In this article we focus on how AI and enriched product data are transforming fashion eCommerce, using the case study of Dressipi and House of Bruar
Autor: Enrico Fantaguzzi
Współzałożyciel i dyrektor
Cyfrowa Akademia Mody
Od butiku do algorytmu: jak sztuczna inteligencja zmienia rolę asystenta sprzedaży mody
In the age of AI, fashion eCommerce is finally catching up with the personalized attention shoppers once received in physical stores. At the heart of this transformation is Dressipi, now part of the MAP Digital, an AI-powered product data enrichment platform that helps fashion brands deliver personalised shopping experiences at scale.
Speaking during a Digital Fashion Academy session, Dressipi co-founder Sarah McVittie described how the platform was inspired by the intuitive skill of a brilliant shop assistant. “I once walked into a boutique in New York and a stylist picked out five outfits for me—all perfect. I bought them all. That kind of intuition is what we’ve worked to replicate using AI.”
I once walked into a boutique in New York and a stylist picked out five outfits for me—all perfect. I bought them all. That kind of intuition is what we’ve worked to replicate using AI.
Sarah McVittie
Decoding Style with Data
Initially a B2C project, Dressipi shifted to B2B after major UK retailers expressed interest in licensing its taxonomy system. Today, Dressipi enhances product feeds with a deep layer of styling, contextual, and dynamic data—including neckline types, fabric flows, and trend alignment—based on thousands of attributes.
This data doesn’t just optimize websites for conversion. It helps retailers understand what drives customer preferences, return rates, and even emotional resonance. For example, women with larger busts are more likely to return a crew-neck top than a scoop neck, these details impact algorithmic product recommendations.
Real-World Impact: The House of Bruar
John Hodge from The House of Bruar, a Scottish luxury retailer, shared how integrating Dressipi’s technology improved their recommendations. “We tested Dressipi against our existing engine and saw a 5% uplift in incremental value,” he said. “Outfits now reflect our brand DNA and are personalized to the customer’s preferences and context.”
Their team also benefited operationally—reducing manual effort for product tagging and visual merchandising while improving consistency across imagery and attributes.
“Outfits now reflect our brand DNA and are personalised to the customer’s preferences and context.”
John Hodge
Returns, SEO, and the Future of Discovery
Beyond styling, Dressipi tackles a critical challenge in fashion eCommerce: returns. By identifying patterns, such as whether returns are due to size, style, or stock fragmentation—the platform optimises for kept items, not just purchases.
Moreover, as search evolves to prioritize natural language (think “What should I wear to a summer wedding in Edinburgh?”), enriched product data becomes crucial. Dressipi supports brands in creating semantic layers for SEO, marketplaces, and even AI chat interfaces, bridging brand language with customer language.
The evolution of Customer Journey with AI
Drawing on the discussion with Sarah and John, the introduction of AI like Dressipi is significantly changing the podróż klienta in online fashion retail, aiming to replicate and enhance the personalised experience previously only available in physical stores.
Here are some key ways the customer journey is evolving:
Natural language search
Shift to Natural Language Interaction: The way customers search is changing from traditional keyword-based queries (like “black dress wedding”) to more natural language phrases (such as “looking for a pencil skirt for work that’s smart but a bit glamorous”). AI helps retailers optimise their product data and metadata to be discoverable through these natural language searches.
Highly personalised recommendations
Generic recommendations like “customers like you bought X” are ineffective in fashion due to variations in size, shape, height, weight, gender, and changing preferences. AI uses detailed product data (physical attributes, context, dynamic trends) and understands how specific features look or feel good on different people. This enables the system to create tailored outfit and product suggestions based on the customer’s taste, lifestyle, colour palette, and even physical characteristics, showing them their “best possible shop window”. The system learns from customer interactions over time.
Reduce ecommerce returns
Improved Fit and Reduced Returns: Understanding factors that contribute to returns, such as sizing issues (15-20% of returns) and style suitability (30-50% of returns), is crucial11. By incorporating fit data and understanding how attributes (like necklines) impact suitability for different body types, AI can help predict return likelihood2…. Algorithms can be optimised for “kept items” rather than just conversion rate, meaning customers are shown products they are more likely to keep12. This leads to a more positive post-purchase experience and reduces the frustration associated with returns.
Increase engagement
Style Guidance and Outfit Inspiration: Customers often need guidance on how to put items together. AI creates on-brand outfit suggestions that help customers visualise how to wear products for different occasions (e.g., work, evening, weekend). This helps customers see the versatility of items and encourages investment. The system understands what occasions are applicable for a specific product and suggests items that go well together8.
Consistent and Relevant Browsing: AI ensures that recommendations and product listings use consistent imagery, avoiding a jumbled look1314. Crucially, it takes into account current stock levels, including specific sizes, so customers are only shown products that are actually available to them. The rich, consistent data also allows for more effective and consistent filters and facets on the website, improving the browsing experience.
Marketing analysis
Deeper Customer Understanding and Segmentation: The detailed data generated by AI allows retailers to move beyond basic demographics and create powerful customer segments based on behaviour, preferences, and lifestyle (e.g., equestrian customer, pro farmer). This understanding enables the delivery of experiences and language tailored to these specific segments, making the customer feel understood and increasing engagement.
Integrated Brand Experience: The AI works closely with the brand’s creative team to absorb and reflect the brand’s DNA and styling principles. This ensures that the personalised recommendations and guidance remain true to the brand’s identity and help build resonant brand experiences online. Tailored landing pages can also be created based on customer intent and context.
Final Takeaway
Fashion retailers are realising that data is not just about optimization—it’s about storytelling, styling, and service. As AI reshapes the fashion landscape, tools like Dressipi are making it possible to scale the art of the sales assistant and build genuinely personal digital shopping experiences.
In essence, AI like Dressipi moves the online customer journey from a potentially overwhelming, static catalogue experience to a dynamic, personalised, and guided interaction that better reflects the nuances of fashion and aims to build a stronger connection with the customer