on its own.
Formal escalation is appropriate when the employee has already heard the gap in plain language, you have given them a realistic chance to improve, and the same failure keeps showing up. At that point, more coaching without consequence is just managerial procrastination.
There is a second trigger, and it is harsher. If trust is broken, if commitments are being fabricated, or if the employee is hiding problems until the last minute, the bridge is already under strain. In those cases, you do not need more patience. You need more structure.
The first-time manager mistake is to delay because the relationship feels personal. It is personal. That is exactly why standards matter. The more personal the relationship, the more damaging it is to stay vague.
How do I keep the team trust intact while I work the issue?
Protect the team by changing load and expectations, not by exposing the person's struggle. The team needs reliability, not a public diagnosis.
Not transparency, but gossip. People often think "being open" means narrating the employee's weakness to everyone around them. It does not. The team does not need the private reasons. It needs a credible plan.
If the underperformer owns a critical path item, move coverage before the deadline slips again. If you keep the scope the same and hope for the best, the rest of the team will read that as denial. Engineers are unusually good at reading managerial avoidance.
A bridge is usually burned in two places. The first is with the employee, when they realize you were not direct. The second is with the team, when they see you protecting optics instead of quality.
In a product org meeting I remember, the manager told the team, "Everything is fine, we are just rebalancing priorities." Everyone knew that was false. The issue was not the rebalancing. The issue was the insult to their intelligence.
Say enough to be credible. Say less than would become gossip. That is the line. If someone needs to know why a dependency changed, tell them the dependency changed. If they do not need to know more, stop there.
The manager who preserves trust is not the one who makes the story sound good. It is the one who makes the operating reality predictable.
What should I document so I do not burn bridges later?
Document the gap, the dates, the feedback, and the next checkpoint. If you cannot reconstruct the conversation two weeks later, you did not manage it, you merely discussed it.
Not a case file against a person, but a memory for a future debrief. That is the right frame. The paper trail is not for punishment. It is for consistency when the room asks what changed.
Write down the exact behavior. Missed dates, broken handoffs, incomplete specs, low-quality analysis. Do not write "attitude" unless the attitude is tied to a concrete action that affected the work.
Write down the date of the conversation and the date of the follow-up. If you said "next Friday," then next Friday is the date. Vague follow-up language is how managers create false flexibility and later resent the result.
Write down what success looks like. Not "improve communication," but "surface risks 48 hours before a deadline." Not "be more proactive," but "send the draft review without being chased twice."
If HR or your manager asks what happened, you want a clean story. In a talent review, the manager who says "we've talked about it" without dates has already lost the room. The one who can name the pattern keeps credibility.
Documentation is not cold. It is the only way to stay fair when the relationship gets tense. Fairness is what preserves the bridge after the work relationship changes.
Preparation Checklist
- Write the gap in one sentence: expected output, actual output, and date.
- Gather three concrete examples from the last 30 days. Anything less is noise.
- Decide the next checkpoint before the first conversation. Do not improvise the timeline.
- Separate skill, scope, and will. If you mix them together, you will diagnose the wrong problem.
- Align with your manager or HRBP on the escalation threshold so you do not freeload on ambiguity.
- Work through a structured preparation system (the PM Interview Playbook covers performance cases, calibration language, and real debrief examples that sound like actual manager conversations).
- Rehearse the hard sentence once, then say it plainly. Do not hide behind a long explanation.
Mistakes to Avoid
- BAD: "You're trying hard, so let's give it another week." GOOD: "The issue is missed commitments on X and Y, and we review again in 14 days."
- BAD: "The team just needs to be patient." GOOD: "I am adjusting the load so the team is protected while we fix the gap."
- BAD: "If I push too hard, I will damage the relationship." GOOD: "If I stay vague, I will damage the relationship more slowly and more completely."
FAQ
- Should I tell the employee they are at risk?
Yes. Hiding the risk does not protect the relationship. It stages the surprise. The bridge usually burns when the employee realizes the manager knew and said nothing.
- How long should I coach before escalating?
Usually 30 to 45 days after the gap is clearly named. If nothing changes inside that window, more conversation without consequence is avoidance, not leadership.
- Can the relationship still be intact if they eventually exit?
Yes, if you stay specific, fair, and boring. The exit only becomes toxic when the manager turns a performance conclusion into a moral verdict.
Ready to build a real interview prep system?
Get the full PM Interview Prep System →
The book is also available on Amazon Kindle.