You may be among the millions of Americans who just purchased an item during the two-day Amazon Prime deals event. Did AI help in the process of convincing you to spend?
Amazon said Prime members bought more than 150 million items from third-party sellers. It didn’t release much more data on the big retail event, and among the things we can’t know for sure is how much generative AI programs may have helped sellers do an even better job of pitching their products than in previous years.
We do know for sure that getting a leg up on the competition is getting easier for e-commerce platform sellers through the latest AI.
Generative AI tools — offered by e-commerce platforms, marketplaces and private companies — can help with some of the more labor-intensive, time-consuming and mundane tasks that sellers tend to hate. The goal of using these tools is to drive more sales with less effort — and angst — on the part of sellers.
AI can be used for many things, from writing impactful product listings to data analytics, but more of the focus of late has been on the product listing side. Amazon, for example, recently rolled out a generative AI tool to help sellers write more robust and effective product descriptions.
A New York Times’ tech reviewer who recently tried out the latest version of OpenAI’s ChatGPT which can “see, hear and speak,” said it did a very good job of writing product listings for items he wanted to sell on Meta’s Facebook Marketplace.
These tools can “spit out the perfect product listing for you that is optimized to your customer base,” says Chris Jones, chief executive and co-founder of AMNI, an AI-powered platform that streamlines procurement, manufacturing and distribution.
It’s obviously early days in the use of AI for e-commerce, and there will be some big hits and misses — as well as risks for any seller than blindly relies on AI. Here’s what sellers need to know about using AI to sell more effectively.
Business owners shouldn’t feel the need to be writers
Creating high-quality e-commerce content often doesn’t come naturally to sellers. There’s a need to create compelling product titles, bullet points and descriptions, which can be time-consuming and frustrating for sellers who don’t have a natural writing ability or the time to devote to these efforts. It can be daunting for sellers to sit in front of a blank screen and figure out what to write.
Beyond just describing they product, they need to create one that’s also well-optimized for Amazon search algorithms so it gets good exposure, said Greg Mercer, chief executive and founder of Jungle Scout, a platform that helps sellers start and scale their e-commerce business.
AI can reduce — to seconds or minutes — these mundane listing tasks that might have taken some sellers three-to-five hours to complete, Mercer said.
Amazon says it will save sellers time and effort
Sellers on Amazon’s competitive third-party marketplace need to provide a brief description of their product in order to allow its new AI tool to generate high-quality content for them to review. For example, they can plug in the item name and whether the product has variations and a brand name. Amazon’s models learn to infer product information from various sources. For example, they can infer a table is round if specifications list a diameter. The models can also infer the collar style of a shirt from its image, the company noted in a blog post about the AI launch.
“In addition to saving sellers time, a more thorough product description also helps improve the shopping experience. Customers will find more complete product information, as the new technology will help sellers provide richer information with less effort,” the company stated.
At eBay, an image is often the starting point
eBay is also working on tools to auto-generate item descriptions, and a revamped image-based “magical” listing tool that leverages AI.
The tool allows sellers to take or upload a photo in the eBay app — only available in Apple’s iOS for the time being — and let AI do the work. From the starting point of a photo, the AI can write titles, descriptions and add important information, such as product-release date, detailed category and sub-category, according to a company blog post. It can also combine with eBay’s other technology to suggest a listing price and shipping cost, the company said.
This latest version, which includes upgrades to a previous iteration based on customer feedback, is being tested by employees. The company said in the blog post that it expects to release the revamped tool to the public in the coming months.
Shopify shows how AI works for much more than product listings
AI can be used to help sellers with much more than just product listings. “A lot of people don’t think of using AI to be their CFO and analyze data for them or to help do competitive research,” Mercer said. “AI is getting a lot more powerful than just writing product listings.”
As Harvard Business School AI guru Karim Lakhani recently said at the CNBC Small Business Playbook event that every small business owner should be using generative AI. “I think about ChatGPT as a thought partner, lowering the cost of cognition and new ideas,” Lakhani said.
Shopify, for example, announced its AI tool Shopify Magic this past summer. It’s a suite of AI-enabled features integrated across the Shopify platform, and specifically designed to enhance commerce, the company said. Merchants receive contextually-relevant support for a range of tasks related to store building, marketing, customer support and back-office management.
For example, merchants can create email campaigns using just a few keywords. They get persuasive subject lines, appealing content and recommended send times to achieve more effective click-through rates, according to a company video.
Shopify Magic also drafts custom replies for a business’s more common customer questions, allowing the merchant to review and edit the content. Those answers are then shared automatically with customers who ask questions, so a merchant doesn’t need to respond in real-time. Another feature creates blog posts for holidays, business milestones or campaign ideas — including the ability to customize the tone of voice and translate the content into different languages, according to a spokeswoman.
“AI creates an environment where an entrepreneur’s expertise, brand, and product can shine, and will help them take something from idea to reality much faster than previously possible,” said Miqdad Jaffer, the company’s director of product, in an email.
Platform-specific tools may get the best sales results
Some sellers use consumer-facing applications such as Microsoft-backed ChatGPT from OpenAI and Google’s rival chatbot Bard — both companies also offer business versions of the AI tools now — to help ease the burden of creating better product listings, but e-commerce professionals said platform-specific tools may be more effective, since they are tailored for that particular platform.
That said, using broadly available tools — whether it is for product listings, analysis of customer reviews, advertising campaigns or financial reporting — would still be better than doing it on your own, Mercer said.
“Anything that can help them not need [human] resources, but achieve the same speed and scale is going to be the name of the game,” says Margo Kahnrose, chief marketing officer of Skai, a provider of data, insights and marketing technology, CNBC reports.
