Webinar Recap: How to Build Compensation Bands Using Real-Time Market Data

Pave recently hosted a webinar explaining how to build compensation bands using real-time market data. Speakers included Meghana Reddy, an advisor at Alamere Compensation and former VP of People at Loom, Joe Bast, a VP of people at Laika, Julie Menge, a people operations and analytics lead at Mux, and Katie Rovelstad, head of benchmarking operations at Pave.

‍

Here are some of the key highlights from the session. Watch the full webinar here.

The anatomy of a compensation band

Source: Alamere.io

‍

A compensation band defines the range of compensation given for a certain role within an organization. These bands exist to provide a fair framework around why companies pay a given amount for a certain role. 

‍

Compensation bands are based on levels and job families: the more senior the level, the higher the pay. You may have different compensation bands for the same level across different roles too. 

‍

Within each compensation band are a minimum salary you’ll pay for the role, a midpoint, and a maximum salary you’ll pay for the role. The midpoints should be based on market data and benchmarked against other companies you’re competing with for talent (more on this later).

‍

While compensation bands  between adjacent levels in the same job family can overlap (if say, a Level 1 employee is getting the maximum in their band, and a Level 2 employee is getting the minimum), ideally an employee will get promoted to a new band rather than max out their current band.

Defining job families

Source: Alamere.io

‍

To figure out how many bands you need, and which compensation band certain roles belong to, it helps to define the job families that exist at your organization. A job family is a group of jobs that involve similar work, training, skills, and expertise. For instance a mobile software Engineer, a backend software engineer may both be categorized in the “software engineers” job family. 

‍

Job families tend to be minimal for small organizations (say, technical vs. non-technical), but grow more complex as the organization scales and more specialists are hired. For example, you may start with one job family for all members of your sales team. Once you start hiring for sales operations, you decide to create a new job family and corresponding compensation band that aligns with the salary expectations and career progression for an employee in a sales ops role.

‍

Typically, the need for a job family arises once you have more than 1 person in a specific role. A large organization may create a job family for every unique role within the organization so they can pay more competitively.

Selecting the right band options

Source: Alamere.io

‍

Now that you understand job families and the structure of a compensation band, it’s time to use data to set up the various compensation bands within your organization.

‍

To determine which data to use when setting up your bands, consider the type of people you’re aiming to hire. What type of background do they have? Where do they currently work? Where are they geographically based? 

‍

Different types of employees have different salary expectations. The goal is to pull from a dataset that allows you to be competitive with the other companies you’re competing with for talent. 

‍

Datasets that can be instructive include the total capital raised from the startup you’re competing with, their valuation, their revenue, and their number of employees. Note that you can also use different datasets for different roles. For example, you may want to hire sales talent away from a large organization, but you’d prefer to hire marketing talent from a smaller startup.

‍

Within each dataset are the various percentiles you can benchmark your band against. Most companies will look to set the midpoint of their band around the 50th percentile (median) of the dataset they’re benchmarking against. However, if you’re competing for hard-to-find talent, you may be required to set your midpoint around a higher percentile. Again, this is a choice that depends on your unique business situation and who you’re competing with for talent.

‍

Another consideration when it comes to salary percentile is pay type. Do you benchmark the percentile against base salary, or total compensation (e.g., insurance, benefits, etc.)? For most job families, it’s advisable to benchmark against base salary. The only exception are roles where a significant portion of income is made via a bonus structure (e.g., sales). 

Choosing range width

Source: Alamere.io

‍

The last piece of the puzzle is selecting the width (distance between min. and max.) of your various compensation bands. There are generally two options organizations can go with here:

‍

• Percentile
• Band width (Percent from midpoint)

‍

Percentile means assigning percentile benchmarks from your dataset to the minimum and maximum range. If, for example, your midpoint is the 50th percentile, you might make your minimum the 25th percentile, and your maximum the 75th percentile.

‍

Band width (or percent from midpoint) means making the upper and lower bounds a percentage higher/lower than the midpoint. Most companies make the minimum up to 10-20% lower than the midpoint, and the maximum up to 10-20% higher than the midpoint.

‍

When it comes to width, higher level bands will generally have wider ranges because senior-level staff require greater salary increases to retain.

How Pave Helps Build Compensation Bands

Pave’s new package, Foundations, leverages our Benchmarking data set and Compensation Bands product to enable organizations to build data-driven compensation bands with ease.

‍

Once you answer a few simple questions, Foundations helps you level your employees, build your bands, and highlight pay inequities. Pave focuses on crunching the numbers so you focus on building a world-class team.

‍

Learn more about Foundations here.

‍