Adobe unveils new ‘Fast Filtered’ AI tool for retailers’ product searches

This personalization feature is currently being mock tested for furniture products and athleisure fashion gear. According to Adobe, these categories appeal to a wide audience while showcasing all the new capabilities in one place. “But the idea is not limited to furniture,” said Oribio. The tech preview at the Adobe Summit in Las Vegas on March 22 is to drum up initial interest from customers who would like to beta test this product.

While Adobe did not share a timeline for this tool to come to market, typically the company’s upcoming products take roughly 12 months to come to market from the time they are previewed by merchants. The company said it will also share updates on generative AI tools like ChatGPT at the Adobe summit.

This is the latest move by the software giant to stake its claim in the commerce space. Adobe owns the enterprise e-commerce platform Adobe Commerce — previously called Magento — which works with retailers like Dick’s Sporting Goods and Home Depot to name a few. By introducing another paid commercial tool for merchants to invest in, this new project presents another way for the company to work directly with retailers and further its commerce ambitions.

Ant Duffin, senior director analyst for digital commerce at Gartner’s marketing practice said the concept Adobe is looking to develop “feels like it is close to what we termed as guided selling or quiz commerce. It’s essentially trying to humanize your product attributes, your product content to create a natural engagement with a machine like you will with say a store assistant,” Duffin explained.

Duffin added that Adobe has made a very significant pivot into the online shopping space, after acquiring Magento in 2018. “They do have a significant focus on personalization to drive personalized digital commerce experiences that then dovetails into the Adobe commerce product,” he said.

Adobe said the upcoming software tool has been under research and testing at Adobe Labs for over a year and will extend across all product categories and all retailers with a level of fine-tuning that doesn’t quite exist in the market yet.

“Our motto is we’re trying to make the online shopping experience feel like you’re with a very good salesperson. Like when you go to the store and you deal with a very experienced professional salesperson. You leave that store with a purchase that you feel good about, and that’s what we’re trying to bring online,” said Oribio.

Using Adobe’s Sensei AI technology, this new project analyzes any Adobe Commerce storefront category to identify both common and unique product details, ranging from measurements and colors to materials and key features. Shoppers can then apply their choice of filters to surface useful results – including specific custom dimensions – and reveal differences between products for quick comparison.

Once a merchant has entered their product descriptions and associated metadata into the system, Adobe’s Fast Filtered model will be able to classify items, find variances between product categories and construct filters.

Oribio said internal discussions with industry stakeholders revealed that, “there’s still a lot of frustration when it comes to online shopping. Especially the more people do it, the more those frustrations come to surface. So there’s still a lot of opportunity to innovate and make shopping online comfortable from home,” he added.

“When you have limitless options, then we really need to find a way as merchants to make it easy for our customers to find something they’ll be happy to have purchased but also quickly,” said Oribio.

A team of four executives at Adobe that work in different areas of the company gathered together to address “some gaps in the commerce experience, not only because we work in commerce, but because we are online shoppers ourselves,” added Oribio.

The idea is to find a “different way to navigate the products essentially to understand what the customer is looking for and drive them through to their recommended product based on what they’re actually looking for” added Duffin.

Oribio said the biggest red flag people have is “what are we doing with this data? Because if you’re logged in then obviously we can track everything that you’re doing in the store. So it is our responsibility to make sure that we build tools to respect people’s privacy.”

Duffin said if you look at the “intersection between personalization engines and digital commerce platforms like Adobe Commerce, what you’re typically finding is they are all looking to leverage AI and ML as a way to help drive those personalized customer experiences and to help to drive that point of conversion.”

While marketers that buy ads on e-commerce websites are becoming more willing to take their hands off the wheel and let the machines make the decisions, this new technology has yet to be fully harnessed by other parts of the retail business. “There is still an element of distrust of the black box that is AI and ML,” said Duffin that Adobe will have to contend with.