Tech Tactics is Sourcing Journal’s series with brands and technology companies to discuss their latest innovations. Here, Nate Barad, vice president of product marketing at Algolia, discusses how the end-to-end artificial intelligence (AI) search solution helps fashion brands make more informed decisions and enhance customer experience with its new Intelligent Fashion Solution tool.
Nate Barad, VP of product marketing, Algolia
Name: Nate Barad
Title: Vice President of product marketing
Company: Algolia
SOURCING JOURNAL: You recently launched the Intelligent Fashion Solutions tool, which tailors search results by factoring in elements like weather, trends and price. What inspired this, and how does it meet the evolving needs of fashion retailers?
Nate Barad: We have extensive experience working with fashion brands, partnering with 75 percent of the world’s top fashion houses, including Everlane and Gymshark. Additionally, with 12 years of Black Friday expertise and a customer base of roughly 17,000, Algolia has a deep understanding of what drives the retail industry. We have clear insights into what makes the best shopping experience, which today revolves around convenience and solution-driven shopping.
For instance, a consumer might search for “a dress for a wedding in Venice in July,” and retailers need to consider current styles, weather and price points when delivering results. At the same time, they must balance consumer desires with business goals like driving urgency, promoting celebrity collaborations and managing inventory across different locations.
Our Intelligent Fashion Solution addresses these challenges, setting a new standard for online fashion retail. It enables a seamless, personalized shopping experience by providing intent-driven results, keeping retailers in tune with fashion trends through powerful indexing and interpreting natural language to improve search accuracy.
You mentioned personalization, which is increasingly important to consumers. Could you dive deeper into how Algolia’s Intelligent Fashion Solution improves the accuracy of product recommendations and search results compared to traditional e-commerce platforms?
N.B.: If we take the wedding guest example even further, this new solution understands the variety of customer inputs so retailers can deliver dress options whether a customer uploads a photo of the bridesmaid dresses, uses “voice” to ask for a “dress that looks like a sunset” or types out “a dress for a wedding in Venice in July.” By embracing the different inputs, we improve search accuracy for fashion e-commerce.
AI also supports accuracy and scale. Fashion catalogues are big, and AI can help sort through 15,000 products in the blink of an eye. As more people search, the AI becomes more intelligent and can understand more signals of intense desire preference. There is a constant learning loop that only improves search and discovery as more people engage with a retailer.
How does this solution alleviate issues like inaccurate product data or inefficient catalog curation?
N.B.: Intelligent Fashion Solution can help retailers create buying urgency on specific campaigns or initiatives. If, for example, a fashion house makes a significant investment to collaborate with a celebrity, they’re likely going to have to demonstrate return on investment (ROI) from that engagement. The same can be said for a discount or sale event. Most other search products out there stop short at how to get someone to act—like buying—without understanding the “why” or what comes next.
Algolia’s solution also creates a beneficial relationship between the digital and in-store teams. This generates digital awareness for in-store opportunities like offering in-store fulfillment for the person who shops online but wants a product immediately.
How can fashion brands maintain creativity and authenticity while relying on AI-powered tools for a more data-driven approach?
N.B.: Fashion brands should combine their creativity and authenticity with AI-powered insights. What I mean by that is, they’re still the fashion experts bringing unique products to life.
However, if AI indicates that handbag A is the best-selling item in the line, maybe move it to the top of the page instead of listing it as the 50th product. Or, if AI suggests that handbag B should be in between an image of handbags D and C to sell more, consider making that adjustment to the site.
Given Algolia’s experience supporting thousands of customers, what are some of the most significant trends you’ve seen in consumer behavior, and how are you adapting your technology to meet these evolving needs?
N.B.: We discussed some of this earlier, but modality and input methods are changing. Consumers want and need different experiences if they’re searching on their phones from a train versus searching on a desktop from their house in the suburbs. We also see more affinity for camera and image search. Consumers are uploading images they find on social media or other channels, and they want search results to match.
Finally, more consumers are using GenAI technologies to generate or create answers to their retail questions, in addition to retrieving results.
As AI-driven solutions become more integrated into fashion e-commerce, how do you see the role of personalization evolving in 2025?
N.B.: Hyper-personalization is heading toward predictive personalization, which anticipates customer needs or desires before they come up. GenAI is also fueling dynamic content to deliver hyper-customized product pages and searches based on customer profiles, as well as product descriptions and personalized content like shopping guides.
Location-aware recommendations and inventory integration are also maturing. Consumers want accurate and real-time product options based on speed and convenience. This entails having insight into their need and geographic location.
Privacy and data are at the core of hyper-personalization, so doubling down on zero-party data and consumer feedback loops are a focus for us in 2025 and beyond. We want to give customers the options they want as transparently as we can be.
