The AI-Powered E-Commerce Shopping Experience

This is the first in a series of short blog posts that will explore the effects that AI is having on the e-commerce shopping experience and what is coming down the road (and soon!).

Let’s start with the advantages of Conversational Shopping. The traditional process of product discovery online includes 2 basic choices – type in a specific short text string (like “summer dress”) or navigate through a series of menu choices (Women’s -> Clothing -> Dresses -> Summer, etc) to get to a page usually ordered in pre-determined vendor hierarchy. Either process requires additional “hunting” to try to find a the closest match to what the client is looking for, often leading to frustration and churn (#1 reason?).

Time-to-Discovery = ? (1min? 2min? 3min?)

Now let’s look at conversational search using natural language processing and AI. The shopper asks a question, posed in the shopping assistant’s chat context window. Here’s an alternate example to the above search:

“I’m after a floral summer dress made of lightweight materials like cotton or linen. It should be brightly colored with a flowy skirt. I’d prefer it to be midi length and have short sleeves. Under $60 please.”

In this scenario, the AI search engine has several key advantages over the above. It has modeled the site’s product catalog, creating “vectors” for each product. It has also augmented & enhanced each product’s data with new and relevant data based on AI-enrichment techniques that extrapolate missing information in the limited amount of data available in the product listing. The agent also has external data added to the catalog’s information package relevant to the current market pulse for the company’s vertical market. This is a process called retrieval augmented generation (RAG).

When the request is submitted, the backend engine rapidly digests and vectorizes the query and quickly delivers nearest vector matches from the catalog. These recommendations, subject of course to the size and target of the site’s catalog, are going to be much better and more targeted to the client’s query than the traditional two search strategies above. Even with smaller catalogs, 1 very close match is better than the hunt & peck process. Revenue goes up, as does consumer satisfaction. Churn goes down.

Time-to-Discovery = ~6 seconds.

Here’s the questions to ask yourself:

Why does chatGPT have ~180M users in less than 2 years?

There are ~24M e-commerce sites worldwide. Which ones would not benefit from an AI-based shopping assistant?

Next up – The Benefits of Adding Multimodal Search to a Conversational Shopping Concierge.

visualAI retail solutions delivers AI-based e-commerce products that fundamentally change the shopping experience. shopperGPT is the platform’s hyper-personalized, multimodal conversational shopping agent.