Setting a Compensation and Leveling System


Nathan Young is the Founder and CEO of Paidwell, where he helps companies support and build competitive compensation programs. Previously, he led People Operations at HelloFresh and SolarWinds. In this guide, he walks through compensation setting and leveling, and optimizing the impact of your comp resources with minimal sophistication.

Table of Contents

Why is compensation setting and leveling important? What are the benefits of investing in it?

Compensation setting and leveling gives you a blueprint for scale – a lot of startups are worried that levels or bands are bureaucratic and limit dynamism. However, good compensation principles apply equally to companies with a handful of employees and those that have scaled to thousands. Companies that don’t think about compensation have to rip out their previous infrastructure to update it every 2-3 years. 

Payroll is a huge expenditure, you need to scrutinize it – for some companies, payroll is 70-80% of overhead. How do you know you’re spending that wisely? A lot of companies apply more scrutiny and discipline to software tool procurement than they do to payroll spend—even though it makes up a much bigger portion of their expenses. Knowing who to hire at what levels can save you a ton of money. 

It will inform and improve your hiring – the comp setting and leveling process helps inform who you’re hiring and how. I have a background in recruiting and 9/10 times, a rec doesn’t get filled because there was a bad intake that makes for an impossible candidate profile. You can’t hire someone with 10+ years of experience right out of school. Good leveling informs your recruiting to help you find the right person with the right skill set at the right compensation. 

What are the components of a thorough compensation system?

A solid compensation philosophy – it should tell you how you set what you pay, what you’re rewarding, and at what level. It says how often you might review comp and through what lens you’ll view comp. 

Who you compare yourself against – there should be a set of principles that come out of your philosophy around, “We pay xxth percentile.”, “We compare ourselves to these types of companies at this size and industry,” etc. 

Break down how to pay different profiles – how will you pay new hires? How are seated employees paid based on their performance? How do you deal with top performers vs. poor performers? 

A good compensation system should clearly differentiate career progression – as employees move up in the job, the rewards should increase as well. This creates an incentive for employees to progress aside from career development and lifestyle. 

The end product is a compensation matrix – a standard version of a matrix would include all functions, and might have groups of levels throughout the business with similar bonus or equity bands. 

You might have a different matrix for each function/department, or for tech vs. non-tech roles – depending on how much complexity and work you have an appetite for, each function might have its own matrix. For example, an HR department might have different benchmarks or practices for HR business partners vs. recruiters vs. systems people. Or, you might say that tech roles don’t have bonuses but they have higher equity and non-tech roles have lower equity with higher bonuses. 

When in a company’s development do you need to define compensation bands for different types of roles?

Whether you state your plan or not, you have a comp plan – even if you don’t set levels explicitly, you’re making decisions and setting expectations that will have a ripple effect down the road. The further you go with an unoptimized or unorganized comp plan, the harder it is to improve.

Start thinking about comp and levels as soon as you have 2+ employees –  as soon as you’re starting to grow, think about what levels you need to hire for, how they should be comped, and their career progression. Even if you’re not thinking about these questions, your employees will. Unless you lay it out for them, they speculate and look at the market; if they don’t like what they find then they might leave. 

Start setting salary bands when you have defined teams – I’ve seen companies with 29 employees that do this, and employees with 10,000s of employees that don’t. When you have more structure to differentiate based on the talent in your business and teams, you can start to set salary bands. 

The longer you ignore comp, the harder it is to optimize it – at some point you’re not just changing a spreadsheet, you’re changing a culture. The more people you have locked into an outmoded compensation system, the harder it is to change. Making that big course correction is a lot harder than making small adjustments along the way. You don’t need to start with the most complicated or sophisticated system right away. But if you’re going to hire a software engineer, you should have a decent handle on what level you need them at. Otherwise, you could grossly under/over-pay. The timing of your investment in comp setting and leveling doesn’t necessarily hinge on headcount, invest in compensation setting when you care about doing it right vs. doing it right now.

What philosophical questions do you need to answer before you begin compensation setting?

Should you pay for performance or pay top of market? 
First, ask yourself:How much effort do you want to spend managing performance vs. compensation? – some startups say they want the best talent in the market so that they don’t have to manage their comp and performance together. They just expect everyone to be an A-Player and pay them as such, and if they fall short you let them go. This is the kind of approach that Netflix or GE would take back in the day: a hardcore, “we hire the best and pay the best” strategy.  Other companies are okay with growing someone into their role and paying them less. Then the more value the employee adds over time, the more money they get paid. 
Advice:I’m a big proponent of pay for performance – it fell out of favor 10 years ago, but now all these companies that weren’t doing ratings or paying for performance are coming back to it. It helps the employees realize what they get paid for, it’s tied to something tangible, and if the company succeeds, they succeed as well. It’s a powerful lever to tie compensation to your strategic goals.

Pay for performance is equitable – if you have the right bands, levels, jobs, and you tie comp to performance, you don’t have to manage DEI in comp. Because everyone at the same level of performance in the same job and level should get paid the same. A really well-designed compensation system is equitable and tied to the long-term strategic needs of the business, so the individuals feel good performance personally. 

Who should you compare yourself to?
First, ask yourself:Where do you want to be competitive? How competitive do you want to be? – ask what kind of company you want to hire from and what percentile you want to be within that. Once you know that, you can shape your data set for comparisons.

Do you want to use comp as a strategic differentiator? – when you’re trying to recruit and retain talent, ask where you want to be and where you can add the type of talent that can get you there. Or, if you want talent that can grow in the role, where do you find them? It could be at companies smaller or larger than you. 
Advice:You can always go up, but it’s almost impossible to go down – a lot of companies want to start in the 90th percentile; but what if there’s a downturn and you have to lay people off? It’s smart to start at the 66th or 75th percentile and you can bring it up later when you’re sure you’re doing well. This is better than having to lay people off or bring comp down because you set aspirational benchmarks. 

In most cases, avoid comparing yourself with a portfolio – companies often put themselves in a portfolio bucket to compare themselves against firms with similar demographic traits (like size, location, headcount, etc.) and look at how they can be competitive in hiring among peers. But 90% of the time, that portfolio isn’t actually where you’re expecting to get your talent from. Even if you’re trying to hire from competitors, if you’re benchmarking the same things then why should someone leave for a similar job at a similar company with similar pay? Most companies are trying to grow, and they hire up-market because that’s where they want to grow to. A portfolio approach makes sense if you’re looking for accounting and HR hires who have experience in your space, or if it’s for your executives because you want them to have experience, knowledge, and perspective in a certain space, industry, or market. 

Should you have a policy of offering cost-of-living (2-3%) increases on an annual basis?
First, ask yourself:What are you paying for? – if you’re paying for performance, you want to budget for increases of 2-3%; but then don’t do set scheduled increases. If you’re paying for performance, that’s antithetical to your performance. The quota isn’t a bad idea if your philosophy is to grow people, and give them the benefit of the doubt with a projected minimum increase regardless. 
Advice:Don’t do it as policy, but leave room in the budget – budget for increases of 2-3% or whatever the cost of living adjustment you normally do is but don’t have a quota policy for cost-of-living adjustments. If a person doesn’t hit their goals, you’d still have to give them an increase, and if you pay for performance this approach can be antithetical to your comp philosophy. 

Commit yourself to reviewing comp and take cost of living into account – I recommend not setting a cost of living adjustment as a policy but reviewing your comp holistically while keeping cost of living adjustments in mind. If you pay for top performers who are at the same salary band and their wages increase each year, those aren’t really cost of living adjustments; they’re cost of the wage market adjustments. 

Using set cost of living adjustments works with employees on fixed incomes – think about hourly roles that have the same salary across a variety of jobs or a sales job where everyone makes $50K plus commission. In these cases, a set cost of living adjustment makes sense.

Rises in cost of living are often temporary – rises in cost of living aren’t something to panic about. We often see it come down, and let’s say it falls in a year, will you take those increases back? Also, CPI takes into account a trailing 12-month calculation, it doesn’t take into account your last pay increase. If cost of living went up 8% and you got a 5% raise this year and last year, your employee comes out on top. 

Take into account total rewards changes – since 2020, a lot of companies have either done hybrid or remote models. The expenses of commuting (gas and time), fell off and it’s worth taking into account the total rewards attendant to that shift when thinking about recent inflation.

Should bonuses be based on individual vs. company performance?
First, ask yourself:What do you want your comp to reward and incentivize? – do you want employees to share in some of the benefits of the company’s performance, or do you want them to realize a cash total, so that if the person hits their target, they get paid (and by hitting their personal targets, the company hits its targets)? If it’s the latter, then the split should be between company and personal performance. It might be a sliding scale based on the scope of impact. For a VP, their personal performance might not matter much because they manage a function, so the company performance is what matters (and vice versa for more junior employees). 
Advice:Think about how the bonus might impact employee behavior and perception – if an entry-level employee’s bonus is 50/50 between their performance and the company’s, then 50% of the bonus may feel uncontrollable and more importantly, they may not even understand it if they’re still trying to figure out their own role. Think about the kind of message you want to send every 6 months when the employee gets a check or not.

I prefer a top-level approach – which is paying out the bonuses as a function of funding, meaning the company performance is tied to some top-level 0-100+% scale. Then their individual performance could be tied to a department or some OKRs or MPO. The idea is to have one number that’s personal, and another influenced by how the company is performing. The employee should be focused on hitting their own goals inside the calendar year because of the bonuses that will be paid within the next 12 months. 

Should you compensate performance with cash, equity or offer a choice?
First, ask yourself:How competitive is hiring? – in 2021 choice was offered a lot because the war for talent was hot and companies were trying to differentiate and get more talent faster. In 2023, I would not recommend offering choice unless a company wants to be more innovative or flexible. There are administrative costs that create a burden to managing choice. But if the economy goes the other way, you may want to be innovative and give talent their choice.
Advice:Blend short-term and long-term incentives – when thinking about your mix, consider blending short-term incentives like bonuses and increases in salary, with some long-term incentives in terms of equity. For each position, consider what performance you want to elicit over what time range

Don’t offer a choice if you can’t predict performance – if you want to offer choice, you need to be able to predict how you’re going to perform in 12 or 24 months. Right now, that magic eight ball is a little murky for some because of the economy. If you’re not certain how you’ll perform, don’t offer a choice.

How often will you adjust employee compensation?
First, ask yourself:How often do I need to be certain about comp? – there are reasons to review comp more and less often. How often do you want to make sure you’re doing the right thing with compensation? When the market is hot, or you’re doing bulk hiring, you might review more often, when it’s more stable, you may review less.
Advice:How often you adjust depends on how hot the job market is – in 2021, companies were reviewing compensation bands once a quarter while the job market was going wild. The comp data was perpetually lagging because it was moving so fast. Now, people are hesitating to give increases because they’re worried about the possibility of layoffs. 

Don’t review comp less than once a year – twice a year is even better until things go back to normal and comp movement is relatively flat with regular 3% increases. Then companies will likely go back to an annual review. Just because you review it doesn’t mean you have to change it. Try to adjust employee comp no less than once a year if you can afford it—most people expect a raise once a year.

If you can’t afford to give raises, don’t give raises be careful about setting expectations you can’t live up to. Come to the table armed with a spot bonus or equity and a darn good reason why you’re not giving them a raise. The lingering smell that’s with us right now coming out of the talent wars is that a lot of employees are not on the same page as their employers regarding whether things are slowing down. Telling someonem “I don’t want to do layoffs, I’m not gonna give you a raise,” is not a positive message that will comfort them. On the flip side, even giving someone a flat 3% right now might also fall flat. 

Comp is likely to cool off in 2023 – you may not want to decrease people’s actual pay, but you may want to move your bands down as the economy is cooling off. If in six months the economy is booming and everything’s great, raise your bands and stay competitive. If reducing your pay in 6 months helps prevent you from laying off in 18 months, reduce your bands.

How should you go about setting and standardizing base, bonus, and equity across different functions and levels?

Mix short-term and long-term incentives – you want a good mix of base, short-term incentives (bonus, commissions, etc.) and long-term incentives (e.g., equity). Then, employees know what they need to do over different timelines to accomplish something together over the next 3-5 years. That’s how you take an employee who doesn’t know if the company is succeeding or not, and create someone engaged in collective success. 

Compensate good employees more (or risk losing them) – a lot of the anti-work, quiet quitting trend can be root-caused to employees who don’t get paid more if they do a good job. By being a top performer, your employees should have access to career development, culture fit, and lifestyle benefits. 

When setting comp for any role, consider:

  • Company goals – consider how to align comp with the long-term goals and strategy of your organization.
  • What’s normal for the role – e.g. a significant portion of sales comp should be commission.
  • Your comp philosophy – e.g. your philosophy might be that only some employees have bonuses but everyone has equity. Or you can commit to pay for performance and everyone needs to have some kind of performance-based bonus.

For bonuses and equity, consider the degree to which employees at different levels should really feel personal and company success – if that’s at the entry level, start bonuses there. If it starts with managers because they’re the ones that feel whether or not they’re contributing, start performance-based comp there.  Equity should be a function of long-term value, so your philosophy might be to get entry-level people to stay 3-5 years with equity. On the flip side, some companies say that entry-level people should focus on their tasks, and they’re not looking at the long-term value of the business, so they don’t give them part of it.

Don’t avoid bonuses because you’re not good at setting goals – if you’re not good at managing performance and you don’t want to offer bonuses because you feel like they’re going to be fluffy, that’s not a good reason. Ask why you stink at setting goals, because you have to figure it out. If you told investors that you’re not committing to targets because you’re not good at setting goals, they’d laugh you out of the room. The best predictor of good performance is clear goals and comp tied to the goals, that’s the best way to make employees live and breathe that target.

There is no wrong answer unless it goes against your comp philosophy – if you say everyone matters and everyone contributes but not everyone gets equity, that’s a bit disingenuous. On the flip side, if you say you want career progression and start bonuses and equity at more senior positions, then everyone will understand why. As long as you’re consistent in your approach and levels and you incentivize the right things, there’s no wrong approach. 

What data do you need to set good compensation guidelines? Where can you source compensation data?

Who you benchmark against determines your data sources – there are data sources for companies in all industries and sizes. When you set your comp philosophy, there should be a set of principles that come out of it including, “We pay xxth percentile and we compare ourselves to these types of companies at this size and industry.” Whatever those guidelines are, find the dataset that satisfies those requirements in the best way.

There are many different data providers – e.g. Culpepper, Radford (great for tech firms), Mercer, Willis Towers Watson (WTW), McLagan (great for FinTech and Financial Services) and many other specialty data providers. The key thing is to find the most comprehensive data set out there that solves your problem with a reasonable investment. 

Aggregate tools are one-size-fits-none – some tools claim to pull in the data from all the different places, but they end up being one-size-fits-none. You won’t be able to scope the position you’re hiring for because the data set doesn’t fit your industry or market, and it doesn’t fit your comp philosophy. 

The data you get has to inform how you pay – if you’re in services or tech and 70-80% of your costs are from people, you have to make an investment in how you spend that 70-80% of OpEx. You should buy the tool that solves your problem and informs how you make that investment. 

How can you get started with free or affordable compensation data?

Look at job postings – if you’re in New York, California, or Colorado, you can go on someone’s job website, and see their pay. 

Take comp stats you find on Google with a grain of salt – those resources are typically self-reported and study after study shows that employee-reported numbers are inflated. However, getting the free tools is better than nothing. 

Add expense and sophistication as you mature – companies usually start with one data set that satisfies their requirements best. From there, some decide they want to fill in the gaps with another provider, and some blend datasets to be really comprehensive and have the best benchmarks for every job. You have to think about impact vs. sophistication and anchor in impact—especially early on. 

How transparent should you be about your compensation within your organization?

Be as transparent as you can manage – what level of transparency can your managers handle? If the management team built a deep comp philosophy, they understand it and how it’s competitive, they can talk about it. It’s not about whether you should talk about it, it’s about what degree you feel comfortable doing it and how you want to manage it. What level of transparency are you comfortable with and can you manage?

Pay transparency is going to grow and expand – look at the recent changes to pay transparency in New York, California, and Colorado. At a minimum, this shows that employees want to know. The NLRB already requires that you can’t stop people from talking about and sharing their pay with each other. 

You can do myth-busting around your process for setting pay – if you don’t want to be super specific and explicit with individual employees, you can bust some of the more insidious myths by sharing process details about your comp philosophy. You may share how you set pay, what goes into it, the ranges, the data you use, etc. 

Is it okay to negotiate with job candidates on compensation?

Be willing to negotiate within bands, especially if you have pay-for-performance – you should have some guidelines around how you’re setting comp and the range should build in flexibility. Good fences make good neighbors—so if you know what the guidelines for a new hire are and you have a range based on performance, qualification, education, and experience, then you should already have some degree of freedom. The key is to level against your internal parity to make sure you’re not hiring someone off the street for more than someone who’s a top performer or similar within your organization. 

Be careful not to let negotiation lead to unfair pay – some demographics are more inclined to ask for more money than others. If you’re paying someone more just because they ask for it, you may create unintended bias in your hiring and compensation process. Not only will that be illegal, but it also makes you a jerk for paying people differently based on demographics—whether intentional or not. 

With the right guidelines, you’ll know acceptable excuses to break range – if you know what the rules are, you should have justifications for extraordinary circumstances that call for you to break them. If you’re breaking the range and you don’t have a good reason why, you shouldn’t. Establish oversight and require approval for exceptions—breaking range should never happen just because a candidate asked for more. Possible reasons to break range include: 

  • Exact experience in your exact industry 
  • A rare skill set
  • Extreme success at past stops

How do you handle one-off comp increase requests (or the threat of a departing good employee)?

Comp increases should be grounded in your comp philosophy – your philosophy should clearly lay out when increases are granted based on performance or something similar. If you have someone that comes with an exceptional request, you’ll know how to manage it based on your philosophy. You might tell them that you don’t give increases when people just ask for more—or you might find a good reason. 

Be careful about giving people increases for threatening to leave – you might say you don’t give increases for people threatening to leave, you give them for people doing better over time. As long as their pay is tied to performance and in line with their peers, why would you pay someone more who’s intending to leave if they’re doing the same job at the same level?

Whether you wait until their scheduled increase is situation dependent – it depends on the person, how well they perform, and how long they’ve been in the business. Ask why they’re bringing it up now? Are they buying a house, or is it because they don’t want to wait three months for the increase? The latter feels like a poor reason, and if you reward it, you’ll get more of it. 

Just because you don’t want to pay a certain level doesn’t change the comp for it – if you under-comp someone or don’t give bonuses to a position that typically receives them, that employee might come to their senses about it at some point. You’re creating risk for turnover when you pay below market rates. 

How do you deal with “exceptions”?

ExceptionHow to handle it
Someone with an inflated titleTry to be honest and right-fit the employee’s title – this is necessary to keep people happy with their compensation. We often hear that titles don’t matter, they’re free. Well, titles are free until two Directors go for a beer and find out that one is paid $50,000 less than the other.

It can cause legal issues for protected demographics – if you have an upset person who’s not compensated at their title level (because they don’t have the right title), there’s going to be a legal issue if they’re in a protected demographic. You’re not building the culture of the company by being dishonest and you’re not helping their career development. Take an honest and competitive approach to titling and comp.
Existing employees whose comp doesn’t fit within the new systemTake an algorithmic approach to adjust prior pay – you should create an algorithm based on budget, market, and performance to translate outdated comp to a new system. This is the fairest way to give better-performing, lower-paid people more of your budget and high-paid, lower-performing people less.

Start out with a budget – you want to end up with everyone in range—but low-comp high performers should get more of your budget. Some companies will budget to just bring everyone into range automatically. If you can’t do that, people who are lowest paid and highest performing should get more of the budget compared to their peers.
Rising stars who are growing very fast and you want to keep them aroundLong-term incentives are crucial – take the total-rewards approach where you think about direct rewards (comp, bonuses, etc.), career development, and lifestyle and culture (which is why this person is in the business to begin with), and hook them with lots of long term incentives. 

They shouldn’t be paid more than peers in the same job with the same performance – however, they may have more opportunities based on their performance over time and they may get increases faster because they’re moving into different jobs and adding more value. I’ve seen some employees who move between jobs quickly and you never really get a sense of their true aptitude. Are they really a high potential A-player or do they just move between jobs every six months?

How do you factor in additional benefits to your compensation package?

Build out training and communications documents to communicate this value to employees – this really goes back to coaching and messaging; make sure that managers and hopefully team members understand that it’s not just about base pay. You want them to consider your whole rewards package. 

Find out what makes your hiring targets tick and tie rewards to the person and culture you want to create – you have to think about your compensation in terms of total rewards. This is made up of direct compensation (base pay, indirect compensation, benefits, time off, sick leave, and other benefits like tuition reimbursement and 401-K, health care, medical, and dental) and indirect benefits and your bigger picture employer value prop (career development, lifestyle, culture, etc.) Some people are willing to take less money and work from home or work on something that has a big impact. Others don’t care about anything else, they just say pay me. Decide the company culture you’re trying to build and think about what the hiring pool you pull from cares about, and direct your total rewards to appeal to them. 

What are the most important things to get right?

Build a compensation setting and leveling foundation early – a lot of startups say they don’t want to do leveling because it feels bureaucratic or reduces flexibility. But whether or not you have a compensation setting and leveling plan explicitly stated, your compensation practices exist in what you pay employees. It’s much easier to improve down the road if you start thinking about it intentionally early on. 

Build a system that can scale – a lot of giant companies will go through leveling and find out that some of their Directors or VPs aren’t really pointed at anything or shouldn’t have the equity they do. If you use compensation setting and leveling to grow intentionally, you can create career paths that help you build the company you want. 

What are common pitfalls?

Don’t take a portfolio approach to benchmarking compensation – don’t set compensation by comparing yourself directly with your peers or similar-sized companies that you’re not actually competing for talent with. Look at the places you’re hiring from.

Don’t overcomplicate it – ask: what’s the highest impact thing you can manage well? You can always make it more complex and optimize it later. Hiring managers need to be able to articulate why a bonus or equity system is in place and recruiters need to be able to quickly pull/calculate base, bonus, and equity for all given positions. More sophisticated isn’t always better.

More Resources

People & TalentPeople Ops

Want free guides?

We feature guides every month in our newsletter

Newsletter Sign Up