Skip to main content
What is a percentile?
Updated over 11 months ago

Percentile - A measure that tells you what percentage of the population falls below a certain value. In compensation terms, an employee ranked in the 85th percentile, with a salary of $150K, means that their salary is higher than 85% of the salaries of other employees with the same Compete roleand Compete Level in other companies.

Understanding percentiles in the context of salary benchmarking is quite straightforward once you break it down.

Imagine you have a list of salaries (sample size) for a particular job (Compete role & level), sorted from the lowest to the highest. The percentile is a way of ranking a specific salary within this list.

Screenshot 2023-12-18 at 11.19.56.png

  • 50th Percentile (Median): This is the middle value. If you are at the 50th percentile, your salary is higher than 50% of the people in that same Compete role and level and lower than the other 50%. It's the median salary.

  • 25th Percentile: If you are at the 25th percentile, your salary is higher than 25% of the people in that Compete role and level , but 75% of people earn more than you. It's below the median.

  • 75th Percentile: Conversely, if you're at the 75th percentile, your salary is higher than 75% of the people in that Compete role and level , and only 25% of people earn more than you in that Compete role and level . This is above the median.

So, in salary benchmarking, if a job's 50th percentile salary is $50,000, it means half of the people in that Compete role and level earn less than $50,000 and the other half earn more.

Target Percentile - The target percentile represents the point in the market pay range where you aim to position your salaries. This percentile is a strategic choice based on factors like industry standards, company policy, and competitive positioning.

By creating a Compensation Philosophy in Compete, you can define target percentiles across different regions (sites/pay zones) and departments.
The compa-ratio will be calculated and compared against the target percentile.

Did this answer your question?