This module is about translating the review conversation into actual decisions – about compensation, development, and what happens next for the people involved.

We'll start with the topic that tends to divide managers most: should you connect performance reviews to the pay conversation?

When to link reviews to compensation – and when to keep them separate

The direction most organisations are moving in is away from tying reviews directly to compensation decisions.

Frederick Herzberg's motivation research showed that pay works mainly as a hygiene factor. Its absence demotivates, but a pay rise doesn't build lasting engagement. When the conversation centres on money, the development discussion tends to vanish.

There's also a tension problem. When an employee knows this conversation directly affects whether they get a raise, and how much, the stakes become too high for honesty. Nobody admits that anything went wrong. Without that, there's no real reflection – and without reflection, nothing actually changes.

Research in the Harvard Business Review ("The Performance Management Revolution," Cappelli and Tavis) points the same way: more companies are moving away from reviews that focus on compensation, shifting toward more frequent development conversations.

And there's a credibility issue worth naming directly. If you officially say reviews don't affect pay while everyone in the room knows they do – don't bother. People will figure it out faster than you expect, and you'll have done more damage than if you'd said nothing at all.

That said, linking reviews to compensation isn't always the wrong call. It works in areas where results are clearly visible and measurable – sales and production being the obvious ones. When there's a sales target with a clear revenue number, or a production deadline with a concrete output, there's little room for interpretation. The employee can see exactly what they did, what effect it had, and what follows from that.

The trap is rewarding the wrong thing.

A salesperson hits their target – even exceeds it – by overpromising what the company can actually deliver. In the short term, the numbers look good. Long-term, however, team frustration grows, collaboration suffers, and client problems accumulate. The same goes for production: reward only speed, and people start cutting corners. Quality drops, and errors increase. The spreadsheet looks fine, while in practice, you're gradually losing control of your standards.

The problem isn't linking reviews to compensation. The problem is an incomplete system.

From working with small and mid-sized companies, I've found the hybrid approach tends to work best: the review is one input among several. Pay decisions are made against a broader picture – team results, current salary relative to the role's range, whether the person is taking on new responsibilities, and the company's financial position. That gives you more fairness, more predictability, and better decisions overall.

But this only works if the criteria are communicated clearly. When the rules aren't transparent, trust erodes, and the perception of unfairness spreads quickly.

How to document review outcomes without creating bureaucracy

Every review should leave a record. That doesn't mean elaborate forms, multi-page summaries, or a new system to manage.

In smaller organisations, heavy documentation doesn't help – it slows things down, leads to half-completed records, and makes it impossible to find the important details when you actually need them.

The rule is simple: documentation should support action, not be an end in itself.

What actually needs to be written down? Three things: a summary of the assessment, the key conclusions, and the specific next steps. That's it. You don't need to transcribe the conversation. What matters is what comes out of it.

The simplest model that works in practice is this: during the meeting, the manager notes three to five key takeaways, two to three concrete actions, and deadlines or check-in points. An email, a shared doc, a simple tool – the format doesn't matter. What matters is whether both people can come back to it quickly and see exactly what was agreed.

The most common mistake: writing things that can't be checked. "Improve communication." "Better time management." "More engagement."

These are useless as documented outcomes. There's no way to come back in three months and know whether anything actually changed.

Every agreed action should answer three questions: what exactly is going to happen, by when, and how will both sides know it's working?

• Instead of "improve communication": "For the next four weeks, you will run a weekly team status meeting and paraphrase the main points from each participant."
• Instead of "better project management": "For every project, you will send the client a written summary at the end of each phase."

Who writes it up? Either the manager, or the manager and employee together. Both sides need to agree on the content and understand the conclusions in the same way. Sending a short summary after the meeting costs five minutes and saves a lot of misunderstanding later.

Without it, three months down the line everyone remembers something different. The employee says, "We never discussed that." The manager says, "It was obvious." Nobody wins, and both people are less willing to engage with the next round.

How often to come back to the conclusions? Documentation only earns its place if you actually return to it. Regular 1:1s are the right mechanism – not to monitor the employee, but to track progress and catch anything that needs adjusting.

A simple test: if after three months you can't remember what was agreed upon, don't know whether anything changed, and have nothing to refer back to, then the documentation isn't doing its job.

Low performance: how to run the conversation and plan what comes next

This is the moment many managers dread. Saying "things are more or less fine" and moving on is much easier than having a difficult conversation.

But avoiding it doesn't make the problem go away. It stays – and it becomes visible to the rest of the team fairly quickly. A low performance result should be an impulse to act, not a conversation to sidestep.

Start with something specific. Opening with "there's room for improvement" or "there are some areas to work on" usually means the employee leaves without knowing what's actually wrong. Which means nothing changes.

Name the issue directly instead: "The team hasn't received a complete project status on time once this quarter." That's concrete. Only when something is concrete is there space for real reflection and the possibility of change.

Provide context. Telling someone what isn't working isn't enough on its own. The employee needs to understand why it matters – what it means for the team, and what it means for the company.

"Delays mean the client is losing confidence and doesn't want to renew." We've come back to Kim Scott's Radical Candor approach several times in this Academy, and this is one of the moments where it matters most. A well-run difficult conversation isn't an attack. It won't damage the relationship – if anything, it shows the employee that you take their work seriously. That's when the conversation starts going somewhere useful.

Agree on what actually happens next. This is the most commonly skipped step. A conversation that ends with "this needs to improve" isn't a plan. If you want something to change, go one level further. Agree on what specifically needs to change, by when, and how you'll both know it's getting better.

Ask about support, too. We tend to assume the employee should already know what to do, but often they don't. Asking "How can I help?" isn't softening the message – it's creating the conditions for change to actually happen.

The most common mistake managers make here is inaction driven by fear of how the employee will react: avoiding the topic, or softening the message until the real point disappears. The result is that the employee doesn't grasp the scale of the situation, doesn't see a reason to change, and the problem persists into the next period.

To close

The review itself changes nothing. What changes things is what you do after – the decisions you make, how you communicate them, and whether you translate them into specific actions.

That goes for all of it: compensation, development, documentation, and the difficult conversations that are easiest to avoid. Keep it simple, keep the rules clear, and follow through. That's when the process starts working.