Yesterday’s Advisor featured consultant Diana Neelman’s five approaches to job evaluation. Today, her colleague, Sarah Schmidt, on data aging and market consensus.
 |
Neelman is a principal and executive vice president with Compensation Resources, Inc. (CRI) in Upper Saddle River, New Jersey. Sara Schmidt, CCP, PHR, is a consultant with the company. Their remarks came at a recent BLR-sponsored webinar.
Getting to Market Consensus
Schmidt says that to get to market consensus, you have to:
- Determine average annual salary increase for jobs
- Consider your philosophy (lead or lag)
- Adjust survey data to account for time differences
- Calculate central tendency
The results will yield a single market value for each job, says Schmidt. Then you compare and analyze internal data to the external (market) data.
Aging Data from External Sources
As data from different sources are often linked to different dates, you must “age” the data to be able to compare them. Most people age to the fiscal year, but you may want to lead or lag, says Schmidt. She offers the following example of aging market data:
Aging Example
Say that survey data indicate an average salary of $20,500, effective 04/01/14. You wish to age the data to 12/01/14 with an annual adjustment factor of 3.0%.
1. Determine the number of months between the survey’s effective date and the adjustment date:
- Survey effective date: 04/01/14
- Adjustment date: 12/01/14
- Difference: 8 months
2. Adjust survey data to account for the time difference:
- Calculate monthly change (3.0% divided by 12 months = 0.25% per month).
- Calculate total change (0.25% multiplied by 8 months = 2.0% total adjustment).
3. Apply the change percentage to the surveyed salary amount to determine the salary for the specified date ($20,500 increased by 2.0% = $20,910).
As you budget for 2015, you may be considering changes to your pay structure, to get maximum value from your employee compensation budget. Join us on November 6 for an in-depth webinar on how to determine pay grades, and laying the groundwork for getting your entire job value puzzle right. Register here!
Calculating Central Tendency
Once data from various sources are aged, to arrive at a market consensus, you need to work with measures of central tendency, says Schmidt.
Mean: simple average.
Median: the middle number.
Trimmed Mean: average after eliminating the high and low data elements.
Market Consensus (MC): the average of the Mean, Median and Trimmed Mean. This figure represents the best estimate of the market value for the position.
Schmidt offers the table below to demonstrate central tendency measurements. The mean uses all the data; the median uses just the middle number; and the trimmed mean leaves out the top and bottom numbers. Those three figures are then averaged to get the market consensus.
|
|
Salary Data |
|
Salary Data |
|
Salary Data |
|
|
$21,000 |
|
$21,000 |
|
$21,000 |
|
|
$21,800 |
|
$21,800 |
|
$21,800 |
|
|
$22,900 |
|
$22,900 |
|
$22,900 |
|
|
$23,500 |
|
$23,500 |
|
$23,500 |
|
|
$27,300 |
|
$27,300 |
|
$27,300 |
|
Mean: |
$23,300 |
Median: |
$22,900 |
Trimmed Mean: |
$22,700 |
|
Mean |
$23,300 |
|
Median |
$22,900 |
|
Trimmed Mean |
$22,700 |
|
Market Consensus |
$23,000 |
Typically, Schmidt says, you may consider that you are “at market” as long as you are within 10% plus or minus of the market consensus.
Whether your goal is to reward performance, time, knowledge or a combination of all three, establishing and solidifying your pay grades is the first step in building an equitable, competitive compensation structure. Join us for an in depth webinar on Assembling a Pay Grade System: A Step-by-step Guide to Getting It Right.
Evaluating Nonbenchmarks
Not all positions can be benchmarked, says Schmidt. Your organization may have unique jobs and hybrid jobs that won’t have sufficient survey matches to make a benchmark determination. Typically, these nonbenchmarks are examined against comparable benchmark jobs, looking at such factors as job scope, internal worth, and position in the hierarchy.
Evaluating Hybrid Jobs
Hybrid jobs are those where the individual has assumed dual or multiple roles. This is especially prevalent due to the recession. They are often not benchmarkable because you created them according to your own special needs, says Schmidt. Here is her suggestion for handling this situation:
- Identify the major responsibilities of the job and assign the percent of time devoted to each.
- Price each individually.
- Blend the results together according to the percentage weight.
However, you have to use common sense, says Schmidt. For example, if one job is high value and one low, you may not want to devalue the higher job too much.
Final Steps
After completing your job pricing, stop to take the time to reconfirm. Take a commonsense look, says Schmidt. Make sure your conclusions make sense.
If you have outliers, look to see where there might be problems, Schmidt says. Maybe there is a good explanation. For example, perhaps you had to hire out of range to get a critical skill, or maybe the incumbent is brand new with no experience and very low salary.
Also, look critically at all your external inputs, says Schmidt.
- Are market data inputs reasonably close?
- Do the market data reflect how you value the position?
Next, determine the difference in pay between market data and current salaries for incumbents.
- Determine the overall average difference with the market.
- Consider market adjustments to specific positions or job groups or the whole salary structure to maintain competitiveness.
As a final caution, Schmidt says that you must, of course, determine the budget impact of your plan.
Tomorrow’s Advisor will explore how companies should approach the problem of being unable to fill important positions despite large applicant pools.