The company started with one guy and a great idea. Soon, things started to grow, and the entrepreneur decided it was time to take a modest salary. Soon, he hired a salesperson and a couple of full time employees. The back corner of the basement started to feel cramped, so they rented a cheap office. Before long, there were 30 employees on the payroll, most of them poached from businesses in the local area.
What started as a seat-of-the-pants payroll system started to look a little messy. Some employees made the same pay as they did in earlier jobs. Others, whose specialties were desperately needed in the burgeoning business, made more than the market might have warranted – but no one knew for sure.
And that’s the point where Bill Erdle of William J. Erdle Compensation Consulting (www.erdleconsulting.net) generally enters the picture. For new businesses, he says, “That’s when I usually get a call.”
This veteran compensation consultant helps get things organized, undoing or avoiding some common mistakes in setting up a compensation system. Here are some of the most common mistakes he sees.
Lack of objectivity versus internal equity. As an outside expert, Erdle doesn’t have a problem using a dispassionate approach when comparing one job against another. But a company employee might. When you choose a salary survey that includes the companies you believe to be your competitors for talent, take it seriously and try to be objective, he suggests.
“A lot of times managers or even HR people try to be objective when evaluating the jobs, but they have a hard time with it. Often, they’ll match an existing job with one that’s a higher level in the survey. Maybe they match the company accountant with a senior accountant in the survey, when the job should be graded and paid as a junior accountant. Part of the issue is they know the incumbent, and the incumbent’s performance may influence how they evaluate the job. You need to focus on the job alone. Performance issues need to be addressed elsewhere.”
At the same time, Erdle says that internal equity is a very strong force for some companies. “You can go through the process of setting up a system based on where you want to pay against your market, but you also have to consider the internal value of the jobs. How the company values the jobs internally ultimately decides how much they want to pay in relation to the market,” he says.
“For example, the market says that software engineers are a highly-paid subcategory of engineers; mechanical engineers, industrial engineers are paid less. But some companies want all engineers in their organization to be graded the same. They want their internal equity to be maintained, and grading them differently would take away from that. You may need to readjust the salary grades for those jobs so that they make sense for your company.”
Choosing and using the right survey. First, says Erdle, learn about the competition. “And I don’t mean competitors in your industry, necessarily,” he says. “I want to know about competitors for your people. I need to find out where the company wants to pay in relation to those other organizations, and then I look for salary surveys that have those organizations in them.”
While free and cheap salary surveys are available online, he warns against them, saying you will likely get what you pay for. “Usually in those surveys incumbents are reporting their own data. They might fill out a form on the internet. People tend to exaggerate the level of their job, and sometimes their pay. And sometimes the survey matches jobs just based on the title, not on the description of the job duties. They don’t often go through and match actual duties.
“I don’t think that is anywhere near as good a survey as one that’s done by trained compensation professionals matching and reporting data. You need a survey that has specific companies represented, and you know which companies are included. That provides face validity to the whole thing. It looks right. So if you’re a high tech company, Intel is in the survey, and so is Tektronix, and so is Mentor Graphics, for example. That goes a long way toward credibility for the survey.”
Once you’ve chosen the right survey, you need to make sure you’re mining the right data from it. “I always look at the total survey because it’s the largest sample size,” Erdle says. But he doesn’t stop there. “You can also select for revenue or geography or a sub-industry. Many times a company doing this themselves will pick exactly the selection they’re in, in terms of size and industry and location. What you really want to compare against is the companies you are losing people to and the ones you’re bringing employees in from.
“You want to analyze these other companies as opposed to only the ones in your specific niche. That might mean a higher data selection than just their current position, as far as revenue or other things. I look at maybe five different data selections, including the total survey, and go through and analyze them for each employee or employee group. If they’re at similar rates of pay that gives some indication of competitiveness.”
What if the pay rates for different data selections aren’t similar? Erdle says that one mistake he often sees is averaging these figures. “Averaging them doesn’t give you an accurate picture,” he says. Each data selection needs to be evaluated in reference to the job you are looking at, and then a decision is made about market competitiveness. Depending on the job, one data selection may be more important to consider than another.
“Even if you did average them, wouldn’t you want to do a weighted average based on the number of incumbents in each data selection instead of just a simple average?” he muses. Aside from the data selections to consider, you will want to ensure you have a large enough sample size, and you would want to analyze other statistics such as the median and salary range midpoint. The median is a much better statistic to use than the average when sample sizes are relatively small, but most people simply average everything, thinking they are doing the right thing.
Rewarding hard work. Typically, you’re not trying to reward hard work, Erdle says; you’re trying to reward an outcome. The reason? Employees can work hard and still not get the job done. “I did a project for a biotech company. They were getting ready for an audit by the FDA, and people were in there all night working. You’d see them during the day, bags under their eyes.
“One of the VPs told me that they like to reward employees for their hard work. I asked him what kind of behavior they were rewarding. ‘You’re rewarding people for working hard and not necessarily for getting the job done,’ I told him. So that’s the behavior they got. I tell clients not to reward based on working hard. If employees can get the job done quickly and efficiently, reward for that.”
Don't Overcomplicate Your Comp System!
Remember that employees won’t respond to a program they don’t understand. Erdle recommends that your compensation system be simple, without a lot of complicated bells and extraneous whistles. “You need to have a system that is simple enough to understand … and to explain,” he says. “I find that complicated systems just don’t work that well. I like to do things that are straightforward and simple.
“This stuff isn’t rocket science at all. You need something that you can explain to managers, and have managers explain to their employees. Once the program is up and running, you’re going to have employees coming in waving some internet survey they found and threatening to leave if they don’t get a pay raise. Make sure you have something explainable so you can stand behind it. It makes sense for everyone.”