One customer-facing marketing company wants to understand shoppers to their core. 

That’s why Bluecore acquired virtual shopping assistant company Alby, which uses generative artificial intelligence to answer consumer questions, both in real time and preemptively. 

Max Bennett, CEO and co-founder of Alby, also co-founded Bluecore over 10 years ago. He eventually left to pursue building the generative AI-powered service, which will now bring him back to his Bluecore roots. 

Fayez Mohamood, CEO and co-founder of Bluecore, said he’s excited that Bluecore’s path has crossed with Alby’s—and that acquiring such a powerful tool just made sense for the customer marketing technology platform. 

“We believe [Alby] represented the future of how shoppers would engage [with] websites, mobile apps—it’s akin to an in-store associate experience,” Mohamood said. “Our visions converged and when the vision converged, it became a matter of, how quickly can we make this happen?” 

Bluecore and Alby began discussing a potential acquisition in June; that deal has now closed, with Bluecore fully acquiring the company. Mohamood declined to disclose the financial details of the transaction. 

Bennett said that Bluecore’s ability to understand the consumer from a variety of angles will enable Alby to provide a more personalized experience. 

Some retailers and brands have begun to implement chatbots into their websites or apps, in an effort to help consumers search for items with natural language queries. Mohamood knows the thirst for chatbot integration is at an all-time high among his clients. He said about three-quarters of his conversations today center around implementing generative AI into companies’ technology abilities. 

“The most [common thing] we hear about is a chatbot. Everyone’s familiar with ChatGPT, so everyone runs to, ‘I’m going to have a chat [function] on my website,’ which we believe is not the optimal experience,” he said. 

For those looking to integrate generative AI in an innovative way, Alby could be a new way forward. Rather than requiring users to input long-winded natural language queries, the technology has the ability to predict what kind of questions a consumer might have. In the case of apparel items, that could be anything from “What is this sweater made of?” to “What is the warranty on this jacket?”

“People don’t shop the way they use ChatGPT. When you go to ChatGPT, you have a long-form query in your head that you’re trying to get help on, some complex task you’re trying to accomplish. That’s not how people shop. People shop in between things. People shop on the go. People shop when they’re on the train, scrolling. The needs for the human being engaging in that activity are very different,” Bennett said. 

The model works in a similar way to Amazon’s Rufus chatbot, which aids shoppers with product discovery and recommendations by offering up prompts when consumers click into it, based on their shopping and viewing history. 

Bennett said Rufus will continue to help shape consumers’ expectations of their experience with e-commerce, which will only bolster Alby’s importance—particularly because Amazon isn’t currently licensing or selling Rufus technology to other retailers or brands. 

“We see [Rufus] as this amazing process of consumers being trained to expect a new type of experience, which is now going to catalyze all these other brands and retailers who have their own platforms and want to go direct to consumer to do something similar,” Bennett explained. 

Bennett said, in testing, 95 percent of consumers using Alby for the first time clicked on an already proposed question; the other 5 percent entered their own, longer-form query. 

Alby runs on both a proprietary machine learning (ML) model and an open-source model. 

The ML model, which is trained on data like ratings and reviews, community Q&A, product attributes and companies’ site data, helps determine likelihood of abandonment, conversion and more. 

The open-source model then handles the dialogue with the consumer, and clients can select from any major model on the market today; Alby can be hot swapped into various large language models (LLMs). Bennett said, today, OpenAI’s LLM, ChatGPT, is the most popular among Alby’s clients, but the system just as easily pairs with other options, including Google’s LLM, Gemini, or Meta’s LLM, Llama. 

The two companies expect to pair Bluecore’s existing recommender system, which helps companies to understand “if, then” patterns about a consumer—that is to say, if a consumer has an interest in one product, then they may also be interested in another—into Alby. That will help with product pairing, discovery and starting conversations with consumers. 

They also expect to add variability to how consumers access Alby. Mohamood and Bennett’s teams have already begun working to integrate Alby into emails and text messages, which they hope will engage consumers in a new way. 

“Today, you get an email with a bunch of product recommendations, and what we’re imagining is, can we make email two-way by enabling people to ask questions?” Bennett explained. 

For instance, a consumer might receive an email about a product they abandoned in their cart, with a list of frequently asked questions to re-engage them. If a consumer clicks into one of the questions, they will be directed to speak with Alby on the brand or retailer’s site. In the case of SMS, consumers will be able to text back and forth with Alby directly. Mohamood said the email capability is likely to be launched in February, with the SMS capabilities following closely behind. 

Mohamood said he expects Alby and other emerging technologies will enhance the consumer experience by making mass marketing less common. Instead, he said, Bluecore’s segmentation combined with Alby’s dynamic messaging could make consumers feel heard, even in the absence of an in-store experience. 

In the future, consumers’ direct engagement with Alby could also improve brands and retailers’ assortment and design decisions. Already, some AI-powered technology providers help companies forecast demand and external considerations that could impact allocation. But by understanding the types of information consumers seek when they shop for a product, companies may be able to solve issues they never knew consumers had, before items ever hit the shelf or the site. 

“Right now, there’s very little insight into, what are the shoppers on your website looking for? In terms of your product, what features are exciting? What do they think they’re missing?” Mohamood said. “Imagine if you could provide all of that insight to teams…buying products or designing products. That is a huge untapped opportunity for marketers and merchandisers to work together.”