Finding the Recipe for GenAI Success in Grocery E-Commerce
Imagine customers sending their entire shopping lists to your website or app — then seeing a screen where they can buy everything at once, with options they’ll most like highlighted for them. Imagine perpetually knowing the perfect inventory balance: skirting both surplus and stock-outs. And imagine easily delivering hyper-personalized marketing: enticing individual shoppers with the right items and offers at just the right time.
You’re probably not doing all this now — but thanks to generative AI (GenAI), it may be the wave of the future. Do you want to ride that wave?
It’s a question many grocery executives have been considering since GenAI exploded onto the scene last year. The technology is teeming with potential for grocery retailers: transforming everything from ecommerce search to inventory management to marketing communications and more. Some emerging applications will be winners; others may be duds.
[RELATED: 4 Reasons for Retailers to Embrace GenAI]
As they weigh use cases, especially customer-facing ones, grocery execs often consider their customer base, asking questions like: Are they early adopters? Are they tech-forward? Will our customers want to adopt this, or will they have qualms? Is it worth the investment?
GenAI Food for Thought
A recent survey from Deloitte sheds some light on shoppers’ mindsets: 53% say they don’t want AI-generated communications from their primary grocer, and just over two in 10 (22%) think their grocer would use GenAI responsibly in customer interactions. In short, there are objections to be overcome. So, is it time to get out of the kitchen?
Not really. Just as you can’t make a dish that pleases everyone, it’s unrealistic to expect all your customers, or even the majority, to immediately dig in. Instead, the better question to ask is: “Is this type of GenAI-based innovation some version of the future? Will it eventually be good enough that many, if not most, shoppers will adopt it?” If you think the answer is no, then stop there.
But if the answer is “yes” or even “probably” — because it makes shoppers’ lives easier, drives important internal efficiencies and/or delivers measurable value — then it’s likely worth starting to consider and experiment. You may draw just a small portion of your users at first, but that’s OK. It’s a part of the adoption curve, and it’s worth incenting and catering to these early adopters. They’ll be instrumental in evangelizing the new feature(s) to help you “cross the chasm” to more mainstream adoption.
The Early-Bird Special
History shows that the companies that are the first to make bets on the wave of the future are often those that survive and thrive. Look at Amazon and the e-commerce curve of adoption — not everyone, or even many people at all, wanted to shop online at the time. But Amazon had a good idea of what the future would look like and won big because of it. It wasn’t that everyone wanted to buy all their products online in 1994. It was that Amazon figured we’d eventually get there and wanted to get the hang of ecommerce early.
The question grocers should be asking themselves is if the same is likely to happen with shopping via GenAI. For example, is shopping via an intelligent AI assistant that “gets you” likely to eventually be a better experience than what people have access to now?
So, when it comes to investigating and implementing GenAI, the goal isn’t — and shouldn’t be — pocketing extra billions tomorrow. It’s about ensuring resilience and positioning yourself for the future. Remember when Netflix offered to sell itself to Blockbuster for $50 million — but Blockbuster “laughed them out of the room”? You want to stay on the crest of where the market is heading, rather than focusing only on today.
Placing Bets
There are lots of exciting and highly promising applications for GenAI in grocery, such as the ones mentioned earlier. Additional survey data from Deloitte shows many grocery execs moving full-speed ahead: eager about GenAI’s potential for managing supply chain logistics and inventory, and citing the most compelling “killer application” being “acting as a customer assistant.” This isn’t just something conceptual; Walmart and Amazon have their own proprietary solutions, and some vendors have begun to offer similar functionality to other grocery retailers.
With such AI-based assistants, shoppers can, as Gartner reports, “search using natural language instead of specific product names or categories.” This opens the door for potential use cases like:
- Recipe generation: Dynamically displaying recipes in response to shopper queries with in-stock, personalized ingredients. (e.g., if the recipe calls for milk, and the shopper tends to buy organic, then organic milk options are shown.) A U.S. grocer trialing this saw a nearly 4% increase in search conversions.
- Longer queries and dialogs: e.g., “Help me plan a barbecue for a party of 15 with two vegetarians.” “I’m on snack duty for my son’s soccer game. What are some popular and healthy snacks and drinks?”
- Innovative online merchandising: Interspersing higher-margin, strategic products among loss leaders like milk, eggs and rotisserie chicken, in response to shopper queries (e.g., “Fill my fridge with the essentials” or “What are things I should make sure I don’t forget to buy at the grocery store?”).
- Better personalization: Taking advantage of grocery’s unique opportunities to know your online customers, even when they’re first-time visitors. In other industries, a typical online cart may have one or two items, but in grocery, it’s often 30. This makes grocers able to “know” their shoppers quickly and allows GenAI to analyze shoppers’ preferences/tendencies as they shop (e.g., for organic, bulk or discount items) to serve up increasingly attractive results that same session.
These types of use cases just scratch the surface of how grocery retailers may apply GenAI. As you put your figurative eggs in the right baskets for you, keep in mind how you’ll stay fresh and future ready.
