RIS Exclusive: How The Home Depot Is Employing Prescriptive Analytics

Jamie Grill-Goodman
Editor in Chief
Jamie Grill-Goodman profile picture

Retail’s competitive environment requires businesses, especially large retail chains like The Home Depot, to vigilantly move into the future of every aspect of the enterprise. Asset protection is no exception.

Shrink, whether from waste, fraud, abuse, training, execution or other reasons, continues to cost retailers about 1.38% of sales. This rate that has remained steady over the last few years and has an estimated $50.6 billion impact on the retail industry, according to NRF’s “2019 National Retail Security Survey.”

For The Home Depot, which boasts 2,291 retail stores and had sales of $108.2 billion and earnings of $11.1 billionin fiscal 2018, shrink can have a large impact on not just the bottom line, but the customer experience. When SKUs are missing due to shrink, employees don’t know to restock those items, resulting in shoppers unable to find what they need on shelves and suffering a poor retail experience.

So when Scott Glenn, VP of Asset Protection, came on board a little over a year ago, and saw Home Depot had a homemade manual process to analyze shrink, he knew there were some great tools out there and began an “exhaustive” search for a better solution.

Glenn formerly held positions with Sears Holdings, Target, TJ Maxx, and Kohl’s, and brought his experience implementing a similar solution to Home Depot.

“Our top deliverables in asset protection are to drive value for our customers and shareholders by reducing shrink and mitigating total retail loss,” says Glenn. “We recognized the need for an enterprise system that would drive these results.”

Eventually the company selected Savanna Prescriptive Analytics to help prevent fraud by identifying opportunities within the retailer’s sales data, both online and in-store. The solution includes the Sales and Exception Based Reporting (EBR) module from Profitect, which is now part of Zebra Technologies Corporation. 

The Home Depot will be leveraging the prescriptive analytics solution to process data from more than 2,000 stores in the United States and Canada.

Glenn says Home Depot signed the deal three weeks ago and the solution provider already has 1900-plus stores worth of data for the retailer. He expects the solution will be fully loaded and running in the next 30 days.

“We’re in the test and learn integration phase at this point,” says Glenn.

He notes his team has been focused on several different things, but doing a very manual, time-consuming process.

“We think we’re going to see pretty significant productivity and efficiency returns that come from this implementation once we get fully uploaded.”

The solution will help Home Depot battle shrink by employing an artificial intelligence (AI) platform that takes in data, specifically transactions from any register as well as online, and has a machine learning capability to identify any behavior anomalies. Anomalies could be from fraud, wrong pricing, a training issue or many other possibilities. Once the solution identifies something, it sends out a guide with a descriptive insight and prescribes an action to take.  

The key to the prescriptive analytics solution is that it shows adoption. If, for example, after two days a training task comes back, the retailer will know the employee isn’t acting on it. Value is added when people at the edge take action.

By offering a prescriptive action, or smart tasks, the system offers workflow for accountability, meaning if the end receivers of the message don’t take the action, The Home Depot will know. This helps the retailer know the adoption of the system, which is the key to its success.

When asked about who will be working with the solution, Glenn says, “we are a crawl, walk, run environment here at The Home Depot.” The initial users of the solution will be the corporate investigations group, approximately 50 employees who will become the super users. As they get better with the technology Home Depot will start to deploy it to their field asset protection team, and in the longer term likely deploy it to the store level asset protection employees and store level operations partners.

“The Home Depot is a big, big machine,” says Glenn. “It is a change management process, some people’s jobs are going to change (they won’t be cut)…and anytime there’s a change that’s implemented at this scale there’s fear. So it’s a lot about communication, expectations, being very transparent with the folks that are engaged with this process.”

Glenn says they have been working through a change management process for more than six months as they prepared to turn a solution on.  The ramp up time has helped employees prepare.

He notes the retailer has weekly meetings now about all the issues that come up and the process has been “as good as we can expect.”

Glenn advisers other retailers looking for a similar solution to “do your homework, kick the tires on it.”

“I’ve implemented this in other companies and I certainly feel like this has been the best experience for us, the most comprehensive experience for us, and actually provides us the best overall product.”