When shoppers walk past displays of produce at a grocery store, they typically don’t think they’re witnessing the result of a high-tech operation. But companies in the grocery business rely heavily on engineering and software development teams to attract customers, run operations and continue to push for innovation within the grocery industry.
Take, for example, the grocery chain Kroger, which owns 84.51, a data insights company that uses retail data to provide insights to the company about customer habits. Other companies, like Albertsons, operate tech divisions in developer hubs like San Francisco to attract tech workers who otherwise may not envision themselves in the industry.
Software development skills are needed in virtually all aspects of the business, said Christina Garcia, senior vice president of engineering at 84.51.
“We are working on science and engineering solutions for merchandising, supply chain, couponing, personalization, loyalty — there isn’t an area of the business that is not supported in some way, shape or form by an engineering team,” Garcia said.
What Is It Like to Work as a Developer in the Grocery Business?
Even something as common as the ability to have customers order items online and pickup at the store, known as shop fulfillment, requires engineering support. What used to be a manual process, with orders printed out and distributed to shop associates, has been accelerated through digitization efforts by reducing the amount of time it takes between when the order is received and when it is ready for pick up, said Reyner Bacallao, senior software engineering manager at Walmart’s technology organization, Walmart Global Tech.
Curious about what engineering roles are like in the industry? Here are some ways developers can have an impact working in the grocery business.
Data Engineering Shapes the Layout of Stores
Software development teams even influence the way items in grocery stores are laid out on the shelves and throughout the stores. They create software that helps other employees, like category managers, visualize and optimize the spacing of shelves.
“There’s a lot of solutions that don’t seem like they would involve engineering, but in order for these individuals to be successful at driving the results they’re looking for, they actually partner with an application,” Garcia said.
For example, teams in charge of merchandising use computer-aided design software to help them visualize the layout of the store and figure out product placement schemes. Data scientists use data insights to offer feedback about how best to optimize the space on shelves to maximize profits. Engineers also contribute to inventory management, creating tools that track purchases regardless of whether they happen in store or online, and making sure stores don’t run out of stock.
Data can also give stores insights into the types of products that should be displayed together, which directly affects the customer experience, Garcia said.
“Like you notice pastas close to the sauces — they’re creating that kind of seamlessness for customers to be able to have a better shopping experience.”
“Like you notice pastas close to the sauces — they’re creating that kind of seamlessness for customers to be able to have a better shopping experience,” she said.
When berries are in season, for instance, data insights may recommend that stores stock angel food cake close to those fruits. That may result from noticing that customers like to buy those items together or as part of a trial to see if placing those items close together will encourage sales of one of the products.
“We can see which products are bought most frequently together and provide that to the merchants,” Garcia said.
Getting Insights From Receipts Takes a Lot of Engineering
At 84.51, much of the work revolves around data. A lot of the effort goes into creating a data mart, or central data repository, with high-quality data, said David Scroggins, who works as a lead software developer there.
High-quality data isn’t just useful to internal teams of data scientists and engineers, but also to the consumer packaged goods companies whose products line the shelves of stores like Kroger. The data insights provided by 84.51 about the performance of different products and the purchase patterns of customers are helpful to consumer packaged goods companies, who can use the insights to inform how they conduct marketing campaigns in the future.
“All of our different solutions, at a very high level, are trying to enable those CPG companies to get insights into what’s happening with their products in Kroger stores, or to maybe create a campaign to do targeted advertising for a given product,” Scroggins said.
But in order to get there, engineers first have to clean the data and get it into a format that’s ready for consumption by data scientists. Scroggins’ team helps with that, taking care of repetitive tasks and problems data scientists may encounter before being able to glean insights for customers inside and outside the company.
“How do we know that data is inaccurate? How do we know what the person actually bought?”
Getting clean data can be a formidable challenge by itself. A lot of data is created during checkout on Kroger’s point-of-sale systems. All of that information, which is also provided to customers in the form of receipts, is saved by the company and fed back through the engineering teams to data scientists.
But even though some of the checkout process is automated, things can still go wrong and introduce faulty or incomplete data into the system. Stores are constantly receiving new inventory, and that can result in items being sold in stores that are not yet in the system. This can lead to inaccurate data during checkout, especially for customers using self-checkout machines. And those inaccuracies become challenges for data engineers down the road.
“How do we know that data is inaccurate? How do we know what the person actually bought?” Scroggins said. “It doesn’t seem like a lot, but a handful of mistakes over 100 purchases really starts to add up when you’re thinking in terms of hundreds of millions of items on a daily basis.”
Software Engineers Power Acquisitions and Dashboards
Software developers working for grocery companies use a lot of the same skills as developers in other industries. Bacallao, senior engineering manager at Walmart Global Tech, said engineers on his team have good front-end and back-end development skills, as well as the ability to manage data.
There are also teams of engineers responsible for handling cloud infrastructure work, which allows other software developers who don’t have the skills for infrastructure to focus on other aspects of their work. Experience with the Java Spring framework is helpful, as well as experience with integrating different systems and building microservices, Bacallao said.
System integration challenges also come up when companies acquire other grocery chains, Garcia said. Companies might discover a great digital asset from the acquisition and adapt it for the entire company, or simply work to integrate the internal systems so customers can get the same experiences and benefits across all stores, like the ability to use loyalty programs and coupons.
“These are 12-, 18-, 24-month journeys that really do fall on the shoulders of engineering,” Garcia said. “It’s just natural that you’re going to need work to marry two different brains. … If we were operating just as separate units, there wouldn’t be that value of being part of the Kroger brand.”
Front-end development skills are just as important, though. Bacallao’s team creates apps for operations engineers, who investigate and resolve technical problems that come up in stores and distribution centers. Those apps are built using a front-end React framework, and give operations managers ways to keep track of their work through user-friendly dashboards and logging.
Some engineering groups are also exploring how to take a more proactive approach to resolving problems. “There’s a lot of development in that area with machine learning and trying to identify the issues before the issue actually occurs,” Bacallao said.
There are many layers to the engineering work that happens in the industry. Many engineers are tasked with supporting other engineering and data teams, who in turn create tools and insights for employees working in the stores, distribution centers and directly with customers.
“The better you understand the sort of concrete action of making a purchase in a grocery store, and how that filters down — the better you understand that, the better you then become at your job.” Scroggins said. “What unites a lot of engineers, or at least the good ones, is this really deep desire to solve problems, and I think the problems you face in this space are some of the more interesting problems I’ve encountered. They’re really complex, really difficult problems — and for me, that’s exciting.”