Salary surveys are great tools to help your organization calculate starting pay, figure increases and avoid pay-discrimination complaints. But using inaccurate surveys is worse than using no surveys at all.
That's why it's important to use only those salary surveys that provide the right data, taking into account your industry, company size, region and other variables.
Here's a five-point checklist to help you spot survey weak spots:
1. Overly broad job titles. Job titles can mean different things in different organizations, even those in the same industry. So a "sales manager" at Company A can have entirely different duties than a sales manager at Company B. That's especially true for IT jobs such as senior systems analyst. Duties often change rapidly while the titles remain the same.
Find out the job descriptions the survey used to match its categories.
2. 'Average' versus 'median' salaries. Many surveys report only "average" salaries, which provide an incomplete picture of pay ranges. Median salary, the salary in the middle of the range, is a better gauge. Median salaries are usually 3 percent to 5 percent below the average.
3. Moldy data. Salary survey figures can be as much as 18 months old. Determine the survey date. Then, "age" the data, when appropriate. For example, if a survey reports how much people expect salaries to increase over the next year, apply the anticipated raise to salaries for the coming 12 months.
4. Improper employer size. Most employers in the survey are bigger or smaller than your organization. That flaw can prompt you to set salaries that are higher or lower than similarly sized organizations in your industry.
Seek out surveys that allow you to slice the data by employer size.
5. Sample size skews results. When a survey's sample size isn't big enough, a handful of results can completely skew the numbers. Always ask about sample size and lean toward the bigger ones. Use several different surveys to make sure you're getting a complete picture.