In Praise of Electronically Monitoring Employees

You might think it’s legitimate and appropriate for management to make sure employees aren’t stealing from the business. Apparently, not everyone agrees.

A little while back, a Steve Lohr post at the New York Times described some research Lamar Pierce, Dan Snow, and I just finished up and published as an MIT Sloan working paper. We had the chance to study a question that business academics and managers alike have been interested in for a while: what happens to performance when companies suddenly get much better information about employee misbehavior? Our results are fascinating, as is the discussion around them.

The companies we studied were restaurant chains, and the misbehavior was theft: servers, bartenders and managers stealing money. This is a big problem, since restaurants are low-margin businesses where theft by the staff can mean the difference between profit and loss. By one estimate, $200 billion each year is lost in the industry due to stealing.

We got to observe what happened at 392 locations across five chains (all of them sit-down places like Applebee’s or Chili’s, although neither of these were part of the research) both before and after they started using Restaurant Guard, a new piece of theft-detection software from NCR. NCR supplied us with the data but did not support the research in any other way. The data covered almost two years and 39 of the 50 states.

Restaurant Guard generates weekly reports for the management of each location that identify, by employee, all instances of suspicious behavior (the service now also includes instantaneous alerts sent to managers’ mobile phones). ‘Suspicious’ here means ‘really suspicious’ — since the accusation of theft is a serious one, the reports only flag behaviors that that match common scam patterns and are difficult to explain any other way. Some of these are quite elaborate. I don’t want to compromise RG by giving away too many details of how it works; if you want a compendium of restaurant industry deviousness, check out the ‘scam bible’ How to Burn Down the House.

The software works by minutely analyzing every transaction that flows through the restaurants’ point of sale (POS) systems. Since staff have to use POS terminals to get anything and everything done — order food, print a check, process payment, and so on — the systems provides a highly detailed and nearly complete record of all activities. NCR combs through all this data and generates reports and alerts for all Restaurant Guard customers.

What happened once Restaurant Guard was in place? Theft clearly went down, but not by much. The average reduction across our locations was a bit less than $25 per week. The real shock was how much overall performance improved. Weekly revenue increased on average by $2975, or 7% of total revenue. Almost $1000 of this increase came from drinks, where margins are highest and theft is most common.

These are huge increases, enough to make a material difference in the overall profitability of the location. And they didn’t fade away over time; they instead persisted for at least several months after Restaurant Guard was adopted.

Why did overall performance improve so much if obvious theft went down so little? I’ll get to that in a minute. First, I want to highlight some comments made in response to Lohr’s post in the NYT. Many people felt that Restaurant Guard was intrusive, unfair, and/or menacing. Comments included:

“Big (restaurant) Brother is watching you.”

“I can’t believe how many commenters are OK with this practice. Just shows how far towards fascism the right wing has pulled us as a country when spying on employees is considered mainstream.”

Let’s clear this up: Restaurant Guard doesn’t engage in surveillance of employees’ personal electronic communications, or any other activity they might reasonably consider private. Instead, it monitors their on-the-job performance, which is exactly one of the things that managers are supposed to do.

It’s ludicrous to say that employees should be free from all monitoring, so are the commenters above just upset that Restaurant Guard does such a good job of it? Do they want people to have a little wiggle room left for theft? Let’s hope not. I’ve written a great deal about the tough economic times faced by working Americans these days, but giving them license to steal from their places of business is not part of any solution I’ll support.

Now, what about those large revenue increases? Where did they come from? Not from the fact that the thieves left (or got fired) and were replaced by honest people; staff turnover didn’t increase significantly once Restaurant Guard was in place. As far as we can tell, performance improved simply because people started doing their jobs better. To oversimplify a bit, once the bad actors saw that theft was closed off to them as an option they realized that their best way for them to take home more money was to hustle more, take better care of customers, and generally be a better restaurant employee. And I imagine that once some people started acting that way the rest of the staff joined in; good behavior is contagious, just as misbehavior is.

The strongest piece of evidence we have for this explanation is the fact that tip percentage went up significantly after Restaurant Guard was in place. It’s hard to see how this would happen if employees got surly or disgruntled about the increased monitoring. Instead, it looks like they started doing the right thing by their employers and their customers. Isn’t that a story we can all applaud?

This entry was posted in Leadership. Bookmark the permalink.

Comments are closed.