Performance Review Best Practices for Small Teams That Actually Work

Seven people. That's the whole team. You are the manager, you also do the work, and somewhere between the payroll spreadsheet and the client fire you're putting out at 6pm, you're supposed to run meaningful performance reviews.

Sound familiar? Good, because small teams have a performance review problem that no enterprise HR playbook solves. The frameworks you'll find online assume a dedicated people team, a calibration committee, and managers who manage full-time. You have none of that. What you have is proximity — you see these people every day, you know their kids' names, and that's precisely what makes an honest review so awkward.

This is a guide to performance review best practices for small teams, built around what actually survives contact with a 5-to-15-person company. No HR department required.

Why performance review best practices for small teams look nothing like the enterprise version

A 12-person company has no HR business partner, no calibration session, and no anonymity. Everyone knows what everyone earns. Everyone knows who left early on Friday.

That changes the mechanics completely. In a large company, the review process is designed to counteract the fact that your manager barely knows you. In a small one, the process has to counteract the opposite problem: your manager knows you too well and likes you too much.

The proximity trap

I ran reviews for a nine-person agency for a stretch of about two years. The first cycle, I gave everyone a "meets expectations" or above. Every single person. Then I looked at the actual project outcomes and realised two of those reviews were wrong — one person had coasted for a quarter and I'd genuinely not noticed, because they were pleasant, reliable in meetings, and easy to be around.

That's proximity bias in its purest form. It doesn't feel like favouritism while you're doing it. It feels like being a decent human being.

The fix isn't to become colder. It's to make the review depend on written evidence collected before the conversation, not on the impression sitting in the room with you. If your process runs on memory, it runs on vibes.

There is no one to check your work

In a bigger organisation, someone else reads your draft ratings and pushes back. In a small team, you are the reviewer, the approver, and the appeal court. Which means the quality of the review is capped by the quality of your own self-discipline.

Practical consequence: build the process so it's hard to skip. Put the review slot in the calendar a quarter ahead. Keep a running notes file per person that you update after every 1:1. If it takes more than four minutes to find an example of a behaviour from three months ago, your notes system is broken.

The cadence that actually holds: one hour a week, not one hour a year

Here's the thing about annual reviews in a small company: they're load-bearing in a way they shouldn't be. There's no compensation committee, no promotion panel, no formal grade. So the yearly meeting ends up carrying pay, promotion, development, and emotional validation all at once. Nobody performs well under that weight.

Split the function. Three layers work for teams under about fifteen people:

Layer	Frequency	Duration	What it covers	Who writes anything down
1:1 check-in	Weekly or fortnightly	25–30 min	Blockers, priorities, workload, a bit of how-are-you	Manager only, a few lines
Quarterly review	Every 3 months	40–45 min	Progress against 2–3 objectives, one strength, one development area	Both sides, one shared page
Annual conversation	Once	60–75 min	Compensation, role evolution, ambition for the year ahead	Manager, formal record

Notice what's missing: the quarterly review is where the assessment happens. By the time you reach the annual conversation, nothing in it should be a surprise. That's the whole point. If someone hears criticism for the first time in their annual review, the process failed eleven months earlier.

How long should a quarterly review take you?

Realistically, 40 minutes per person if you've kept 1:1 notes: 15 minutes to re-read them and draft two or three observations, 45 minutes in conversation, 5 minutes writing the outcome down. For a team of eight, that's around one working day per quarter, spread across a fortnight.

Compare that to the alternative. An annual process in a small team usually collapses into two days of panic in December, plus the emotional cost of a conversation you've been avoiding since March. The light version isn't less rigorous. It's just spread out enough to survive.

What a good review actually contains

Strip out the personality language and a useful performance review answers four questions, in this order:

1. What did this person actually deliver, and against what expectation?
2. Which specific behaviours helped — and which got in the way?
3. What changes next quarter, in concrete terms?
4. What do they want, and what are you doing about it?

Question four gets skipped constantly in small teams, and it's the one that drives retention. If your best designer has told you twice they want to lead a project and it's never happened, they will notice — and they will start quietly looking.

What are the 5 C's of performance management?

The five C's are a common framework: clarity, context, consistency, communication, and coaching. Clarity means expectations are explicit before work starts. Context means the person understands why their work matters to the business. Consistency means the same standard applies to everyone. Communication means feedback flows both ways and often. Coaching means the manager develops capability rather than simply judging output.

Does it hold up? Mostly, yes. I'd push back on one thing: in a team of six, consistency is the hardest of the five to actually achieve, because your sample is too small to average out. You can apply the same rubric to two people and still reach different outcomes depending on how visible their work was. Which is why the second evidence source matters so much.

Two evidence sources, always

Never rely on your own observation alone. Use:

• A short self-assessment from the employee, sent three working days ahead. Not a form with 20 questions — three prompts: what went well, what was hard, what do you want next quarter.
• Peer input on a rotating basis, asking one or two colleagues a single question about collaboration. In a small team, do it openly and consistently rather than as an anonymous 360, because anonymity in a team of seven is a polite fiction.

This is the part most small-team managers skip, and it's the cheapest bias correction available. Three sentences from a colleague will surface things you genuinely didn't see. And the worked example:

I once had a developer on a team of five who looked quiet and low-output to me — few meetings, not much visibility. Peer input revealed he'd been quietly unblocking two other people every week, spending something like 30% of his time on it. My rating was going to be "meets expectations." It should have been well above. Without a second source, I'd have written a review that ranged from inaccurate to actively demotivating.

What not to say during a performance review

Some phrases are small-team landmines, and they tend to come out precisely when you're uncomfortable:

• "Everyone feels that way about you." Vague, unaccountable, impossible to respond to. In a seven-person team it also means the person will spend the next week trying to guess who said what.
• "You're like family here." Sounds warm, quietly signals that standards are personal rather than professional.
• "You're doing great — no notes." Feels kind. Communicates that you didn't prepare.
• "We might need to think about your future here." Never say this unless you have a concrete, documented performance concern and a clear next step. In a small company, a sentence like that can trigger a resignation by lunchtime.
• Comparing to a colleague by name. "You're not as fast as Marc" tells the person nothing they can act on.

The pattern behind all five: no specific behaviour and no observable consequence. If you can't describe what happened and what effect it had, you're not giving feedback — you're transmitting a mood.

One more, less obvious: avoid opening with "How do you think you did?" and then using their answer to soften your own assessment. It's a common coaching trick that, in practice, trains people to lowball themselves so the conversation stays pleasant.

Handling the cases small teams get wrong

Underperformance in a tiny team

This is the hardest scenario, because there's nowhere to hide someone. A person who's struggling is visible to every colleague, every day, and the rest of the team is quietly watching whether you'll act.

What works: name the gap in the quarterly review, not the annual one, and describe it in terms of delivered work against agreement. Then agree on two specific changes and a check-in date four to six weeks out. Write it down and send it to them the same day. That record protects both of you.

What doesn't: waiting for the annual review to "give them a fair chance." Six months of silence followed by a poor rating isn't fairness, it's ambush.

The high performer who wants a title you can't offer

Small teams hit this constantly. Someone's ready for more scope, and the org chart has no rung above them. I've made this mistake: I stalled for about four months, saying we'd "look at it," and by the time I came back with a proposal the person had already been interviewing.

Honest options are scope, money, autonomy, or a named project. Say plainly which ones are on the table and which aren't. Vagueness is the thing that costs you the person.

What are the 5 words for performance review?

There's no fixed official set of five words — but the formulation I find most useful as a memory anchor is: clarity, evidence, dialogue, development, follow-through. Clarity of expectation beforehand. Evidence rather than impression. Dialogue rather than a verdict delivered from above. Development as the purpose, not the afterthought. And follow-through, because a review with no logged commitments evaporates within a fortnight.

If you remember only one of the five, make it the last one.

What are five best practices for high-performing teams?

Strip away the framework names and the research that keeps resurfacing points the same way: psychological safety, clear goals, regular feedback, recognition, and development opportunity.

For a small team, translate those into behaviours, because abstract principles don't survive a busy Tuesday:

1. Safety — a person can say "I don't know how to do this" in a team meeting without it becoming a performance issue
2. Goals — two or three written objectives per quarter, visible to everyone, not tucked in a document
3. Feedback — weekly, short, specific, and mostly positive by volume, because most behaviour is fine
4. Recognition — named publicly, quickly. Small teams are excellent at this when they bother
5. Development — something concrete each quarter, even if it's just a stretch task or a budget for a course

The recognition point is worth dwelling on. In a twelve-person company, praise from the founder or manager lands harder than any bonus scheme you could design, and it costs nothing. Yet it's the first thing to disappear when the quarter gets busy.

A working template you can set up in an afternoon

Keep the paperwork embarrassingly short. One shared page per person, four sections:

• Objective and how it's measured — two lines, written by the manager, agreed in the previous review
• What happened — a handful of dated bullet points from your 1:1 notes, both wins and misses
• Assessment — one paragraph, in plain language, no rating scale if you can avoid it (a five-point scale in a team of seven is just numerology)
• Commitments — two or three items, each with an owner and a date

Skip the rating scale if you can. In a small team, numbers create false precision and endless arguments about the difference between a 3 and a 4. A written paragraph forces you to be specific. And specificity is the entire value of the exercise.

One rule I'd defend to the end: the employee reads the assessment before the meeting, or at the start of it. Delivering a written verdict verbally, in real time, gives you the pleasure of watching their face and gives them nothing to work with.

The review is the smallest part of it

Here's what took me longest to accept: the quarterly review document has almost no impact on its own. What changes how a small team performs is the thirty-minute conversation you have every week, and whether the notes from it are honest enough to be useful three months later.

The review just makes that record visible. If the page says nothing surprising, you've done the work. If it does surprise the person reading it, no template in the world will rescue the situation.

So the real question isn't which framework you adopt. It's whether, in your team of eight, an honest sentence about someone's work can be said out loud in June rather than saved up for December. Most small companies can't. The ones that can tend to keep their people a lot longer.