Compensation surveys are often used to determine the appropriate market pricing when posting a new job or when evaluating an organization’s pay structure, Terry Pasteris explained during a recent BLR webinar.

Surveys can be a key component when it comes to creating competitive pay structures for an organization. That is why it is so important to ensure that the survey data is reliable and accurate.

Here are seven considerations when determining the data integrity in a compensation survey:

1. Sponsor reputation. Most surveys are conducted by a third party. If that third party is an organization with a good reputation, such as a well-known name in your industry or a well-established survey company, that is a good sign. It’s not fool-proof, however.
2. Quality of the data collection process. "All surveys—and most of them are online now—use some sort of format for collecting data." Pasteris noted. "Look at the quality of the data collection process to see whether or not you think that there’s quality data going in. If there’s not, whatever comes out of the survey may not be so useful."
3. Adequate job descriptions. Good compensation surveys have a description of a job, not just a title. The same title can mean different things in different organizations. For this reason, it is important to ensure that you match jobs by description, not just the title. Be especially aware of title inflation.
4. Large enough survey sample. Generally, the more organizations that are reported in the survey, the better the data. Large sample size can give an improved level of accuracy due to the averaging effect across the numbers. In other words, the effect of extreme outliers is diminished.
5. Competitor participation. It is important to know that the organizations you compete with (for talent, not necessarily for customers) are also in the survey.
6. Year over year reliability. Generally, we know that the market goes up each year—though the last few recession years have diminished this trend. Year-on-year reliability is simply referencing the data reported in the compensation survey over multiple years to ensure the trend makes sense.
7. Logical data trending. Much like year-on-year data, other trends can be evaluated to ensure that the data makes sense. For example, check some job families to ensure that the lower levels do in fact show lower pay rates than higher levels in the same job family. There should be a logical progression of scope and level changes and reasonable pay differences accordingly.

For more information on evaluating compensation survey options, order the webinar recording of "Pay Surveys: How Benchmarking Data Can Help You Competitively Price Your Jobs." To register for a future webinar, visit http://catalog.blr.com/audio.

Terry Pasteris is president of TLMP Consulting Group. She is both a Global Remuneration Professional (GRP) and a Certified Compensation Professional (CCP). Ms. Pasteris’ compensation work includes developing cash, benefit and equity programs and she has also developed performance management, staffing and communication solutions.