How to enhance personalisation by optimising product feeds for trends

Today, fashion retailers face a rapidly changing landscape characterised by rising labour and material costs, persistent shipping delays, and shifting consumer expectations. As consumers increasingly seek ethical and sustainable products while balancing tighter budgets, optimising product feeds for timely and relevant recommendations becomes crucial. Rising costs in daily essentials may lead to more understated clothing purchases, and new sustainability guidelines are pushing for greater supply chain accountability. Despite these challenges, the industry has shown remarkable resilience. To truly thrive, companies must leverage personalisation to stay ahead of trends and meet customers where they are. This enhances the shopping experience, boosts conversions, brand loyalty, and engagement, and ensures retailers can navigate the complexities of the modern market effectively. Personalisation has been tried and tested and stands out as one of the best tools retailers have to leverage trends – whether seasonal or industry-shifting. It enables brands to serve decisive customers who know what they want and engage those who need assistance with their purchase. Tools like affinity allocation allow retailers to present relevant, hyper-targeted banners based on a visitor’s preferences, boosting conversions, brand loyalty, and engagement. For instance, personalised hero banners can showcase boho style with floral print dresses and wide-brim hats to one customer, and new arrivals in the Y2K style to another, aligning with their user affinity. However, effective personalisation requires an up-to-date foundation of product data to serve relevant content and timely recommendations. In the fast-paced world of social media, fashion trends evolve quickly, risking outdated or irrelevant product attributes. Now is the time for a spring cleaning of product data. Brands should identify and eliminate non-specific attributes and create new, actionable ones that align with shifting consumer preferences. Users and key stakeholders will notice the difference. Ready product feeds for GenAI success There is huge potential for GenAI in the fashion industry, especially with how brands could utilise digital personal shoppers onsite or in apps. Having comprehensive and accurate product data becomes increasingly important. And the ability for these tools to generate relevant and appealing content can heavily rely on the quality of data they are provided, beyond the product images. The more detailed and up-to-date the data, the more effectively GenAI can tailor its outputs to meet individual consumer preferences, especially if brands are looking to base it on customers current and historical affinities. For instance, if a customer searches for an ‘wedding outfit’ a customer interested in certain brands, materials or sustainable fashion trends will receive more relevant suggestions if the product feed includes specific attributes like ‘brand’, ‘organic cotton’, ‘recycled materials’ and ‘fair trade’. In contrast, incomplete or outdated product data can lead to irrelevant recommendations, having a negative impact on the consumer experience. Brands will need to ensure that their product feeds are meticulously curated with regular updates and comprehensive attributes reflecting current trends and consumer interests to leverage the full potential of GenAI. This will enhance AI-driven personalisation, leading to improved customer satisfaction and loyalty. Tidy up product data for optimised recommendation results A great example of this is home24. After acquiring over 300,000 products, home24’s chief product officer, Gianluca Randisi, undertook the technical, time-consuming project of streamlining the company’s product feed. This effort significantly reduced costs and tripled the company’s share of revenue, demonstrating the transformative impact of a clean product feed. To achieve similar results, brands must clear their product feeds of duplicate, unused, irrelevant, or inconsistent product attributes. For example, a site visitor looking for an academic blazer to match their tennis skirt might come up short in the recommendations if the products are poorly categorised. Likewise, someone might miss out on items that embody specific trends if they’re sifting through broadly tagged products. Ensure keywords reference useful differentiators like colour, material, specific campaigns, fit, and sale status. Avoid overly specific descriptors, fashion brand marketing terms, and redundant terms in different languages. Focus on aligning keywords and attributes with specific trends, personalities, or styles. Foster cross-department collaboration Product feed cleaning is rarely a one-person task. Product development, buying, and marketing teams should meet internally twice a year to identify emerging trends, colours, pattern styles, and product types that need to be added to the feed. Determine whether existing product attributes suffice or if new ones are necessary. Establishing a structured process integrated into seasonal cycles will make maintenance easier. With these elements in place, retailers can create a mapping file. Product feed mapping for personalisation Consider creating a mapping file for emerging fashion trends, such as Y2K throwback fashion and sustainable clothing. Y2K fashion attributes: asymmetry; metallic fabrics; oversized baggy cuts; low-waist jeans; cargo pants; rhinestones; chunky sneakers; cropped bandeau tops; miniskirts; denim; headscarves and bandanas; and patterns like flames, dragons, and yin and yang. Then there are also sustainability fashion attributes, such as vegan; PVC-free; organic cotton; recycled materials; PFC-free; Gruner Knopf; emission-neutral; Fairtrade and FSC. For sustainability, use a “True” or “False” column to identify whether an item meets sustainable criteria. Similarly, use an additional True/False column for trending items. This enables dynamic serving of recommendations to users with specific interests, like Y2K fashion or sustainable apparel. Incorporate demographic modifiers to account for different audience segments, ensuring trends are relevant across diverse groups. An optimised product feed means improved personalisation Cleaning up product data can be daunting, but neglecting it can render your product database unreliable. The effort is well worth the investment, as consumers reward brands that get personalisation right. A significant 72% of shoppers expect businesses to recognise them and know their tastes. Building relationships based on personalisation fosters long- term loyalty and engagement, positively impacting conversions, product sales, and customer retention across the company. By optimising the product feed and personalising offerings, brands can better meet consumer demands and navigate the ever-evolving fashion landscape. Retailers that embrace this strategy, will watch their brand thrive in 2024 and beyond.