Compensation planning is like "extra hard Sudoku" except once you've figured it all out... you still have to convince each square that it's the right solution

Always challenging, never a straightforward "satisfying" outcome.

What is compensation planning and why do you care?

Every organization has some fixed budget. Yes there's a little wiggle room, but there's also a breaking point.

Compensation is "buying time with salaries": when you are in charge of an organization you owe it to them, to everyone, to spend responsibly.

It is not sustainable to spend everything today and having nothing for tomorrow. Leadership is providing some measure of predictability.

You must devote at least a couple of hours (or days) to planning and budgeting your largest expense of your most precious resource.

So you have N people, and you have budget X. And if you mess it up really badly you'll either have 0 people or 0 dollars (actually you can go negative but that's very unpleasant).

Fundamentals of compensation planning

Classically you keep salaries (the majority of spending at most successful companies) separate from "everything else" (real estate, vendors, infrastructure). Start with last month's total payroll x 12. That's your baseline spend.

Ask your HR or Board for clarity on any existing or historical policies on how this has been handled previously (or should be handled going forward).

Usually you are not doing this in complete isolation

What your organization needs

Before you start planning, make sure you first learn what your organization needs for the coming year:

Ask your manager
Ask your CEO
Ask your CFO
Ask your Board

There will be conflicting, even contradictory answers. That's okay =]

Everyone's predicting the future, and that's hard.

Ideally there will be a theme, and finding overlaps gives you a direction/pattern.

Is this a growth year with hiring? Maybe it's a "hold tight" year with a slightly shrinking budget.

Even in a shrinking budget there may be projects or opportunities the organization is willing to invest in.

Understand what your company wants to pay for, and where your team members (skills, experience, etc.) may fit in that picture.

Identify which key people (in priority order) would most impact the company if they left. Have you done succession planning?

More importantly: which people have a track record of success and you believe represent the future of your organization?

What your people need

Gather your data about your people and their compensation history

When were they hired?
When were they promoted within your organization?
When have they received raises?
Has there been a trend to their compensation (plateau, spikes, exponentially up)?

Gather your data (hopefully from your regular one-on-ones) about what your people most prioritize.

People want different things (either stated or revealed preferences), so have private notes of their priority ranked list of these things...

cold hard cash
equity
bigger bonus when things go well
flexibility in working hours
flexibility in location/remote
more PTO
more autonomy and/or responsibilities in their work

Some Example Personas

Alice was hired 4 years ago and prefers cash above all else. She has had regular raises every year at 3%. She's still at the low end of the salary band for her level - and you really want to retain her steady troubleshooting skills and institutional knowledge.

Bob was hired 2 years ago and prefers equity and working remote. He's only had one raise during a promotion of 7%. His performance is a roller coaster, and even though he's recently had a promotion he said he thinks he's under-levelled.

Charlie was hired 3 years ago and has never had a raise. Charlie likes PTO. Charlie's well liked by everyone. He's also at the top of his salary band and privately co-workers wonder if he's pulling his weight.

Good Practices

Having distinct levels with responsibilities in a career ladder is a data driven approach to evaluating performance and promotions.

there are many career ladders published on GitHub - consider using even a most rudimentary version of just two categories: "new grad" and "senior"

You should strive to be fair and objective. Not just because it's the right thing to do - but because that's a reproducible framework that creates predictable outcomes.

Salary bands for a given level are a common practice and available through a variety of sources. (see Milkovich et al.'s Compensation)

If you don't have salary bands you may have trouble recognizing that someone is underpaid (or overpaid) compared to market benchmarks.

Remove your emotions from the process. There is a place for emotions, but not in the current calculus of budgets and planning. The feelings come when you are delivering these decisions to real people.

Framework = rules based, and these rules should be statistically relevant. Rules for one-offs will add a lot unnecessary complexity.

Use a "Parity" principle: keeping people in the same levels or cohorts in a very close (or identical values)

Build your framework

Start with:

What does the company need? (specific skills? leadership? an upcoming key project/feature?)
What are the factors you care most about? (tenure? retention? innovation?)

Create your custom spreadsheet with the header columns.

First pass strategy: ignore the framework for a moment and just write down what feels right for each person.

This gut-check baseline helps you see where your instincts diverge from your rules.

put the basic info in - level, current salary, proposed raise
- have a running sum at the bottom to see what the total is
then remodel and tweak variables

Use a rating system (i.e. performance on a scale of 1 to 5) that maps to percentages:

1: 0% raise
2: 1% raise
3: 3% raise
4: 5% raise
5: 7% raise

of course change those values to whatever you and your organization find best

Raises should be meaningful to the individual for "a very good performer"

Create distinct gaps between the "low performers" and "high performers"

Bonuses are for specific effort in a given time period or year, Salary/Compensation is paying someone for ongoing and future responsibilities

Review

Now that you have your draft...

Ask HR for data

industry comparables
(anonymized?) other departments/groups in your company
Historical trends?
Anomalies or unconscious bias?

Ask for feedback from trusted places:

review with frontline managers (they have to deliver/live with your decisions)
"what did I miss?"
"what did you think about ratings/framework?"
Are you at risk of losing high performers?
examine individual cases

Nuances

Do look for an opportunity to correct an unconscious/undesired trend - just because someone else made a mistake earlier doesn't let you off the hook

When you have to fix things, small corrections (sometimes over multiple years) are preferred over one large change that looks like a "shadow promotion"

Focus on professional performance, Avoid life/family decisions as input: you don't know every invidual person's situation.

If you skew things and then later learn someone else had "more need" would that be fair? Would you need to ask/pry into every employee's personal and private situations?

Raises as absolute numbers or percentages: most people can understand both so pick one and be consistent.

If you use percentages, at larger salaries this can become surprisingly large absolute-number raises

When someone is getting a raise or promotion, make sure the amount they're receiving "makes a difference".

Giving someone a 1% raise sometimes feels like the worst of both worlds: you are using up budget and the person may get the sense "we don't value you".

Either give a larger raise like 3% or explain "HR/company default for middle performers is 1%", here are the things you can do to be a high performer.

International and Company Patterns

If you have employees spread internationally it may be very difficult to compare without regard to the local economy. Many innovative organizations have attempted to have a "single global salary" and eventually capitulated to economic realities.

Your organization may deliberately choose to not be competitive to the market. You will have to indicate that losing people will affect capacity/timelines, and extra costs for recruiting and hiring a replacement.

Is there an implicit pattern sought or enforced by your company?

The expectation over a large enough distribution of a bell curve or "bottom 10% get 0?"

it's confronting these types of practices that remind you how important your role is in understanding what the organization says vs does - and how your team are impacted

Adjusting to fit your budget

When your compensation proposal doesn't fit your budget, you've got to tweak some things :|

People who are at the top of a salary band are less likely to feel the a small absolute reduction.

If you have excess (lucky you!), consider distributing that to the "bottom" earners first as that may have the most relative impact

Use round numbers: we're human! - This also means, if you need to round down, do so on the higher end first

What if I'm handed a finished compensation plan?

Do your homework. If you spot a serious error you can raise it with your manager and potentially prevent an embarrassing or even consequential mistake.

Showing interest and doing this exercise shows you are preparing to take on more responsibility.

When is it done?

Sleep on your decisions. Review them again with fresh eyes.

These numbers can affect people's life decisions: a progression in their career, where they live, if they get a new car, where their kids go to school or university, etc. They deserve your best when you're making such critical decisions.

If I got this raise from my manager how would I feel?

If I was a manager who had to deliver this compensation change would I know what to say and how explain it?

Can the organization deal with the consequences if someone is unhappy and leaves?

Another round of (hopefully small) tweaks and reviews should go much faster.

Reality bites: sometimes you do this perfectly and then the organization comes back and asks you to reduce your proposal by 10%.

No one ever said the work was easy, but someone has to do it, and do it well.

Delivering the changes

Once it is all approved and you have printed compensation change (or promotion!) letters, practice sharing the news.

Each person deserves an individual conversation, even if you feel like the change was routine or that it was "a large raise and that should make them happy".

(Research confirms that pay systems affect different people in different ways "Personnel Psychology" from Fulmer et. al.)

Do not start with assumptions, every person has unique circumstances and reacts differently
Do give them the letter with the number and at least a minute to read and process
Do listen and observe their reaction, and ask them if they have any questions or feedback
Do not offer any promises to change things, but do take notes and look into concrete concerns

sometimes people need a day or week to fully process, be open and prepared for them to bring those questions about a week later

Part of the reason you put in all the effort to a framework and process is to be prepared to persuade each part of the "Sudoku Puzzle" why the solution you've chosen works.

And sometimes, when you've done this work with care and delivered it with empathy, even if they're not happy with the result, they will understand that you worked hard on their behalf.

Compensation Planning is Sudoku with Feelings