Lorem ipsum dolor sit amet, consectetur adipiscing elit. Vivamus convallis sem tellus, vitae egestas felis vestibule ut.

Error message details.

Reuse Permissions

Request permission to republish or redistribute SHRM content and materials.

The Art of Setting Pay

Identify the best salary survey data, then use it with care.

Page Content

0513cover.gifEarly in her tenure with United Grinding Technologies Inc., Director of Corporate Human Resources Christine Taylor, SPHR, faced the daunting challenge of developing a formal salary structure for the company’s 140 employees. The trouble was that the jobs and the skills required for the grinding-machine business, based in Miamisburg, Ohio, were not like comparable jobs in most salary surveys. “It was very hard to find exact matches,” Taylor says. “We were not able to find benchmark jobs.”

Taylor worked with managers to develop detailed job descriptions listing the skills required for each position, then used the job descriptions to find survey data that was a close fit. She ended up relying on compensation data from the U.S. Bureau of Labor Statistics (BLS) as well as industry and professional association surveys to set median salaries and consistent salary ranges for most positions. Because she could not find exact matches for all of United Grinding’s positions, Taylor searched until she found two or three jobs requiring skills that were comparable to the manufacturing skills that her company needed. “It is more of an art than a science,” Taylor says.

Compensation is by far the largest cost item for most companies, so it demands careful consideration. Compensation professionals should find meaningful, reliable compensation data and then use the data to benchmark jobs and set pay levels according to the going market rate and the company’s compensation philosophy.

As Taylor’s experience illustrates, this process is not always straightforward. “A good benchmark job is a bridge toll taker because everyone knows what they do—they sit in a booth and take tolls,” says Ira Winsten, director of compensation and benefits for CenterPoint Energy Inc., an energy distribution company based in Houston. “It is when you get into higher levels and roles specific to the organization that it becomes more difficult to benchmark jobs.”

Choosing the Surveys

Although there has been some consolidation among compensation survey providers in the past few years, there is still an array of choices for compensation data, including consulting companies, industry and trade associations, and the BLS’s pay and benefits surveys. No matter the source, compensation professionals should evaluate each provider and its data with care to ensure that the data are timely and valid.

There are a growing number of free—usually Internet-based—salary data sources, but HR professionals should carefully weigh whether to use information from them. Any data that is self-reported by individual job holders about their own positions and salaries may not be as reliable as data produced by a more rigorous survey process.

“When data are reviewed by third-party professionals, those professionals can look at the job matches and spot any data inconsistencies, which makes that data more reliable than the self-reported data in some Internet surveys,” says Andrea Averill, a principal with Strategic Rewards Consulting Group in Philadelphia.

Compensation specialists who choose to use information from free sources should compare it with at least three other data sources to validate it, advises LoriAnn Penman, SPHR, director of human resources at Spectrum Comm Inc., a government contractor based in Newport News, Va. Free data from online sources tend to be at least a year old, she cautions.

Furthermore, surveys with fewer than 10 or 15 companies “do not necessarily provide a good sense of what the overall market looks like,” says Jason Adwin, vice president at Sibson Consulting in New York City. However, he notes an exception: A custom study of the largest 15 companies in a specific industry or labor market is likely to be more meaningful than a general study of 15 companies.

In addition, the cost of compensation surveys always raises concern among budget-conscious HR leaders. While free BLS data is obviously the most cost-effective option, some business groups and professional associations offer discounts or certain data for free to organizations that participate in salary surveys, so compensation professionals should ask about this when approaching a new survey source. Other providers, such as the Employers Resource Association, offer annual memberships, sometimes based on organization size, that allow access to all or a certain amount of salary data.

Prices for individual surveys vary from a few hundred dollars for professional association data to several thousand dollars or more for custom surveys. The key is determining what data the organization requires and finding a way to access it within budget.

HR professionals should push back if senior managers try to use budgets as an excuse not to invest in strong compensation data: Setting pay at the correct levels remains critical to an organization’s ability to attract and retain employees, and survey research ensures the right compensation investment.

Using the Data

Once HR professionals have the data they need, the next step is to benchmark jobs to find the best matches. The survey data shouldn’t be taken as the be-all and end-all when setting compensation.

When data are viewed too narrowly, compensation professionals do not leave themselves enough flexibility, Adwin says. Data are “to be used to make informed decisions based on the specific individual’s performance and experience,” he notes. For example, the data might indicate that a chief financial officer should be paid a certain amount. But a CFO with no experience and a CFO with 25 years of experience add different value to the organization, and compensation professionals need to consider that and other relevant factors.

Also, it might be necessary to make adjustments when a job in a survey does not exactly match an employer’s position. In fact, if the company’s job does not match at least 70 percent of the survey job’s description, compensation professionals should consider whether that survey job is a comparable match.

If compensation specialists cannot find a match for a specific job, they can use internal indicators, such as the job’s level in the organization, along with survey data, to price the job. For example, if specialists are pricing jobs A, B and C and have strong matches for jobs A and C but no match for job B, they can set pay levels by comparing the requirements of A and C with B. In this case, if job B’s responsibilities are higher than job A’s but lower than job C’s, a compensation professional can use the pay data on A and C to set pay for B.

Another consideration: Titles vary from company to company, and it is important to match job requirements rather than titles when using survey data. Some surveys include all jobs reported as, for example, an accounting manager without detailing any responsibilities. But the accounting manager in one company might be overseeing a large department, while the accounting manager in another organization might be limited to support work such as handling accounts payable. “If you see a survey that defines the role and the key job responsibilities, then that tends to signal a good survey,” Averill says.

HR professionals may also need to consider whether their organizations have hybrid positions, which have become more popular postrecession as organizations collapse and merge certain departments and functions. HR professionals whose organizations have hybrid positions might have to apply adjustments or discounts to market data to reflect the actual responsibilities of a given job.

“You might have a director of sales who is also doing some operations work—maybe 70 percent directing sales efforts and 30 percent operations,” explains Spectrum’s Penman. In that case, HR professionals can make sure overall compensation reflects those diverse job requirements by basing pay on that breakdown—70 percent of pay based on comparable sales positions and 30 percent based on comparable operations positions.

This is the case at Birmingham-based Children’s of Alabama medical center. Its compensation specialists rely heavily on health care industry data for clinical jobs, but their data searches often move beyond that.

“Our information technology jobs often require a clinical degree,” says Suzanne Thorn, SPHR, the organization’s director of HR services. Thus, Thorn looks for a comparable job that requires experience and a Bachelor of Science degree in nursing while also looking at information technology job data to see how the position fits into the IT structure.

As a final step, HR professionals should document the rationale behind compensation decisions. “Someone has to know the process well to communicate and explain it to employees if necessary,” says Don McDermott, president of compensation consulting firm D.G. McDermott Associates LLC in Red Bank, N.J. “You don’t need to go into all the details, but you need to make them feel comfortable that you have done a good analytical job of interpreting the data.”

The author is a New Jersey-based business and financial writer.


​An organization run by AI is not a futuristic concept. Such technology is already a part of many workplaces and will continue to shape the labor market and HR. Here's how employers and employees can successfully manage generative AI and other AI-powered systems.