A Beginner's Guide to Payroll Analytics

Payroll analytics can help you pinpoint who may be lying on their time card, where you have too many employees, and even who needs a raise. The Ascent goes through several key metrics.

Most of the time, when you hear about analytics, it’s while watching sports, and an analyst who's never even played the game is criticizing your team. He brings up some fancy-sounding acronyms to show that accuracy and efficiency are down and bad decisions are up.

From then on, you likely don’t associate analytics with good things in your mind.

I've experienced this, too -- Aaron Rodgers is still a top-tier quarterback and just needs better wide receivers, obviously -- but I think analytics can actually be very useful in a business setting.

Payroll analytics will improve how you do payroll, help you discover if someone is reporting too many hours, and help you ensure your employees won’t leave you hanging by jumping to a competitor. It can even let you know if one of your departments is overstaffed.

In this article, we’ll go over several examples of payroll analytics and the metrics you use for payroll analysis. We’ll also discuss how to use business analytics to drive incentive pay.

Track errors and fraud

Every once in a while, an employee will let you know their paycheck should’ve been bigger because they forgot to include hours on their timecard. Less often, they’ll let you know they were overpaid for the week. These payroll mistakes aren’t a big deal if they really only happen every once in a while.

However, it’s still helpful to run a report to see how often you are doing more than one payroll run each period and try to pinpoint the cause. If you’re constantly having to repeat the entire payroll process to add shifts for the same person, they may need more training on how time tracking works on your payroll software.

Additionally, the time and money spent doing second or third manual payroll runs adds up over time.

A quick and easy fraud check is to create a payroll analytics report that shows weekly pay trends for employees over time. Any big spikes that aren’t associated with raises could be because the employee is reporting more hours than they actually worked.

The employee may try to hide this by claiming they forgot to add a shift so that you don’t notice during the payroll reconciliation.

Compare revenue and compensation costs

When your small business reaches a big enough size, you will eventually need to start departmentalizing. It makes sense for some overhead costs, such as accounting and HR, to be centralized, but costs like research and development, materials, and direct labor should be charged to their individual departments.

Picture a big box retailer -- the pharmacy, electronics, grocery, and clothes departments will all have their own supervisors and should all have their own profit and loss statements.

When you’re working on the budget for the year, use market research and your own experience to set the expected labor cost for the department. At the end of the year, if the actual labor cost is far higher than it should be, it may be time to move people around or lay some people off.

Conversely, a super low labor cost isn’t necessarily a good thing because it could mean that the employees in that department are overworked and ready to leave.

Budget for new hires

Think about the last time you hired someone new to a big position. How did you decide how much to pay them? It’s likely that you used some combination of what the last person in the position made and what the new hire was currently making.

This approach makes sense -- you probably won’t be able to hire someone if you don’t pay them more than they make now -- but it’s only part of the equation.

Over time, if you’re constantly paying more than market rates to hire new people, you’re going to have higher-than-market costs and not be able to compete.

Each new employee, above a reasonable pay threshold, should be budgeted for. Especially in human capital-intensive industries.

If new hires are pushing your labor costs far above your budget, it may not be sustainable. Consider focusing on hiring lower-level employees for that department who can grow into the higher-level positions, or even consider outsourcing some activities to third parties.

Benchmark pay

The other side of budgeting for new hires is making sure your current employees are earning a reasonable market rate. Employees who are perpetually underpaid will likely be looking for new work.

Increasing their pay to the market level, or at least close to it, is probably less expensive than attempting to hire and train a new person.

There are plenty of websites that allow you to find the market wage in your area for each position. You may also be able to reach out to a friendly headhunter to learn what they’re seeing.

If you discover that a key employee is overpaid and they are irreplaceable, it’s not the end of the world, since being irreplaceable is worth a lot of money.

However, if you find out an easily replaceable employee is overpaid, think about ways to increase their responsibilities and keep these numbers in mind when hiring future employees.

Measure attendance

One of the first grown-up jobs I had was at a factory as the spreadsheet jockey (not my official title). I worked on making spreadsheets that the management team didn’t have the computer expertise to do.

The main project I completed was designing an attendance tracker. The company had just finalized a new union contract that spelled out the exact procedure for tardiness and unexcused absences.

After one incident, there was a verbal warning; after two, a written warning; three, a meeting with the manager; and finally, dismissal after a fourth.

Up until that point, supervisors had just tried to remember which employees were late or absent and when. As you’ve probably guessed, this was not an effective strategy.

The supervisors generally worked 60 hours per week and had endless fires to fight that pushed tardiness out of their minds. Even when supervisors knew an employee should be dismissed, there was never documentation that the proper steps had been taken with each incident.

The spreadsheet was simple to make and it went a long way toward making sure employees got to work on time and basically ended unexcused absences. Employees who had long been able to get away with strolling in whenever they wanted stopped after their first verbal warning.

Seemingly small things, like tardiness, can get swept under the rug when you’re running a business because it’s so hard to keep track of everything at once. Payroll analytics will help you monitor the small things without having to expend a lot of extra effort.

Bonus: Use analytics to calculate incentive pay

We’ve talked a lot about data analytics in payroll. Now let’s discuss how other business analytics can affect how your payroll works.

Incentive pay has been around for centuries. The original Charles Schwab used it to great effect to make Andrew Carnegie a Robber Baron and then to run the shipyards for the U.S. Navy in World War I.

Not all positions would necessarily benefit from adding incentive pay, but most would. Here are a few principles to follow when setting incentive pay:

• Don’t be cheap: I once went whole hog for an entire shift, when I worked at KFC, to set the record for drive-through time. I was then awarded a nice, shiny $10 gift card to a grocery store. Not worth it.
• Use an easy-to-understand metric: Agree with your employees ahead of time on what the payroll metrics will be. If it is a salesperson, use a commission per sale. If it is a supervisor, use production per shift. If it’s a department head, use profitability. If it is yourself and other owners, use return on capital for the business. It is imperative that the metric is easily understood by the employee who is being paid.
• Control for employees with bad intentions: Make sure the metrics are not easily manipulated. You don’t want a salesperson to drop prices below profitability levels just to make the sale or supervisors who will sacrifice long-term business to hit production goals with a shoddy product.

What’s payroll WAR?

Wins above replacement (WAR) is one of the most popular analytics metrics in sports. The aforementioned analysts go through a bunch of statistical formulas to determine which stats best predict wins, and then drill down to the individual contributions of each player.

In this scenario, an average player is considered "replacement level," so wins above replacement shows how far above average your favorite player is.

Now you can merge data management, payroll, and analytics programs and calculate which of your employees has a high WAR based on the productivity of their department and their attendance record, and pay them accordingly.

At the end of last season, the last teams standing were among those that used analytics the most, including the Super Bowl-winning Kansas City Chiefs. Your industry probably doesn’t have a Super Bowl, but you can use analytics to devastate the competition in the same way.