Performance Management as a Practice

Supporting Growth and Careers was about the long arc — helping people become more over months and years. This closing page of the part is about the shorter, steadier loop that runs underneath that arc: how you keep the everyday work honest, on track, and free of nasty surprises. It ties the whole part together, because performance management only works when it stands on the motivation, trust, and coaching you have already built.

Most people hear “performance management” and picture a form, a rating, an awkward once-a-year meeting. That is the symptom of a broken practice, not the practice itself. This page reframes it: performance management is a continuous conversation, and the annual review is just a summary of conversations you have already had. Done right, the review is boring — because nothing in it is new.

The core rule: no surprises at review time

Here is the single rule the whole page rests on. If something shows up in a formal review that the person is hearing for the first time, the manager has already failed. Not the employee — the manager. Because a review is a summary, and a summary that contains surprises means the real conversations never happened.

Think about what a surprise at review time actually reveals. It means the manager saw a problem (or a strength) weeks or months ago, said nothing, and saved it up for a scheduled meeting. That is not fair, and it is not useful. It is not fair because the person had no chance to fix a problem they didn’t know they had. It is not useful because the moment to act on a mistake is when it happens, not six months later when the details are cold and the work is long shipped.

So the test of a good performance practice is simple: the annual review should be a formality. The person should be able to write it themselves and get it roughly right, because every point in it has already been said out loud, in the moment, when it mattered.

   THE BROKEN MODEL                    THE PRACTICE
   ────────────────                    ────────────
   silence ......... silence           feedback → feedback → feedback
   silence ......... silence      vs.  → feedback → feedback → ...
        ↓                                        ↓
   one big REVIEW  (surprise!)         a REVIEW that surprises no one

The common trap — saving it up for the review

The most common performance-management mistake is not being too harsh — it is being silent, then dumping months of stored-up feedback in one sitting. A manager notices a barista consistently mis-ringing orders in March, says nothing to avoid friction, and raises it in the September review. By then the habit is set, the person feels ambushed, and the manager looks like they were quietly keeping score. Every point you save for later loses most of its value and gains resentment. If it is worth saying in the review, it was worth saying the week it happened.

Clear expectations: people cannot meet a bar they were never shown

Before you can fairly judge anyone’s performance, they have to know what “good” looks like. This sounds obvious and is constantly violated. A huge share of what managers experience as “underperformance” is really just an expectation that was never made explicit — the manager had a clear picture in their head and assumed it was shared.

Clear expectations are the foundation of fair performance management for one simple reason: you cannot hold someone accountable to a standard you never showed them. If a warehouse picker was never told that orders over a certain value need a second check, then a missed check is not a performance failure — it is a communication failure, and it is yours.

Clear expectations cover three things, and vagueness in any one of them causes trouble:

• What — the actual outcome or standard. “Respond to patient call bells within two minutes,” not “be responsive.”
• How well — what good looks like versus just acceptable. Show an example if you can. People calibrate far better against a concrete sample than a description.
• By when — the timeframe, and how often you’ll check in on it.

In practice — the invisible standard

A team lead is frustrated that a developer keeps shipping code without tests. In the review, she raises it as a performance concern. The developer is genuinely bewildered: on his last team, a separate QA group wrote all the tests, and no one ever told him the expectation had changed. He wasn’t ignoring a standard — he never knew it existed. Ten minutes of explicit expectation-setting on day one would have prevented six months of quiet friction. The lesson generalizes far beyond software: most “why won’t they just…?” frustration is an unstated expectation wearing a disguise.

The ongoing loop: frequent, specific, timely feedback

If expectations are the foundation, feedback is the loop that keeps someone on the standard between the moment they learn it and the moment it becomes a habit. Feedback is not a special event you brace for. It is a steady stream, and it flows in both directions — toward what is working and toward what needs to change.

Three qualities make feedback actually work:

• Frequent. Small and often beats large and rare. A quick “that handover was really clear, thank you” today does more than a paragraph in a review. Frequency also lowers the stakes — when feedback is normal, no single instance feels like a verdict.
• Specific. “Good job” tells someone nothing they can repeat. “The way you summarized the three options before recommending one made the decision easy — do that again” tells them exactly what to keep doing. Specific feedback names the behavior and its effect.
• Timely. Give it close to the event, while the details are fresh for both of you. Feedback decays fast. A week later, “remember that call on Tuesday…?” is already a memory test.

Notice that positive feedback matters just as much as corrective feedback — arguably more, because it tells people which of their many behaviors to double down on. A practice that only ever speaks up to correct trains people to hear your voice as a threat. Catch people doing things right, and say so, specifically, often.

The evidence — feedback close to the moment

Research on skill development consistently finds that feedback works best when it is close in time to the performance it refers to, and when it is specific enough to act on. Vague, delayed praise or criticism generally produces little change in behavior; timely, concrete input produces more. Studies of high-performing teams also tend to find that a healthy ratio of positive to corrective feedback sustains engagement better than a stream of criticism alone. Treat these as directional, not precise, but the pattern is robust: little, often, specific, and soon beats big, rare, and late.

Diagnose before you respond: four different gaps

When someone is not meeting the bar, the instinct is to treat it as one thing — “a performance problem” — and reach for one response, usually pressure. That is a mistake, because underperformance has different causes, and each cause needs a different response. Applying pressure to the wrong cause makes things worse.

Before responding, diagnose which of these you are actually looking at:

  THE GAP                  THE QUESTION                    THE RESPONSE
  ───────                  ────────────                    ────────────
  Unclear-expectations     "Did they know the bar?"        Set/restate the expectation
  Skill                    "Can they do it yet?"           Train, coach, pair them up
  Motivation               "Do they want to?"              Reconnect to what drives them
  Genuine performance      "Clear bar, able, willing —     Address directly, with a path
                            still not meeting it?"          (and consequences if it persists)

• An unclear-expectations gap. They never knew the standard. The fix is not feedback about their performance — it is to set the expectation clearly, now. (See the section above.)
• A skill gap. They know the bar and want to hit it but can’t yet — they lack a specific ability. Pressure is cruel here; it just makes an anxious person fail faster. The response is coaching, training, or pairing them with someone who has the skill.
• A motivation gap. They know how and could do it, but something has drained the will — boredom, burnout, a sense the work doesn’t matter, a grievance. The response is a conversation about what actually drives them, not a lecture about output. Piling on pressure a demotivated person will often push them out the door.
• A genuine performance problem. The expectation was clear, the skill is there, the motivation is present — and they still aren’t meeting the bar. This is the case where direct performance action is the right tool. It is also rarer than managers assume, which is exactly why you diagnose first.

The discipline is to ask the questions in order and stop at the first “no.” Most apparent performance problems dissolve at the first or second question — which is good news, because those are far easier to fix than the last.

In practice — same symptom, three different fixes

A nurse manager has three staff nurses who are all slow to update patient charts. It looks like one problem. On closer look: one was never told charts must be updated before end of shift (unclear expectation — fixed in a two-minute conversation); one is new and still fumbling the software (skill gap — fixed by pairing her with a fast colleague for a week); the third is a strong nurse quietly furious about being passed over for a role, and has stopped caring (motivation gap — needs a real conversation, not chart training). One symptom, three causes, three completely different responses. Had the manager “cracked down on charting” across the board, she’d have insulted the first, terrified the second, and lost the third.

Address underperformance early, directly, and with a path

When you do land on a genuine performance problem, the failure mode is to let it drift — to hope it self-corrects, to soften the message until it disappears, to avoid the discomfort. Drift is the enemy. A small gap addressed early is a coaching conversation. The same gap ignored for six months becomes a crisis, a formal process, sometimes an exit — and along the way it quietly tells the rest of the team that the standard doesn’t really matter.

Addressing it well has three parts:

1. A clear picture of the gap. Describe the specific difference between the expectation and the reality, using concrete examples, without character judgment. “The standard is orders double-checked over £500; on these three occasions that didn’t happen” — not “you’re careless.” Name the behavior and its effect, not the person’s worth.
2. A path to close it. Underperformance addressed without a path is just criticism. The person should leave the conversation knowing exactly what “back on track” looks like, what support they’ll get, and by when. This is where you become their ally against the problem, not their adversary.
3. Honest stakes, if they apply. If continued underperformance has real consequences — a formal plan, a role change, an exit — the kind thing is to say so plainly and early, not to spring it later. Vagueness here is not kindness; it robs the person of the chance to respond while they still can.

Do this early, while the gap is small and the goodwill is intact, and most performance problems get solved in the conversation rather than in a process. That is the whole point of treating performance as a practice: you handle small things while they are still small.

The common trap — softening the message until it vanishes

Trying to be kind, managers often wrap the hard message in so much cushioning that the person walks out thinking everything is fine. “You’re doing great overall, just maybe keep an eye on the deadlines, no big deal really” — when deadlines are the actual serious problem. The person can’t fix what they didn’t hear. Real kindness is clarity: state the gap plainly, then be warm about the path. Be hard on the problem and soft on the person — not soft on the problem to protect the person’s feelings, which only guarantees a harder conversation later.

Case study — Netflix and radical candor, 2009 onward

When Netflix published its culture deck in 2009, one of its most-discussed principles was that feedback should be frequent, direct, and given close to the moment — the opposite of the annual-review model. The lesson people took from it wasn’t “be brutal”; the healthier reading, later sharpened in the idea of radical candor, is that caring about someone and being direct with them are not opposites — you can only afford to be direct because you’ve shown you care. Withholding hard truths to seem nice, on that view, is a quiet form of disrespect. Treat the specifics as one company’s approach, not a universal rulebook, but the core move travels everywhere: the trust you build in the earlier pages of this part is exactly what lets your direct feedback land as help rather than attack.

How it all connects

Performance management is not a separate skill bolted onto the end of this part — it is what happens when the rest of the part is working. Clear expectations depend on your ability to describe outcomes, from delegation. Feedback lands as help rather than threat only on a base of trust. Diagnosing a skill gap versus a motivation gap uses everything from coaching and what actually motivates people. And the whole thing points back at the part’s promise: reliable outcomes, no chaos, no burnout, and no one ever ambushed by a review. The revision page ties the six topics back into the single system they form.

Try this

Pick one person you manage and audit your performance practice with them this week. Answer three questions honestly: (1) Could they state, right now, what “good” looks like in their main responsibility? (2) When did you last give them specific feedback — positive or corrective — within a few days of the thing it referred to? (3) Is there anything you’d raise in a review that you haven’t raised in the moment? If the answer to (3) is yes, that is your homework: have that conversation this week, while it’s still small — and use the gap / diagnose / path structure to do it.

Reflect

1. Think of a review (yours or one you gave) that contained a surprise. What earlier, in-the-moment conversation should have happened instead — and why didn’t it?
2. For someone you manage, could they actually state what “good” looks like in their role? If not, is what you’ve been calling underperformance really an unclear-expectations gap?
3. Which do you give more of — positive or corrective feedback — and how specific and timely is either? What would change if you caught people doing things right, out loud, more often?
4. Recall a time you treated underperformance as one thing. Was it really a skill gap, a motivation gap, or unclear expectations in disguise? How would the right diagnosis have changed your response?
5. Is there a performance problem you’re currently letting drift because the conversation feels uncomfortable? What is the drift costing — for the person, the team, and the standard?

Show reflections

1. The useful insight is almost always that the in-the-moment conversation felt awkward, so it got postponed into the review — where it became far more awkward. Naming why you avoided it (fear of conflict, hoping it would resolve itself) tells you the exact habit to build: raise small things when they’re small.
2. If they can’t state the bar, you’ve likely mislabeled a communication failure as a performance failure. The fix is fast and freeing — set the expectation clearly and watch how much “underperformance” simply disappears.
3. Most managers under-give positive, specific feedback and over-rely on staying quiet until something’s wrong. Shifting toward frequent, specific praise doesn’t just feel nicer — it tells people which behaviors to repeat, which is how standards actually spread.
4. This is the highest-leverage reflection in the page. A skill gap met with pressure, or a motivation gap met with training, makes things worse. Getting the diagnosis right usually means the response is easier than the one you were about to reach for.
5. Drift always has a cost, even when it’s invisible: the person loses the chance to fix it early, the team learns the standard is optional, and you inherit a bigger conversation later. Naming the cost is usually what makes the uncomfortable conversation feel worth having now.