Salary Management and Compensation Analysis: A Guide for SMBs

Why salary management is harder than it looks

In most small and mid-sized businesses, pay decisions are made faster than they are documented. A salary is agreed in an offer letter, a raise is approved in a one-to-one, a counteroffer is matched to keep someone from leaving. Each decision is reasonable on its own. The problem is that, taken together, they add up to a pay structure that nobody designed and nobody can fully explain.

The symptoms are familiar to any HR leader. Two people in the same role, hired eighteen months apart, sit 12% apart on salary with no documented reason. Bonuses land unevenly because there was never a clear framework. The total people budget always seems to creep over plan, but no single decision looks like the culprit. And when a valued employee asks the hardest question in compensation, "why does my colleague earn more for the same work?", the honest answer is that the data to settle it does not exist in a usable form.

This guide is about replacing that drift with structure. It covers the building blocks of modern compensation, pay bands, salary benchmarking, compa-ratio, pay equity analysis and budget planning, in plain terms a 10 to 500 employee company can actually use. It is written for the UK and US market, where pay transparency expectations are rising fast and where getting compensation wrong is increasingly a legal as well as a retention risk. Towards the end, it explains honestly where Treegarden's HR module fits and what it does.

Structured pay versus the spreadsheet trap

The default tool for compensation in a growing company is a spreadsheet, and for a while it works. The trouble starts when the spreadsheet becomes the source of truth. It gets copied, emailed and edited in parallel; nobody is certain which version is current; sensitive salary figures end up in five inboxes; and the formulas that calculate averages and differences are only as reliable as the last person who touched them.

Moving to structured compensation management does not mean buying enterprise software you do not need. It means agreeing on a few things and holding to them: a defined pay band for each role, a single record of what each person earns and why, and a small set of metrics you review on a regular cadence. Three benefits follow directly from that discipline.

Equity and defensibility. When pay sits inside a banded structure, unjustified gaps become visible instead of hidden. You can show that differences map to experience, performance or location rather than to who negotiated hardest. In both the UK and the US, equal pay for equal work is a legal expectation, and a structured record is what lets you demonstrate it if challenged.

Retention. Pay dissatisfaction is one of the most common reasons people leave, but the driver is usually fairness, not the absolute number. Payscale's research found that improving an employee's perception that they are paid fairly can cut their intent to leave by around 27%, even when the underlying salary does not change. Structure and transparency are what move that perception. (See Payscale research summarised by NJBIA.)

Predictable budgeting. Without a single dataset, every salary review is a guessing exercise. With one, you can model the cost of a planned raise, see the aggregate impact before you commit and walk into a finance conversation with numbers you can defend.

Start with pay bands

A pay band is the foundation everything else rests on. It is a defined salary range, a minimum, a midpoint and a maximum, attached to a role or a level rather than to a person. The midpoint represents what you expect to pay a fully competent performer at market rate. The minimum is roughly where someone new to the role starts; the maximum is the top of what the role is worth before a promotion is warranted.

Bands turn pay into a deliberate decision. Instead of asking "what number will this candidate accept?", you ask "where in the band does this person sit given their experience and our market position?" That single shift removes most of the inconsistency that creeps in over time, and it gives managers a clear, fair answer when an employee asks how their salary was set.

Building bands does not have to be a six-month project. For each role, gather two or three external reference points (more on benchmarking below), set a midpoint near the market median, and define a range around it, commonly the minimum at about 80 to 85% of midpoint and the maximum at about 115 to 120%. Group similar roles into the same band where it makes sense, so you are maintaining a manageable number of bands, not one per job title.

Measure with compa-ratio and range penetration

Once bands exist, two metrics do most of the analytical work. Compa-ratio is an employee's salary divided by the band midpoint, expressed as a percentage. A compa-ratio of 100% means the person sits exactly at midpoint. Below 90% usually flags someone who is underpaid relative to the role and may be a flight risk; above 110% flags a salary that has moved ahead of the band, often through repeated retention raises, and should be watched.

Range penetration answers a related question: how far through the band, from minimum to maximum, does this salary sit? It is calculated as (salary minus minimum) divided by (maximum minus minimum). A new joiner might sit at 10 to 20% penetration, a seasoned performer at 60 to 80%. Looking at penetration across a team shows whether your distribution makes sense, for example whether your most experienced people really are paid further up the band than your juniors.

You do not need either metric to be precise to three decimal places. What matters is the pattern. When you sort a team by compa-ratio and the order does not match your sense of who contributes most, you have found something worth investigating, perhaps an overdue raise, perhaps a salary that was inflated by a counteroffer two years ago.

Run a pay equity analysis

Pay equity does not mean everyone earns the same. It means that differences in pay are explained by legitimate, job-related factors, experience, performance, the complexity of the role, location, rather than by gender, ethnicity or who happened to negotiate hardest. A pay equity analysis is the structured check that the gaps in your data are the explainable kind.

The method is straightforward in principle. Group employees doing the same or comparable work, then look at whether pay differences within each group track the legitimate factors. Where a gap remains after you account for experience, performance and location, you have an unexplained gap, and that is the one that creates both fairness and legal exposure. A practical workflow is to surface the largest unexplained gaps, document the reason for each (sometimes there is a good one), and cost out what it would take to correct the rest.

This is no longer just good practice; it is moving toward law. In the UK, gender pay gap reporting is already mandatory for larger employers, and the EU Pay Transparency Directive (Directive 2023/970), which member states must transpose into national law by 7 June 2026, will require employers to justify pay gaps above 5% that cannot be explained by objective criteria, and to publish pay ranges in job adverts. In the US, a growing number of states and cities now mandate salary ranges in job postings. The direction of travel is clear: companies will increasingly have to show their pay is fair, not just assert it. (Sources: European Council on pay transparency.)

Look at total compensation, not just base salary

Base salary is only part of what an employee actually receives. Health insurance, pension or 401(k) contributions, paid time off, bonuses, equity, professional development budgets and other perks all carry real financial value. When you reason about pay, and especially when you compare an internal salary to an external offer, you need the whole picture, not just the number on the contract.

Keeping each benefit recorded with its value, eligibility and period lets you calculate a total compensation figure per employee: the full investment the business makes in that person. That figure matters most in retention conversations. When someone arrives with a competing offer that quotes a higher base, a total compensation view often shows your package is closer, or ahead, once benefits and employer contributions are counted. Many employees simply do not realise what their full package is worth until it is laid out for them.

Plan the salary review budget deliberately

The annual review cycle is one of the most demanding processes in HR. You are balancing employee expectations against a fixed budget, prioritising raises for critical roles and trying to keep the outcome fair across the whole organisation. Run on circulated spreadsheets, it takes weeks and produces a result nobody fully trusts.

Done deliberately, budget planning is a structured exercise. Start from the total approved pool and distribute it down by department, role or level. Then model the trade-offs before you commit: an even 4% across the board versus differentiated raises of 2 to 8% based on performance and compa-ratio. Each option has a different total cost and a different effect on equity, and you want to see both before the proposal reaches finance, not after.

A clean cycle also assigns clear roles. Managers propose increases within the budget allocated to them and within the relevant bands; HR reviews the proposals, checks equity and consolidates; finance approves the total. Every proposal is attributable, so a year later you start from a documented baseline instead of a blank sheet.

Benchmark against the market

A salary can be perfectly fair internally and still uncompetitive externally, or the reverse: you can be paying well above market without realising it. Salary benchmarking is how you anchor your bands to reality. It compares your pay for a role to what the wider market pays for the same role, at the same level, in the same location.

Good benchmarking uses ranges rather than a single magic number. For a given role you want a sense of the spread, roughly the lower quartile, the median and the upper quartile, so you can decide where you want to sit. A company competing hard for scarce engineering talent might target the upper quartile for those roles while paying around the median for roles where supply is plentiful. That is a strategic choice, and benchmarking is what lets you make it on purpose.

Benchmarking is as important for retention as it is for hiring. If an experienced employee's salary has slipped well below the market range for their role, their flight risk rises whether or not they have started looking. Reviewing your team against the market on a regular cadence lets you act before a resignation forces a reactive, and usually more expensive, counteroffer.

A worked example: reviewing pay for a 50-person team

To make this concrete, here is how the pieces fit together in a single review cycle. Sarah leads HR at a 50-person software company and has been given approval for a 4% total salary increase budget.

Step 1, understand where you stand. Sarah looks at the current distribution by team and against the market. She sees that the QA team sits low against benchmarks while sales sits above the median, and a quick equity check surfaces two unexplained gaps between people in the same role.

Step 2, model the options. She builds three scenarios: a flat 4% for everyone, a performance-differentiated spread of 2 to 8%, and a hybrid that fixes the two equity gaps first and then distributes what remains by performance. For each, she checks the total cost and the effect on those compa-ratios.

Step 3, involve managers. Having chosen the hybrid approach, Sarah pushes a budget allocation to each manager. Managers propose individual raises within their allocation and within the relevant bands, with each person's current position visible to them.

Step 4, review and approve. Sarah checks that the two equity gaps are closed and the total stays within 4%, produces a summary by team for the finance review, and on approval the new salaries are recorded. The reasoning behind every decision is documented, so next year she starts from a baseline instead of a blank spreadsheet.

The value is not just speed. It is that every figure in that final summary can be explained, which is exactly what you need when finance, an employee, or one day a regulator asks how the numbers were arrived at.

Where Treegarden fits

Treegarden is an ATS and HR platform for UK and US SMBs, and compensation lives inside its HR module. It is not a standalone enterprise compensation suite, and it is honest to say so: if you are a global company running complex equity, long-term incentive and multi-currency comp plans across thousands of staff, you will want specialist software. For a 10 to 500 person business that wants to bring structure to pay, it does the things this guide describes.

Concretely, the HR module keeps a single compensation record per employee, benchmarks salaries against market ranges by role, level and location, runs pay equity gap analysis and flags salary anomalies that sit outside role norms. It supports budget planning with real-time spend tracking and scenario modelling for review cycles, with manager proposals routed through approval workflows. Because compensation sits next to the ATS, an agreed salary at hire flows straight into the employee's record without re-keying. And because salary is among the most sensitive data you hold, access is governed by field-level permissions and logged, with data handled in line with GDPR.

The point of all of it is simple. Every salary decision sends a signal about how much the business values a person and how fairly it treats its teams. Made on a stale spreadsheet, those decisions are guesses. Made on structured data, they are choices you can stand behind.