Compensation Philosophy: What It Is and How to Write One
What a compensation philosophy is and how to write one. The six decisions, lead vs meet vs lag, and a copy-paste template a small business can use today.
Compensation Philosophy
The one-page document that decides every pay conversation you will ever have
An employee once asked me, reasonably, why a colleague who had joined after her was earning more for the same job. I had an answer. The answer was that he had negotiated harder.
I did not say that out loud, because saying it out loud makes it obvious what it is: not a compensation decision, but the absence of one. What I actually had was a pay structure determined by how comfortable each person felt asking for money, which is a structure that systematically rewards confidence rather than contribution, and which produces exactly the disparities you would expect it to.
A compensation philosophy is the document that would have prevented that conversation. It is one page, it takes an afternoon, and almost no company with fewer than fifty people has one. This covers what it is, how it differs from a strategy and a policy, the six decisions it has to contain, what a small business can realistically commit to, and a template you can fill in today.
What Is a Compensation Philosophy?
A compensation philosophy is a formal statement of the principles a company uses to decide how it pays people. It is the reasoning, written down, before you need it.
The most useful way to think about it: the philosophy is the answer you give when someone asks why. Why does this role pay what it pays. Why does she earn more than him. Why did my raise come in at that number. If you have no philosophy, you have no answer, and the absence of an answer is itself an answer that employees will interpret unfavorably.
Philosophy vs Strategy vs Policy vs Plan
These four words get used interchangeably and they are four different layers. Here is the distinction, which is genuinely useful once you see it.
| Layer | It Answers | Example |
|---|---|---|
| Philosophy | Why. The principles and values behind how you pay. | We pay at the market median for all roles, we do not negotiate individually, and everyone in the same role at the same level is paid the same. |
| Strategy | What. The decisions that implement the principles. | We benchmark annually against national wage data for our industry, we use three levels per role, and variable pay is capped at 15 percent of base outside sales. |
| Policy | How. The operational rules and procedures. | Offers above the band midpoint require founder approval. Pay reviews happen every March. Exceptions are documented in writing with a stated reason. |
| Plan | The mechanics. The actual numbers and structures. | The salary bands themselves. The bonus formula. The equity vesting schedule. The merit matrix. |
Small companies typically have a plan and nothing else: some numbers, no reasoning. That works right up until someone asks a question the numbers cannot answer, which happens sooner than founders expect and usually at the worst moment.
From Philosophy to Compensation Strategy
A compensation strategy is what the philosophy becomes once you attach numbers and dates to it. The philosophy says you pay around the median and do not negotiate individually. The strategy names the data you benchmark against, how many levels a role has, how wide the bands are, and what share of payroll next year's increases get.
Strategic compensation planning is mostly a calendar problem at this size. Pick the month you review pay. Put the benchmark refresh one month before it, and the budget conversation with whoever runs your finances one month before that. Three appointments a year keep the philosophy connected to the money that has to fund it.
The compensation model sits in between the two: mostly base salary, base plus a bonus tied to company results, or a commission-weighted structure for sales roles. Name the model in the philosophy and build its mechanics in the compensation plan, where levels, bands, and the review cycle get their detail.
Why It Matters More Than You Think
The intuitive belief is that employees care about how much they are paid. The research suggests they care more about whether the process was fair, and the gap is not small.
Sit with that second finding for a moment, because it is the more actionable one. Most employees who think they are underpaid are wrong. They are not underpaid. They are uninformed, and in the absence of information they have assumed the worst. That is not a pay problem you can solve with money. It is a communication problem, and a philosophy is the cheapest available fix.
There is a second, quieter reason. Research from WorldatWork has found that while the large majority of organizations report having a compensation philosophy, only about a third believe most employees understand it. A philosophy nobody has read is functionally the same as no philosophy at all. Writing it is half the job. Saying it out loud is the other half.
Lead, Meet, or Lag the Market
This is the central decision, and everything else in the philosophy is downstream of it. Four options, and one of them is what most companies actually do.
Notice that all four are defensible. There is no correct answer, and a company that lags the market honestly, with a clearly stated tradeoff, is on much firmer ground than one that claims to meet the market and does not. The failure mode is not choosing badly. It is not choosing at all, and then paying whatever each individual candidate managed to extract.
The Six Decisions You Have to Make
A philosophy is really six decisions in a trench coat. Make all six and the document writes itself.
The one people skip is governance, and it is the one that undoes everything else. If you do not write down who decides pay and how exceptions are made, then exceptions are made by whoever asks, and after enough exceptions you no longer have a philosophy. You have a philosophy-shaped document and a pay structure that contradicts it.
Do You Need a Compensation Committee?
Not in the formal sense, and probably not for years. A compensation committee is a subcommittee of the board that approves executive pay and oversees the structure underneath it, and at listed companies every member generally has to be an independent director under the SEC listing standards for compensation committees. None of that machinery is built for a company of twelve people.
The function is worth borrowing even when the committee is not. If one person sets pay, approves the exceptions, and reviews the results, nothing checks the philosophy except the memory of the person who wrote it. A founder and whoever runs the finances, meeting twice a year with every off-pattern pay decision in front of them, do the same job.
The formal version starts to matter when outside investment brings a real board, or when you hire executives whose packages include equity. At that point write a short charter: who sits on it, what it approves, and what it only reviews. Until then the governance line in your philosophy carries the same weight in one sentence.
How to Benchmark Without Benchmark Data
"We pay at the median" is an empty sentence until you can finish it: the median of what, measured where, for which version of this job. This is the step small companies skip, and skipping it quietly turns the philosophy back into the thing it was written to replace, which is a number somebody felt was about right.
You do not need a subscription to do this honestly. You need three sources that disagree with each other, because the disagreement is where the information is.
| Source | What It Gives You | Where It Misleads |
|---|---|---|
| BLS Occupational Employment and Wage Statistics (OEWS) | Free annual wage data by occupation, published for the nation, each state, and most metro areas, reporting the 10th, 25th, 50th, 75th and 90th percentile alongside the mean. | It describes an occupation, not your job. Every experience level sits in one distribution, so the 75th percentile usually describes a senior person rather than a well-paid junior one. It is also collected and published on a lag, so it tells you about last year's market. |
| Job postings that carry ranges by law | Live employer intent. In states that require pay ranges in postings, you can read what real companies are willing to commit to in writing, today, for a named title in a named place. | Posted ranges are wide on purpose and frequently span more than one level. Treat the bottom of a posted range as the number an employer would pay a candidate it had doubts about, not as the market floor. |
| Free published salary surveys for your industry or function | Role-specific detail that federal occupation codes cannot capture, often broken out by company size or stage, which is the cut that matters most to a small employer. | Self-selected respondents. Surveys run by recruiters, vendors or communities skew toward the population that reads them, which is generally the better paid and better networked end of the field. |
| Candidates and peers | What people are actually declining offers over, which is the only genuinely current signal you have access to. | A sample of two or three, weighted toward whoever spoke most confidently. Useful as a sanity check on the other sources and close to useless on its own. |
Two mechanics matter more than the sources themselves. The first is that percentiles in occupational data mix levels together. If you want to lead the market for a mid-level role, you cannot simply take the 75th percentile of the occupation, because a large share of the people above that line are people with ten more years of experience than the person you are hiring. Occupational percentiles tell you the shape of the whole distribution; your positioning target applies within a level, not across all of them.
The second is that you have to fix your geographic rule before you look anything up, not after. National data and metro data for the same occupation can differ by a third or more, and if you pull whichever number supports the offer you were already planning to make, you have benchmarked nothing. Decide the rule in the next section first, then choose the geography of your data to match it.
The arithmetic that follows is unglamorous, and that is the point. Suppose you are hiring an office manager and your three usable sources come back at roughly $58,000, $64,000 and $71,000. You are not looking for the average of those. You are looking for what explains the spread: the high figure is a metro posting for a role that also runs payroll, the low one is a national occupational median that includes part-year and part-time workers, and the middle one is a state-level figure for your actual metro. Once you can explain the spread, one of those numbers is obviously closer to your job than the others, and that number is your midpoint. From the midpoint the band builds itself: a common construction puts the minimum at about 85 percent of midpoint and the maximum at about 115 percent, which turns a $64,000 midpoint into a band of roughly $54,400 to $73,600.
Compa-ratio is also what makes the internal equity clause checkable rather than aspirational. "Same role, same level, same range" is a promise you cannot audit by staring at a list of salaries, because the salaries are supposed to differ somewhat. You audit it by listing compa-ratios in a single column and looking for the ones you cannot explain. If the two people doing identical work sit at 0.92 and 1.11, the philosophy has already failed and you now have the number that proves it.
The Compensation Metrics Worth Watching
Five numbers tell you whether the philosophy still describes what you actually do, and every one of them comes out of a spreadsheet you already keep. None of them need software, and none of them take longer than an afternoon once a year.
| Metric | How You Work It Out | What It Tells You |
|---|---|---|
| Compa-ratio | Salary divided by the midpoint of the band, per person | Whether individuals sit where the philosophy says they should. Materially above or below 1.0 needs a reason you can state out loud. |
| Range penetration | Salary minus band minimum, divided by band maximum minus band minimum | Where somebody sits inside their band as a percentage. Easier to read than compa-ratio when the bands are wide. |
| Share of employees outside their band | People above the maximum or below the minimum, divided by headcount | How stale the bands are. A rising number almost always means the market moved and the bands did not. |
| Spread within a level | Highest salary minus lowest, for the same role at the same level | Whether the internal equity clause is real. This is the number employees work out for themselves if you never publish it. |
| Increase budget as a share of payroll | Total planned increases divided by current annual payroll | What the philosophy costs next year, settled before individual conversations start rather than after them. |
Read them together rather than one at a time, in the month before your review cycle. A set of compa-ratios that looks orderly on its own can sit beside a spread within one level that does not, and the second number is the one an employee notices first. The formal version of that fourth row is a pay equity analysis.
Remote Pay: Pick the Rule Before You Need It
The geographic decision feels abstract right up until the first time you hire someone in a much cheaper city, or an existing employee tells you they are moving to a much more expensive one. Both events force an answer, and an answer produced under that pressure is a rule shaped entirely by one person's circumstances, which is how companies end up with a policy they cannot apply to the next person.
| Model | How It Works | Where It Breaks |
|---|---|---|
| Single national rate | One rate per role and level for the entire country, regardless of where the employee lives. Location simply is not an input. | It is the most expensive model, because you are effectively paying everyone at the rate of your most costly market. It also loses candidates at the very top of the market, since a national rate set to be affordable sits below local pay in the highest-cost metros. |
| Location tiers | Two to four zones, each carrying a stated percentage differential from a base zone. An employee's tier follows their home address. | The boundaries. Someone living forty minutes outside a tier-one metro who does everything a tier-one employee does will not accept the tier-two number, and you are left defending a line drawn on a map rather than a principle. |
| Local market indexing | Each employee's pay is set against the market for their own metro, so the phrase 'we pay market' means literally what it says. | The highest administrative load, and it embeds a fairness question you have to be willing to answer aloud: two people doing identical work at identical quality are paid differently for reasons that have nothing to do with the work. |
Whichever model you pick, the clause the philosophy actually needs is the one about movement, because that is the case people ask about. Three things have to be stated: whether pay changes when an employee relocates, whether it can move down as well as up, and when the change takes effect. The version most small companies can live with is that pay is set by location at hire, that a move to a higher-paying market is reviewed at the next scheduled compensation cycle rather than immediately, and that a move to a lower-paying market does not trigger a cut. That is not the most rigorous answer, but it is one you can apply consistently, which beats a rigorous rule you quietly waive for the first person who tests it.
What the Law Already Decides for You
A compensation philosophy is a voluntary document with involuntary consequences. Several bodies of law reach into it, and they constrain the wording more than most founders expect.
The Equal Pay Act. Federal law requires equal pay for substantially equal work within the same establishment, and a pay difference between employees of different sexes doing that work is lawful only where it flows from a seniority system, a merit system, a system measuring earnings by quantity or quality of production, or a factor other than sex. Those are affirmative defenses, which is the part that matters: the burden of proving one sits with the employer. "He negotiated harder" is not a system. It is the absence of a system, which is exactly what the defense requires you to produce. The Ninth Circuit held in Rizo v. Yovino that prior salary alone cannot serve as a factor other than sex, and several state equal pay laws go further still, requiring any such factor to be job-related and consistent with business necessity. Read in that light, the internal equity clause in your philosophy is not only a values statement. It is the documentation of the system you would otherwise have to invent after the fact.
Salary history bans. A large and growing number of states and cities, including California, New York, Massachusetts, Illinois, Colorado and Washington, prohibit private employers from asking applicants what they currently earn, and some also require you to hand over the range for the role on request. The variation is in the details rather than the direction of travel, so a company hiring across state lines is generally better off adopting a single rule that it never asks, and stating that rule in the philosophy where recruiters and hiring managers will actually see it.
Pay range disclosure in postings. Colorado moved first and applies its requirement to every employer with at least one employee in the state. California and Washington reach employers with 15 or more employees, and New York reaches employers with four or more. Hawaii, Illinois, Minnesota, Maryland, New Jersey, Vermont and Massachusetts have their own versions, with their own thresholds and effective dates. Several of these statutes reach roles that could be performed in the state, which means a single fully remote posting can trigger more than one at once. This area has changed almost every year since 2021, so verify the current text rather than relying on any list, including this one.
The right to talk about pay. Under Section 7 of the National Labor Relations Act, non-supervisory employees have a protected right to discuss their wages with each other, in almost every private workplace, unionized or not. A handbook rule forbidding it is unlawful whether or not you ever enforce it. The practical consequence for your philosophy is worth being blunt about: employees are going to compare numbers. The only decision you actually control is whether the explanation they hear comes from you in advance or from each other afterwards.
What a 10-Person Company Can Actually Commit To
Nearly every guide on this topic is written for a company with a compensation team, an equity program, and a benchmarking subscription. If you have ten employees and none of those things, here is the honest version.
| Decision | What Enterprise Guides Suggest | What You Can Actually Do |
|---|---|---|
| Market positioning | Percentile targets by job family, refreshed against multiple benchmark datasets. | Pick one target for the whole company. 'We aim to pay around the market median' is a complete and honest philosophy. |
| Benchmarking | A paid benchmarking subscription with peer-group filtering. | Free federal wage data, job board ranges, and a few honest conversations. It is not perfect and it is vastly better than guessing, which is the current method. |
| Pay mix | Modeled base, bonus, equity, and long-term incentive by level. | Mostly base salary, and say so. Simplicity is a feature. A small company does not need a variable pay structure to have a philosophy. |
| Internal equity | Formal job architecture with levels and bands. | A commitment: people doing the same job at the same level are paid the same. You do not need job architecture to keep that promise. You need to keep it. |
| Transparency | Published ranges in every posting, internal ranges visible to all. | At minimum: every employee can read the philosophy and understand how their pay was set. That is already more than most small companies offer. |
| Governance | A compensation committee with documented delegation of authority. | One named person decides, reviews happen once a year in a stated month, and exceptions are written down with a reason. That is governance. |
A Template You Can Fill In Today
The philosophy itself is six sentences and you have just read what goes in each of them, so write it in your own words rather than mine. The document worth having a form for is the one that records where you departed from it. A philosophy is almost never repealed. It dies by accumulating exceptions that nobody wrote down, until the file on the shared drive describes a company you no longer are. One entry per departure, and one look back over all of them once a year.
The line most people cannot fill in is the one about precedent, and that is the line worth sitting with. An exception you would repeat for the next person is not an exception at all. It is the rule you follow, and the document has quietly stopped describing you. Finding that out over a year of entries is considerably cheaper than finding it out in a compensation conversation with an employee who is already upset.
What Real Philosophies Look Like
Three real approaches, deliberately chosen because they are very different from each other. All three are defensible. None of them is the right answer for you.
| Company | The Approach | What It Costs Them |
|---|---|---|
| Buffer | Radical transparency. Buffer publishes its salary formula and every employee's compensation publicly, and has done for years. | Total loss of negotiating flexibility, and every pay decision permanently visible. In exchange they get enormous trust and a recruiting signal nobody can fake. |
| Basecamp | No negotiation. Everyone in the same role at the same level is paid identically, benchmarked to top-of-market rates, with no location adjustment. | They pay significantly more than they would have to, especially outside expensive cities. In exchange they eliminate negotiation-driven inequity entirely. |
| The typical small business | Unstated. Each offer negotiated individually against whatever the candidate asked for. | It appears to cost nothing, which is why it is so common. It actually costs a pay structure nobody can explain and an equity problem accruing quietly in the background. |
The third row is the one to sit with. It is the default, it is what most companies with fewer than fifty people are doing right now, and it feels free precisely because its costs arrive late and arrive disguised as something else. You can inspect the first approach yourself: Buffer's public salary data is still online.
Three Compensation Philosophy Examples, Written Out
Those three rows describe what companies do. Here is what the document itself sounds like at the size you are probably working at. None of the three below is a real business, and each is deliberately short enough that an owner could say it out loud without reaching for the file.
A twelve-person accounting firm, meeting the market: "We pay around the median for our metro, refreshed against federal wage data each January. Pay is mostly base salary. Two people in the same role at the same level sit in the same range, whatever either of them asked for at hire. I approve every offer, and pay is reviewed in March."
A six-person software company that cannot match larger salaries, lagging honestly: "We pay below the market median on cash and we say so in the first conversation. What we offer instead is equity, four-day weeks, and scope you would wait years for elsewhere. Ranges are visible internally. We never ask what you earned before."
A twenty-person field services business, hybrid: "We lead the market for licensed technicians, because an open van costs us jobs, and we meet it for everyone else. That difference is deliberate and we will say it out loud. Technicians carry a monthly target bonus. Remote staff sit on one national band, wherever they live."
What makes all three usable is that each contains at least one sentence somebody could hold you to. "We pay competitively" contains none, which is why it is the most common opening line in this genre and the least useful sentence in it.
Writing Yours in Six Steps
Ways This Goes Wrong
| Mistake | What Happens | The Fix |
|---|---|---|
| Never writing one | Pay is set by negotiation. Your structure systematically rewards confidence over contribution, and you cannot explain any individual number without embarrassment. | Write one page before your next hire. The next offer is another precedent, and precedents are what you are accumulating whether you intend to or not. |
| Writing one nobody reads | You have a document and the same problem. Research finds most organizations have a philosophy but only a minority believe employees understand it. | Share it. Walk the team through it once. Put it where a new hire will see it during onboarding. |
| Being vague to avoid committing | 'We pay competitively' commits to nothing and answers no question. Vague philosophies are long because specific ones are short. | Name a target. Not 'competitive', but 'around the market median'. Vagueness is a decision you have deferred, not avoided. |
| Making exceptions without recording them | After enough undocumented exceptions the philosophy no longer describes your company, and you cannot tell when that happened. | Every exception written down, with a stated reason and a named approver. If you cannot write the reason, it is not an exception. It is a breach. |
| Letting salary history set pay | You import someone else's pay inequity into your company and compound it. This is also prohibited in a growing number of jurisdictions. | Set pay from the role and the level, never from what the person previously earned. State this explicitly in the philosophy. |
| Never revisiting it | The market moves, you hire remotely, you enter a new state, and the document quietly stops being true while everyone still believes it. | Review annually in a named month, and treat any material business change as a trigger regardless of the calendar. |
Keeping It Alive Without an HR Team
Writing the philosophy is an afternoon. Keeping it true is the actual work, and it fails at small companies for a completely predictable reason: the document is in a folder and the pay decisions are in people's heads.
| What Needs a Home | Why | What Happens Without It |
|---|---|---|
| The philosophy itself, where employees can read it | A philosophy nobody has read does not do the thing philosophies do, which is answer the question before it gets asked. | You have the document and the same problem, which is the most common outcome by a wide margin. |
| Each employee's level and range | Internal equity is a promise you can only keep if you can see who is in which role at which level. | The commitment quietly becomes false and you find out from the person it failed. |
| Every exception, with its reason | Exceptions are how a philosophy dies. They are individually reasonable and cumulatively fatal. | You cannot tell whether your pay structure still matches your stated principles, because nobody recorded the divergence. |
| The annual review, with a date and an owner | Markets move and companies change. A philosophy that is never revisited is eventually a description of a company that no longer exists. | It stops being true and everyone keeps citing it, which is worse than having none. |
This is where FirstHR fits. Document management keeps the philosophy where employees can actually read it rather than where it was emailed once. Employee profiles hold role, level, and compensation, so the internal equity commitment is something you can check rather than something you hope is true. And because exceptions become recorded events rather than remembered conversations, you can tell whether your stated philosophy still matches what you are actually doing.
FirstHR is not a compensation benchmarking tool and does not tell you what the market pays. What it holds is the philosophy, the structure it implies, and the record of where you departed from it. Any explicit commitment on pay equity or transparency is worth running past counsel before you publish it, since it also engages the Equal Pay Act and a growing patchwork of state pay transparency law.
Frequently Asked Questions
What is a compensation philosophy?
A compensation philosophy is a formal statement of the principles a company uses to decide how it pays people. It answers the why behind pay: where you aim to sit relative to the market, how you balance salary against bonus and equity, whether people doing the same job are paid the same, what you will tell employees about pay, and who makes the decisions. It is typically one page. It is not a salary structure or a list of pay bands, though it is the document those things should be built from.
What is the difference between a compensation philosophy and a compensation strategy?
The philosophy is the why, the strategy is the what, and the policy is the how. The philosophy states your principles: we pay at market, we do not negotiate individually, we are transparent about ranges. The strategy translates that into decisions: which benchmark data we use, what our pay bands are, how we structure variable pay. The policy is the operational rules: who approves an offer, when reviews happen, how exceptions are documented. Most small companies conflate all three, which is why their pay decisions feel arbitrary.
Does a small business need a compensation philosophy?
Yes, and arguably more than a large one, because a small company has no structure to hide behind. Every offer you make sets a precedent, and precedents made in isolation do not add up to a policy. They add up to an inconsistency you will eventually have to explain to someone who is upset. A one-page philosophy written before your fifth hire is dramatically cheaper than a pay equity problem discovered at your twentieth.
What does lead, meet, or lag the market mean?
These are the three market positioning options. Leading means paying above the market median, often around the 75th percentile, to win talent that is scarce or expensive. Meeting means paying around the median, at roughly the 50th percentile, which is the most common choice. Lagging means paying below the median on cash, and offsetting it with something else such as equity, flexibility, or mission. Most companies are actually hybrid: they lead for a few critical roles and meet for everyone else.
What should a compensation philosophy include?
Six decisions at minimum. Your market positioning target. Your pay mix, meaning the balance of base salary against bonus, commission, and equity. Your internal equity commitment, meaning whether people in the same role at the same level are paid the same. Your transparency stance, meaning what you will actually tell people. Your geographic approach for remote employees. And your governance: who decides, how often you review, and how exceptions are handled.
How long should a compensation philosophy be?
One page. Possibly less. The purpose of the document is that people can read it, understand it, and repeat it accurately to each other, and a five-page document fails all three tests. Length is often a sign the author was avoiding a decision: vague philosophies are long because specific ones are short. If you cannot state your market positioning in a sentence, you have not decided it yet.
Who writes the compensation philosophy?
At a small company, the founder or owner, because it is fundamentally a statement of values and budget rather than an HR technical exercise. It should be reviewed by whoever runs your finances, since it commits money, and it is worth having employment counsel look at any explicit commitments around pay equity or transparency before you publish them. What you should not do is delegate it entirely, because the philosophy is a decision about what kind of company you are.
Should I make my compensation philosophy public?
You should at least make it internal. Publishing it externally is a real choice with real consequences: it becomes a promise candidates will hold you to and competitors will read. Some companies publish everything, including individual salaries. Most do not. The minimum useful commitment is that every employee can read the philosophy and understand how their own pay was determined, even if you never publish a single number.
How often should I review my compensation philosophy?
Annually is a sensible cadence, and any material change to the business is a trigger regardless of the calendar. Hiring in a new state, hiring your first remote employee, entering a market with a very different wage level, or a significant change in your funding all justify a revisit. The philosophy should be stable enough that people can rely on it, and reviewed often enough that it does not quietly stop being true.
What if I cannot afford to pay at market?
Then say so, honestly, and be clear about what you offer instead. A lag position is legitimate when the tradeoff is real and named: more equity, more flexibility, more responsibility earlier, better benefits, work that matters. It becomes a problem when it is unstated, because then employees fill the silence with their own explanation, and the explanation they choose is rarely generous. The dishonest version of lagging is not paying below market, it is pretending you are not.