Benefits and Compensation

How HR Can Get the Most from Salary Surveys

There are plenty of salary surveys available and, arguably, there’s not a lot of difference between them. Most provide a number of incumbents, a mean, and multiple percentiles. Some use number of employees to describe company size and some use annual revenue. Most are broken down by industry and geographic location. Some provide position level charts, though some don’t.

salary survey

Timurpix / iStock / Getty Images Plus

Cost is one differentiator, with prices for salary surveys that vary from free (if you participate) to tens of thousands of dollars. Whether your salary data is a good buy or costs an arm and a leg, the value of your salary survey can be enhanced or hindered by how you use it.

Pick A Number

One of the first things you’ll need to do is to decide which piece of market data to use. As mentioned above, most salary surveys provide an average (the mean) for each demographic as well as several percentiles, typically: the 10th, 25th, 50th, 75th, and 90th. The challenge is to determine which one of those numbers is right for your organization.

Having a well thought out pay philosophy in place before you begin conducting a salary market analysis helps smooth the process of using market data to set pay rates. If you don’t have a pay philosophy in place, think about how competitive you want to be in attracting and retaining talent. Best practice is that paying at the 50th percentile is paying at market, which is a good plan if you want to be moderately competitive. If you want to attract and retain the cream of the crop, however, consider paying at a higher percentile.

Everybody Has One

In general, salary surveys focus on jobs that can be found in many organizations—benchmark jobs—for example: accountant, receptionist, sales rep … you get the picture. Benchmark jobs provide the most incumbents in any given survey and the more incumbents, the more valid the data.

The more specific the job, the fewer incumbents, which is one of the reasons why the rule of thumb is if the job as described in the survey is at least a 70% match to your job, it’s a viable indicator of the market rate for your position.

The Thingamadidgeter

Every organization has nonbenchmark jobs and finding any kind of match for them in a salary survey can be a challenge. If that’s the boat you’re in, you still have options for finding viable market data. If the job is specific to your industry, contact the relevant trade group(s) to get salary data. Many of them provide salary data at no charge if you participate in their survey.

Another option is to break the job down into its major elements, find market data for jobs that fit those portions, and blend the data. Be sure to weight the data depending on importance and how much time is spent on each portion. Using the survey mean for the elements of the position, it would look like this:

Job title: Thingamadidgeter



Survey Mean


Weighted Rate

Element A





Element B

Facility Maintenance




Element C






Blended Rate:


Though blending salary data can provide a market rate for those hard to find jobs, be mindful of how you implement it. Using the example shown above, if you have an accountant that is really good at her job and nonrelated responsibilities are added to her plate, would you downgrade her salary because the market rate for the additional responsibilities is less than the rate for an accountant? I don’t recommend it, otherwise you very well may lose a really good accountant.

If you want to attract and retain the most qualified employees, use the salary data for the job with highest level of duties and responsibilities in that mixed bag of job requirements. If all of the thingamadidgeter position’s elements are similar, however, or if the highest level element is only a small portion of the position, then blending the data can be a good path to the market rate you need.

If It Walks Like a Duck …

If you’ve searched your salary survey from top to bottom and still can’t find the job title you need for viable market data, consider “comparable” in lieu of “matchy-matchy.” We all know that one employer’s number cruncher can be another employer’s bean counter, and let’s face it some employers can be quite creative with job titles, so it’s important to look beyond titles and dig into job descriptions when reviewing salary survey data.

With that said, there’s still a better than average chance that some of your jobs may not be included in your salary survey. You could opt to acquire another survey that gives you a broader range of jobs or, especially if you’re on a tight budget, consider looking for comparable jobs. In other words, look at jobs that require knowledge, skills, and abilities that are comparable to your position. Be sure to consider educational and experience level requirements as well.

Jumping Off the Roof

Were you a kid that told your parents your reason for doing something they didn’t allow was that all your friends were doing it? That logic didn’t get you out of hot water then, and it may not cut the mustard with your senior management now.

Paying the market rate your salary survey indicates can be a bit like jumping off the roof because all your friends did. Don’t get me wrong, salary surveys provide a useful indicator of the going rate for many, many positions but they’re not all you need when it comes to setting pay rates.

Just because your survey says employers in your demographic are paying a certain amount doesn’t mean it’s what’s best for your organization. You also have to consider internal equity and your budget. Realistically, you can’t pay more for that stellar candidate than the company can afford. Additionally, if the new thingamadidgeter will be in the same job as other employees in your organization, you need to pay the new recruit comparably or you might run afoul of those pesky employment laws.

A Snapshot in Time

Surveying thousands of employers and corralling all that data takes a lot of time and effort so it’s important to be aware that it provides a snapshot of the market at the time the participants completed the survey questionnaire.

Every good salary survey displays its effective date, which allows for aging the data from the effective date to the present, or beyond, depending on whether your market position is to lag, lead-lag, or lead the market.

Out in Left Field

Some salary surveys include “outliers” in their data but I’m a bit leery of using them. I recommend passing up data points that are either a lot lower or higher than you expected. One way to identify demographics with outliers is to compare the mean to the 50th percentile.

They’ll never be the same but they should at least be in the same ballpark. If they are too far apart, there may be outliers in the salary data. Outliers can also skew the percentile data, which should be a relatively steady progression.

In addition to traditional salary surveys, there are online salary data providers that assert their data is up to the minute. My understanding, however, is that much of it is provided by employees rather than their Human Resources department. As a comp professional I’m skeptical of using employee provided data. It’s just too easy for employees to consider only their job title and overlook the elements of the job itself when reporting their pay data.

Additional Resources from

Sharon McKnightSharon L. McKnight, CCP, SPHR, draws from more than 20 years of management experience, including 6 years as a director of human resources, to develop compensation administration tools and write about compensation issues. Her experience in both operational and HR management provides her with a practical approach to providing online resources that address the challenges facing compensation and human resource professionals.