Having operated at the center of retail supply chains for decades, SPS Commerce Inc. is applying artificial intelligence to data compiled on trading partner transactions to generate demand forecasts, persona-based marketing, and other supply chain enhancements.

The company, which reported $536.9 million in 2023 revenue, up 19% from 2022, has 120,000 customers across 85 countries. Companies using the SPS retail network include W.W. Grainger Inc., The Home Depot Inc. and Target Corp.

Jason Popillion, director of technology, SPS Commerce Inc.

“SPS sits in the middle of the transactions that happen in retail — including basic transactions like purchase orders, ship notices, invoices — all those things that are key to allowing commerce to happen,” says Jason Popillion, director of technology, adding, “We also have things like inventory files in ancillary documents that show what inventory is available on hand at any moment.”

“What we’re able to do with all that data — which is really super interesting about AI — is dissect it and get analysis done in ways that we haven’t been able to do before.”

He adds: “Transactional data tell many stories about the life and cycle of our economy, customer sentiment and buying habits … what products seem to be stronger in times of the year, or how you might look for things in the coming future.”

Popillion says SPS defined its data structure to make it more understandable by AI. “We create a schema that says, ‘this is the data that we’re going to be working with; these are all the pieces that we have available.’”

As a result, “AI did a very, very, very good job of understanding the data” and produced useful machine learning evaluations related to supply chain operations, he adds.

For the last several years, SPS Commerce has been experimenting with AI. “Early tests have primarily focused on internal projects with the goal of increasing efficiency of the work we do in service of our customers along with validating effectiveness of potential market-facing solutions,” Popillion says. “In every test we run our goal is the same — find ways to help our customers run the most efficient supply chain possible.”

So far, the company says it has found AI produces levels of efficiency, including speed and accuracy, ten times higher than non-AI methods for such efforts as making demand forecasts.

For example, SPS used AI to predict order volumes for a future time period based on past order volumes over an extended time period.

“If you’re trying to think about how to plan for the upcoming holiday season, what we’re doing is looking at the last two years and see what the data is showing us, to give you an idea of what you can expect in this upcoming season,” Popillion says.

Other methods SPS is using with AI include:

• Faster onboarding of supplier product data. Using an in-house developed AI chat client tool, SPS is enabling its internal supply chain professionals to help expedite commerce for retailers by more quickly and accurately compiling retailers’ requirements for onboarding merchants’ new suppliers into the SPS retail network. “When you’re trying to get connectivity to your supplier base that’s feeding your commerce engine, the sooner you can get them on board, the sooner you can acknowledge the revenues from those suppliers,” Popillion says.

• Customer personas. To help retailers tailor marketing communications, SPS uses AI to customize marketing language to the needs of managers in particular roles in companies, such as director of sales, or who have communication preferences based on their U.S. location, such as the East Coast or in the South or West.

Popillion, who is a Certified Information Systems Security Professional, says SPS has addressed security concerns many people have regarding AI by working with established AI technology providers Microsoft Corp. and Amazon Web Services and by setting data management standards.

“It’s all about data,” he says. “It’s about training individuals so they know how to use it effectively, and it’s about creating the right access policies to allow only [authorized] people to have access to it.”

He adds that he’s noticed a shift in thinking among companies concerning AI. As with the early days of the internet, when companies realized they could use the web to grow their businesses and make them more efficient, organizations realize AI offers practical ways to improve business operations — for example, possibly generating a demand forecast in minutes instead of hours.

For now, SPS is using its AI technology internally to provide value-added services to customers as it continues to develop it, Popillion says. But he adds that SPS also expects to introduce as early as later this year AI applications that its customers will directly access and use.

Paul Demery is a Digital Commerce 360 contributing editor covering B2B digital commerce technology and strategy. [email protected].

Sign up

Sign up for a complimentary subscription to Digital Commerce 360 B2B News, published 4x/week. It covers technology and business trends in the growing B2B ecommerce industry. Contact Mark Brohan, senior vice president of B2B and Market Research, at [email protected]. Follow him on Twitter @markbrohan. Follow us on LinkedIn, Twitter, Facebook and YouTube.