To combat returns, Zalando's AI analyzes a user's photos, purchase history, and data from similar customers to predict the correct size for a new item. The ultimate goal is to become so confident in its predictions that for some products, the user won't even need to select a size at all.

Zalando didn't try to stock everything at once. They first targeted very specific, high-intent searches like "Adidas Zamba size 46." Winning these allowed them to build capital and inventory to then capture broader searches for brands, categories, and eventually entire markets, creating a scalable growth loop.

Platforms like Axio go beyond spotting trends by analyzing customer pain points from negative reviews on sites like Amazon. This identifies specific product flaws and reveals clear, data-backed opportunities for creating superior products.

Zalando leveraged its extensive European logistics footprint, initially built for its B2C business, into a new B2B revenue stream. Brands can now use this infrastructure to manage their own e-commerce fulfillment across the continent, avoiding massive CapEx and gaining network benefits.

A common mistake is building a visually impressive data product (like Google Earth) that is interesting but doesn't solve a core, recurring business problem. The most valuable products (like Google Maps) are less about novelty and more about solving a frequent, practical need.

The software practice of analyzing user clicks can be applied to any business. For retail, identify your top-spending customers and reverse-engineer their entire journey, from their first store visit to their big purchase. This helps find common patterns—like interacting with a specific employee—that can be replicated for all customers.

For brands with distributed networks, a central marketing platform provides crucial visibility into what local teams are actually creating. Tracking metrics like content generation and channel preferences uncovers trends that are otherwise invisible, allowing central marketing to understand ROI and learn from frontline experiments.

Instead of traditional market research tools, scrape Google Maps data. Analyze business listings, review volume, and sentiment to find niches with high customer demand but low satisfaction, signaling a clear market gap for a new or improved service.

Brands miss opportunities by testing product, packaging, and advertising in silos. Connecting these data sources creates a powerful feedback loop. For example, a consumer insight about desirable packaging can be directly incorporated into an ad campaign, but only if the data is unified.