Competency and Culture
Never Discuss the Raise and the Growth Plan in the Same Meeting
Once the salary number enters the room, the feedback stops landing. Why pay and development conversations belong in separate meetings.
About This Series
This is the fifth post in our seven-part Competency and Culture series. The first four posts were about getting the judgement right: fair ratings, a score for behaviour as well as results, values written as observable actions, and hiring for what people add. This post is about what happens next, in the room, when that judgement is shared. Even a perfectly calibrated rating can be wasted if it is delivered in the wrong conversation. Get this right, and your feedback starts to change behaviour instead of just justifying a number.
Every manager has seen it. Twenty minutes of careful, thoughtful feedback: what went well, what to work on, where the person could be in a year. Then the salary number. And the only thing the employee remembers, walking out of the room, is the number.
It is not that people are shallow. It is that pay is personal, and it carries a verdict. A number that is higher or lower than expected answers the question every employee is really asking in a review: "What does this company think of me?" Once that question is answered, there is no attention left for anything else.
Most review processes still try to do both jobs in one meeting. They look at the past, decide the pay, and plan the future, all in the same hour. It feels efficient. It is why so much feedback never lands.
What happens to feedback once pay enters the room?
It gets heard as a justification for the number, not as advice for the future.
This is one of the best-supported findings in workplace research. In 1996, psychologists Avraham Kluger and Angelo DeNisi reviewed more than 600 measured effects of feedback on performance. Most people assume feedback helps. They found that in more than a third of cases, feedback made performance worse.
The pattern behind it was consistent. Feedback helped when it kept attention on the task: what to do differently next time. It hurt when it pulled attention towards the self: status, self-image, how the person was being judged. Pay does exactly that. It turns a conversation about work into a conversation about worth.
You can see it in how people behave in the room. When the number is good, they relax and stop listening, because the verdict is in. When the number is disappointing, they get defensive, and every piece of development feedback sounds like a reason they were paid less. Either way, the growth plan you prepared does not get the attention it deserves.
Managers feel it too. When a manager knows the pay decision is coming at the end of the meeting, they soften the feedback to protect it. Or they sharpen it to justify a low number. Neither version is the honest, useful feedback the employee needed. Feedback only lands when it is safe to hear, which is the point we made in The Best Teams Are Not Fearless.
What does recent data say about how reviews are working?
Not well. In 2023, Gallup surveyed 135 chief people officers at Fortune 500 companies and 18,665 employees. Only 2% of the people officers strongly agreed that their performance system inspires employees to improve.
The employee data shows why. 56% of employees said they formally review their goals with their manager once a year or less. When companies held quarterly progress conversations instead, employees were 90% more likely to be engaged, and 2.1 times as likely to feel the process was fair and transparent.
Gallup's own recommendation from that research is the argument of this post: leave pay and promotion out of performance reviews, and discuss them in separate conversations, so that managers can focus the review on feedback and development.
| One combined review | Two separate conversations |
|---|---|
| Once a year, backward-looking | Development often (at least quarterly), pay once or twice a year |
| The number dominates the meeting | Development gets the whole meeting |
| Managers soften or sharpen feedback to fit the pay decision | Feedback can be honest, because nothing is being defended |
| Employees listen for the verdict | Employees listen for what to do next |
What this looks like in practice
look at your current review template. If the rating, the pay increase, and the development plan are on the same page, discussed in the same meeting, you have designed a conversation where the development plan loses.
How do companies that do this well split the two conversations?
The best-known example is Google. In Work Rules!, Laszlo Bock, Google's former head of people operations, described keeping performance conversations and pay conversations apart, with pay discussed about a month after performance feedback. The reasoning was simple: people cannot absorb developmental feedback while they are waiting to hear their number.
You do not need to be a large company to do this. We saw it work at a high-growth company one of us worked in.
From our experience
The appraisal and the performance cycle were two separate exercises, run at different times. Performance still decided the outcome: your rating fed directly into your increase and your promotion case. But the conversations were kept apart on purpose.
The performance conversation was about what went well, what did not, and what to build next. The appraisal conversation was about the number and the reasoning behind it. Because managers were not trying to deliver both at once, employees came into development conversations ready to listen, and managers stopped softening feedback to protect a pay decision already made.
The important detail is the first sentence of the second paragraph. Separating the conversations did not separate pay from performance. It separated the discussion of pay from the discussion of growth. Those are very different things.
If pay is decided separately, how is it still tied to performance?
This is the question founders ask most, and it is a fair one. If you stop talking about pay in the review, does performance stop mattering for pay? No. What changes is the order and the forum, not the link.
A simple sequence works for most growing companies:
| Step | What happens | Who is in the room |
|---|---|---|
| 1. Development conversations | Regular, forward-looking check-ins on goals, behaviour, and growth. No numbers | Manager and employee |
| 2. Rating and calibration | Ratings are proposed with evidence and calibrated across teams | Managers and leadership |
| 3. Pay and promotion decisions | Calibrated ratings feed pay, equity, and promotion, alongside market data and budget | Leadership |
| 4. Pay conversation | A short, separate meeting: the decision and the reasoning behind it | Manager and employee |
The link between performance and pay is still there, and it is stronger, because the rating going into the decision has been calibrated. What has gone is the moment where the employee has to process a verdict and a development plan at the same time.
Two cautions. First, the pay conversation still needs an explanation. "Here is your increase" with no reasoning feels arbitrary, and arbitrary feels unfair. Second, the development conversations have to be real. If they only happen once a year, you have not separated the conversations, you have just moved the problem.
What this looks like in practice
hold the development conversation first, with no numbers in it. Share pay and promotion outcomes two to four weeks later, in a short, separate meeting that explains the decision.
What should the development conversation actually sound like?
Once pay is out of the room, the development conversation has space to do its job. Two simple frameworks help managers use that space well. The first looks back. The second looks forward. (If you have read our earlier post on performance reviews, these sit well alongside the 4As feedback model we described there.)
Looking back: Situation, Behaviour, Impact. Developed by the Center for Creative Leadership, this keeps feedback anchored in facts rather than judgements about character.
| Step | What the manager does | Example |
|---|---|---|
| Situation | Names a specific moment | "In Tuesday's planning review with the product team..." |
| Behaviour | Describes only what was seen or heard | "...you presented the delivery date without mentioning the dependency on the data team." |
| Impact | Explains the effect | "The product team committed to a customer date we now cannot hit, and we had to walk it back." |
Looking forward: GROW. Created by Sir John Whitmore, this turns feedback into a plan the employee owns.
| Step | Purpose | Example question |
|---|---|---|
| Goal | Agree what good looks like | "What would it look like if risks like this were always raised before we commit?" |
| Reality | Understand what is happening now | "What made it hard to raise the dependency in that meeting?" |
| Options | Let the employee generate the ideas | "What could you do differently in the next planning review?" |
| Will (or Way forward) | Agree a specific commitment | "Which of those will you try, and when should we check in on how it went?" |
Used together, the two frameworks change the feel of the conversation. It stops being an audit of the past year and becomes a working session on the next quarter. And because there is no number waiting at the end of the meeting, the employee can engage with it honestly.
What this looks like in practice
ask managers to bring one SBI example and one GROW question to every development conversation. It takes ten minutes of preparation, and it keeps the meeting about behaviour, not about the person's worth.
How does this connect to the rest of the series?
Separating the conversations only works if the rating behind the pay decision can be trusted.
It connects back to our first post. When pay decisions move away from the review meeting, they depend even more on ratings that have been checked across raters and teams. A strict manager's uncalibrated score costs someone money, not just a line of feedback.
It connects to our second and third posts. A development conversation needs something concrete to develop: behaviour described clearly, at the right level, with results and behaviour scored separately.
And it connects forward to calibration. In our next post, we look at the step between the development conversation and the pay decision: how to calibrate ratings honestly, why adding more raters does not solve the problem, and how to stop calibration from turning into forced ranking.
The reframe underneath all of it is simple. A pay conversation answers "what am I worth to you?" A development conversation answers "how do I get better?" Both questions matter. Asked in the same hour, only the first one gets heard.
Frequently asked questions
- Why separate the pay conversation from the development conversation?
- Because once money enters the room, development stops being heard. When a rating decides a raise, the employee defends a number instead of reflecting on how to grow, so the two goals undercut each other in a single meeting.
- If pay is decided separately, how is it still tied to performance?
- Performance still drives pay; the two conversations simply happen at different times. The company assesses performance and sets compensation on one track, and holds the forward-looking growth conversation on another, so each gets a clear hearing.
- What does the data say about how reviews are working?
- Not well. Gallup found only 2% of chief people officers strongly agreed their system inspires improvement, but when companies replaced the annual review with quarterly progress conversations, employees were 90% more likely to be engaged and 2.1 times as likely to find the process fair and transparent.
