Our definition of pay benchmarking is obtaining robust data on typical reward packages for comparable roles in the external market.
What does good pay benchmarking look like?
Good pay benchmarking is both reliable (based on a large sample) and valid (based on a relevant sample).
So it’s important that your data source is defendable, your job matching is accurate, and the data’s statistical methodology is robust.
Benchmarking is a process to provide you with data to enable sound, fair and transparent reward decisions that are right for your business.
What are the steps?
1.Choose your data source
The main ways to obtain data:
- Job adverts: using one, or a handful, of job adverts lacks reliability as the samples are extremely small. Pay often being an ‘up to’ figure which is related to the very specific role described. Your role may have the same title but could be very different in scope and accountability. The salary stated is designed to attract and may not be what is subsequently offered.
- Crowd-sourced: lots of social media sites offer pay data. It is collated from users and usually self-reported. Although this information can be very up to date, matching is based on job title and pay is in danger of being inflated or misquoted. Data is often reported as an average which may mask extremes of pay in the sample such as regional variances and difference in job role.
- Published surveys: data is collected directly from employers, usually by a third party and they can be cross-industry or sector specific. You will usually need to buy these but some are published for free. The better surveys have a large participant group and a robust matching methodology so that companies and roles of all sizes can be matched in using more than just job titles. Clear data cleansing and mathematical processes enable reporting of a variety of statistics. You will often be required to participate by submitting your data before getting some back. The surveys may also provide additional data such as total cash and typical benefits.
- Private or club surveys: these are similar to the published surveys above but participation is restricted to particular group of organisations. The participants themselves may pay for and steer the survey but an independent third party will usually manage the data and reporting. Private surveys are particularly useful for industry groups where roles are specialist and therefore unlikely to be found in a general survey, or specific locations.
We recommend the use of surveys rather than job adverts or crowd-sourced data. We often review the survey market for our clients to find the best one for their needs. You can ensure the quality of any data source by checking it meets the reliable, valid, defendable and robust tests I’ve listed above.
2. Match your roles
Matching your roles into a survey can be a laborious task but it’s important to get right if you want your survey results to be accurate and accepted within the business (and of course if you want the survey to be reliable for other participants!).
Survey methodologies vary but the principle here is don’t match by job title alone. Understand the role content and level, and match accordingly. If you’re able, check your matches with line managers. If they are involved in the process, they are more likely to accept the outcomes.
A good survey will give you a number of data points: percentiles, quartiles, median and averages. This gives insight into the range of pay for your roles. Understand your market positioning on base pay and total cash (base + bonus) and what it might cost the organisation to make any necessary adjustments to pay.
The important thing is to interpret the data in line with what’s happening in your sector, your business and against your reward strategy. For example, should your reward budget be directed to particular hot skills, certain groups such as high performers or flight risks?
Our golden rule is that the market median is not the answer. It is simply what other companies pay other people for doing roles that are similar (but not exactly the same) as the one you are looking at.
All it provides is an indication of a competitive range of pay. You’ll need to decide how much to pay and this depends on other factors as detailed in your reward strategy as well of course as affordability.
Need help pay benchmarking?
Robust pay benchmarking is essential for informed pay decisions. If you need help with it, we’d be delighted to hear from you.
This short blog is part of the Verditer Reward Quarter. It’s where we bring clarity to specific topics that combine to create the world of reward. Think of it like exploring the distinct cultural zones (or Quarters) that make up a city.
Click here for more:
Reward Quarter No. 1: Reward Strategy
Reward Quarter No. 2: Job Evaluation