{"id":8644,"date":"2023-07-19T17:50:16","date_gmt":"2023-07-19T15:50:16","guid":{"rendered":"\/blog\/?p=8644"},"modified":"2025-04-24T13:48:53","modified_gmt":"2025-04-24T11:48:53","slug":"atlas-ai-the-new-frontier-in-ecommerce-product-discovery","status":"publish","type":"post","link":"\/blog\/atlas-ai-the-new-frontier-in-ecommerce-product-discovery\/","title":{"rendered":"Atlas AI: the new frontier in eCommerce product discovery"},"content":{"rendered":"

AI \u2013 no other promise has sparked human imagination more than the prospect of creating a new artificial form of intelligence. <\/span>And rightfully so, Large Language and other AI models have and will have a dramatic impact on a lot of important areas like science, education, and of course eCommerce. <\/span>So it’s no wonder that eCommerce technology vendors are using the label AI with the same frequency as Sundar Pichai at the last Google I\/O.<\/span><\/p>\n

\"\"<\/a><\/p>\n

Image Author: Maurizio Pesce<\/a>\u00a0 Source: flickr.com<\/a>\u00a0 License: Attribution 2.0 Generic<\/a>\u00a0<\/em><\/h6>\n


\nThe problem is that it\u2018s much easier to talk about AI than actually creating a powerful model that integrates into your solution in an impactful way. That\u2019s why we developed this blog post with a no bullish!t methodology – 100% factual and in collaboration with the smartest brains at FactFinder (our AI researchers and developers), to give you genuine insights into how our AI, Atlas, is different and why it can make a big impact on your eCommerce goals. <\/span>\u00a0<\/span><\/p>\n

But before we get into that, let\u2019s explore why other models have struggled to live up to their promises.<\/span><\/p>\n

Why product discovery is a hard problem for AI<\/h3>\n

Thinking from first principles, breaking down eCommerce to its most basic level, it\u2019s nothing more than matching demand and supply in a shop on the internet. Easy right? In a perfect world, consumers should specify what they want, and the shop would show the best possible fit for their needs. Unfortunately, it\u2019s not that simple. There are hundreds of problems to retrieve and rank the best possible products, which, according to The Harris Poll, cost online shops $300 billion a year alone in the US.<\/span><\/strong><\/span><\/p>\n

Current AI models can solve these problems only to some degree due to 3 main reasons:<\/strong><\/h5>\n
    \n
  1. Search queries are difficult to understand for a machine<\/span><\/b>
    \nModeling language is not an easy task in eCommerce. There are many reasons for this including typos, homonyms (e.g. fan has two meanings), words with the same basis (e.g. water and watermelon), shoppers using different words to describe the same item (e.g. mobile phone and smartphone) and so on.\u00a0 <\/span>\u00a0<\/span><\/li>\n
  2. Sparse interactional data is not suitable for real 1:1 personalization<\/span><\/b>
    \nMost customers only make a handful of purchases in a particular shop. In the context of thousands or millions of products, it\u2019s extremely difficult for an AI model to draw any conclusions from this to personalize the experience to the individual, due to a lack of interaction data. For example, if you bought a toaster in April and a pillow in October this doesn\u2019t provide the AI with enough data to accurately predict your next purchase.\u00a0<\/span>\u00a0<\/span><\/li>\n
  3. Personalization by segmentation doesn\u2019t work<\/span><\/b>
    \nThis lack of data (Problem 2) is why most personalization solutions are trying to create shopper segments that behave the same. The problem is that we live in an individualized world and demographic data is not the best predictor of preferences. As you can see in the image below King Charles and Ozzy Osbourne would fall in the same segment – and it\u2019s unlikely they buy the same stuff.<\/span><\/li>\n<\/ol>\n

    \"\"<\/a><\/span><\/p>\n

    Truly understanding your customers is hard. The upside is that companies that have the AI technology that can overcome these challenges and understand their shoppers needs, see significant improvements in their onsite conversion for every visitor entering their shop – yielding growth and higher profitability. So these are problems worth solving.<\/span>
    \n<\/i><\/p>\n

    “We ourselves are surprised at how quickly the switch to FactFinder has been reflected in sales figures, as our\u202fsearch-generated conversion rate has increased by 20 percent\u202fsince go-live.”<\/span> <\/i><\/span>– David B\u00fcschler, Online Marketing Manager Globus Baumarkt<\/em><\/span><\/p><\/blockquote>\n

    How Atlas AI is unique – an inverted approach to product discovery<\/h3>\n

    We have approached AI in a different way \u2013 we inverted the problem. While all other solutions focus on modeling language (understanding the query and matching it with the assortment). Our AI focuses on learning your products so that it gains a human-like understanding of your assortment. This means that the queries and interactions from shoppers can be processed in a much smarter way.<\/span><\/p>\n

    \u201cIt really works to tackle much of life by inversion where you just twist the thing around backwards and answer it that way. All kinds of problems that look so difficult, if you turn them around, they are quickly solved.\u201d<\/span><\/i> – Charlie Munger, Investor and Philanthropist<\/span><\/p><\/blockquote>\n

    We\u2019ve called this unique model Atlas AI, as it creates <\/span>contextual neural maps of your assortment. <\/span><\/p>\n

    The model is trained with your vectorized product data.<\/span> This allows us to extract all the features and attributes from a product using a series of embeddings. <\/span>This training data is used to create a large number of neural networks that act as a detailed representation of your assortment. <\/span><\/p>\n

    Atlas groups products in a similar context together and thereby creates contextual maps of your assortment. Distances between them on the map represent differences between the products in that specific dimension. <\/span><\/p>\n

    The neural maps it creates have several layers so that it is able to understand your products and their relationships in a very granular way. The first layer could be as simple as grouping the products within the same category (eg. Smartphones).<\/span><\/p>\n

    \"\"<\/a>
    \n<\/span><\/span><\/p>\n

    O<\/span>nce the first layer<\/span> is created <\/span>the model creat<\/span>e<\/span>s deeper layer<\/span>s<\/span> (<\/span>eg.<\/span> screensize)<\/span>, <\/span>followed by <\/span>another <\/span>(<\/span>eg.<\/span> price) and so on until it has a<\/span> contextual, <\/span>human-like understanding of your assortment<\/span> \u2013 how deep the model gets <\/span>depends on your specific catalog<\/span>.<\/span><\/span><\/p>\n

    \"\"<\/a>Now <\/span>when<\/span> shoppers<\/span> enter y<\/span>our shop<\/span> and search or click on <\/span>stuff,<\/span> they <\/span>are<\/span> not interacting with your <\/span>actual <\/span>products<\/span>,<\/span> they are<\/span> interacting with <\/span>your <\/span>Atlas<\/span> AI<\/span> model<\/span>.<\/span> And as <\/span>Atlas<\/span> understand<\/span>s your assortment in a <\/span>contextual<\/span>,<\/span> human-like fashion<\/span>,<\/span> it learns from just a single click<\/span>,<\/span> solving the sparse interaction data problem <\/span>mentioned<\/span> earlier<\/span>.<\/span> This is the basis for<\/span> 1:1<\/span> relevance<\/span> in an unprecedented way<\/span>.<\/span><\/span>\u00a0<\/span> <\/span><\/p>\n

    What are the benefits for online shops?<\/h3>\n

    Based on its contextual understanding of your assortment Atlas AI has 4 main benefits over other solutions: <\/span><\/p>\n

    1. It creates general relevance – automatic, fast and flexible<\/span><\/b>
    \nWith your assortment now deeply understood and mapped into detailed neural networks understanding human language and behavior gets a lot<\/span> easier. So once Atlas has learned how shoppers behave on the site it automatically creates general relevance in your search results.<\/span> The benefits:\u00a0<\/span><\/p>\n