Expert Interview: Alejandra Garcia, Product Leader
.png)
March 2025
Introduction
At Fabric, we love to learn from industry leaders.
I recently spoke with Alejandra about her perspective on personalization. Alejandra is an advisor to Fabric and actively contributes to product development.
Alejandra has spent 10 years leading Product and Data teams, working at the intersection of data, AI, and customer experiences, most recently at Zalando and Hello Fresh.
As Head of Product for Zalando, she developed AI solutions for personalized recommendations and will now be leading menu planning and recommendations at Hello Fresh.
She has a wealth of knowledge on e-commerce and personalization. I documented some particularly interesting excerpts from our conversation.
How do you see personalization shifting from broad segmentation to individual-level experiences?
Personalization has evolved from basic demographic segmentation to highly individual experiences powered by AI and real-time data.
Traditionally, brands segmented customers based on factors like age, location, or purchase history. Today, advancements in machine learning and granular user data enable brands to create tailored experiences at an individual level.
Retailers are under increasing pressure to deliver dynamic and innovative experiences, especially as younger generations expect more personalization and excitement. With competition moving faster than ever, personalization is no longer just a competitive advantage, but a business necessity.
Big tech data, such as Google searches, social media interactions, and browsing behavior, provides context for understanding consumer intent. When combined with first-party data like purchase history and preferences, it helps brands predict needs, refine recommendations, and create hyper-personalized journeys.
The key, however, is not just collecting more data, but using it intelligently to add value to the customer experience.
What are the biggest challenges in delivering personalization at scale, especially as consumer expectations rise?
The biggest challenges in scaling personalization are data quality, real-time processing, and balancing personalization with privacy. Many organizations struggle with siloed data, making it difficult to build a unified customer view. Additionally, real-time personalization requires robust infrastructure and AI capabilities, which can be costly and complex to implement.
Another key challenge is keeping customer data up to date and prioritizing recency. Consumer preferences and trends shift rapidly, especially in fashion and retail. A recommendation that was relevant last month might not be today. Ensuring that personalization models continuously learn from fresh data—while avoiding outdated assumptions—is critical to maintaining relevance and engagement.
Consumer expectations have also shifted—customers now expect relevance without feeling like they’re being overly tracked. Striking the right balance between personalization and privacy is crucial. Brands must be transparent about data usage and offer value in return for consumer trust.
How do you prioritize which data sources or signals to incorporate first when building personalized experiences?
Prioritization depends on three key factors: relevance, reliability, and accessibility.
- Start with first-party data – The most valuable insights often come from direct customer interactions—purchase history, browsing behavior, and feedback. These are highly relevant and privacy-compliant. However, most brands have limited, outdated first party data and that's where using off-site data can accelerate personalization.
- Start using contextual off-site signals – Search data, social engagement, and external data sources can enhance personalization and should be carefully validated to avoid misleading correlations.
- Emphasize real-time insights – Prioritizing fresh data ensures recommendations remain relevant. Recency is especially crucial in fashion, where intent shift rapidly.
- Test and iterate – Not all data signals will drive meaningful impact. Running experiments helps determine which inputs genuinely improve customer experience and business outcomes.
Ultimately, the goal is to build a data strategy that enhances personalization while maintaining transparency and trust.
How do you ensure that personalization efforts remain ethical and respect consumers’ privacy?
The foundation of ethical personalization is ensuring transparency, giving control, and offering clear value.
Think about it this way: When a brand suggests a product based on your preferences, you want to feel like it's genuinely helpful—not like you're being watched or manipulated. So, how do we make that happen?
First, be upfront about data usage—let customers know what data is being collected and how it’ll improve their experience. It’s about trust.
Second, give people control over their data, letting them choose what they share and even how deeply they want to personalize their experience. For instance, I might opt-in to share my search history but decline to share my social activity.
It’s all about respecting individual preferences.
A privacy-first approach also includes:
- Optimising data collection: Collect and use what you need—don’t hoard data just because you can.
- Clear opt-ins: Let customers actively decide how much personalization they want, so they feel in control.
- Compliance: Make sure your practices align with regulations like GDPR to protect both the customer and the business.
When done right, personalization feels like a helpful assistant rather than an unwanted guest—enhancing the experience, not overwhelming it.
What key metrics do you track to gauge the effectiveness of personalization—conversion rates, user satisfaction, or something else?
Measuring personalization success requires aligning customer experience with business outcomes, but the right metrics depend on the specific experience being optimized.
While conversion rates and engagement metrics are important, they don’t tell the full story.
Before experimentation, it’s crucial to define and agree on the right success metrics based on the personalization goal. Some key options include:
- Customer retention & satisfaction – Are customers happy with their purchases? Do they return to your site?
- Discovery time & time to purchase – Is personalization making the shopping experience more efficient?
- Return rates – Are recommendations leading to better buying decisions?
- Customer lifetime value proxies – Repeat purchases, basket size, and engagement beyond a single transaction.
- Inclusivity of impact – Are we improving experiences for a broad customer base, not just a subset?
Personalization should create lasting value, not just short-term engagement spikes. By selecting the right KPIs upfront, we ensure optimization efforts drive meaningful impact.
How do you see fashion evolving over the next 3–5 years with emerging technologies like generative AI? What are some promising or disruptive trends that could redefine how brands engage with consumers?
Over the next 3-5 years, I anticipate significant shifts in fashion, especially with the rise of generative AI. This technology has the potential to revolutionize design, personalization, and even production. We can expect:
- Hyper-Personalized Shopping Experiences – With advanced AI, customers will experience deeply individualized shopping where AI generates tailored recommendations, virtual try-ons, and product designs based on personal preferences, body types, and even emotional states.
- AI-Driven Design & Customization – Generative AI will enable designers to create new styles, collections, and clothing based on evolving trends and data from consumer behavior. Consumers could even design their own pieces, blending creativity with AI insights.
- Sustainability Improvements – Generative AI can reduce waste by optimizing production processes, creating on-demand products and reducing overproduction. It could also predict trends more accurately, minimizing unsold stock.
- Virtual and Augmented Fashion – As the metaverse and AR/VR technologies evolve, fashion will expand beyond physical retail, offering virtual wardrobes and clothing experiences for online environments.
Generative AI and related technologies will create a more agile, personalized, and sustainable fashion ecosystem, making it a truly exciting time for both customers and businesses.
How do you foster a culture of experimentation and iteration within an organization?
Fostering a culture of experimentation starts with creating a safe environment where failure is seen as part of the learning process.
Key actions include:
- Clear Vision and Alignment – Ensure that everyone understands the goal of experimentation: to learn and improve. Alignment on what success looks like and how experimentation ties into broader objectives is essential.
- Data-Driven Mindset – Encourage data-driven decision-making by using experiments to generate insights that can guide strategy, not just validate assumptions.
- Empowering Teams – Provide teams with the autonomy to test ideas, along with the tools and resources to do so effectively. Teams should feel empowered to experiment and iterate quickly.
- Celebrate Learnings – Recognize not only the successful outcomes but also the insights gained from unsuccessful experiments. This ensures that experimentation is valued as a way to drive continuous improvement.
By integrating experimentation into the company’s DNA, teams will become more agile and able to innovate faster.
Having worked with ML products for years, I can say that experimentation is crucial for optimizing models, as they will never be perfect. A/B testing of new customer experiences is equally important for continuously refining personalization and improving engagement.
The most valuable experiments are those that combine insights from various touchpoints in the customer journey. It’s not just about success or failure, but about gathering actionable insights that shape and improve the overall experience.