Using AI to scale boutique-level personalization, with Ewoud Frielink
.png)
The new standard in ecommerce: Personal, predictive, and powered by AI → Get the full episode here
“We want to be a personal boutique at scale. The challenge is bringing that boutique-style service into the digital world without losing the personal touch.”
— Ewoud Frielink, CTO of Omoda
In today’s ecommerce market, personalization isn’t a differentiator—it’s a demand. Yet for most retailers, achieving true personalization at scale remains elusive. As online shopping becomes the dominant revenue channel, the pressure is on to replicate the intimacy of in-store experiences across digital touchpoints.
That’s exactly what Ewoud Frielink, CTO of Dutch fashion retailer Omoda, is doing. With over 150 years of brand legacy and 80% of revenue coming from online sales, Omoda isn’t just adapting to the digital age—they’re shaping it.
In a recent episode of Ecommerce Toolbox: Expert Perspectives, Ewoud sat down with host Kailin Noivo to share how Omoda is using AI, large language models, and automation to transform ecommerce from transactional to relational. The strategies he outlines aren’t just about scaling operations—they’re about deepening customer connection and unlocking new levels of operational efficiency.
🎧 Listen to the full conversation on Apple, Spotify, or YouTube
From clicks to connection—Why AI personalization has evolved
Traditional personalization relied on basic behaviors—clicks, past purchases, or generic product recommendations. Omoda’s vision goes far deeper.
Our goal is to be a “personal boutique at scale,” Ewoud explains. That means crafting digital journeys that feel intuitive, empathetic, and tailored—just like your favorite in-store stylist might provide.
To achieve this, Omoda launched an AI-powered stylist that uses large language models (LLMs) to offer natural, contextual style advice. Shoppers can ask for help choosing an outfit for a wedding or pairing a red blouse for a dinner—and the tool responds with curated suggestions, not product dumps.
- 80% of users rate the stylist’s suggestions as helpful.
- Customers are volunteering richer context than ever, giving Omoda deeper data on intent, occasion, and preferences.
This isn’t personalization-as-usual. It’s AI as an emotional connector—and it’s changing how customers shop.
.png)
Lean-back shopping: From search-and-scroll to proactive recommendations
Where most digital journeys still rely on active search, Omoda is investing in “lean-back ecommerce.” That means:
- Generating full outfit recommendations based on purchase history, wishlists, and browsing behavior
- Delivering style inspiration before the customer even asks
- Shifting from reactive personalization to proactive curation
The implications for conversion rates, session time, and lifetime value are massive. In a digital world increasingly defined by decision fatigue, lean-back shopping reduces friction and increases delight.
Strategic takeaway: AI should anticipate needs—not just respond to queries. That’s how digital becomes truly human.
90% AI-generated product data: Automating for quality and scale
Behind the scenes, Omoda has built a custom PIM system where 90% of product attributes are generated by AI. This isn’t just a cost-saving measure—it’s a play for consistency, findability, and conversion.
Manual entry is prone to error. But AI enables:
- Clean, structured data across thousands of SKUs
- Faster time-to-site for new products
- Improved search relevance and product discoverability
By ensuring data hygiene at scale, Omoda sets the foundation for performance across search, personalization, and UX.
AI in customer support: Insight over replacement
Unlike companies rushing to replace human agents with bots, Omoda’s approach to AI in customer service is focused on diagnostics, not displacement.
They use AI to:
- Analyze why customers reach out
- Surface recurring pain points
- Identify opportunities for process or UX improvements
The result? A more strategic support function, where human agents can focus on nuanced interactions, and systemic issues are resolved upstream.
The next frontier: AI-generated imagery for faster merchandising
One of the most future-forward experiments at Omoda is the use of AI-generated product imagery. The idea: eliminate manual studio shoots by generating styled visuals from iPhone snapshots taken during the buying process.
This would allow:
- Faster content creation (before inventory hits the warehouse)
- Cost savings on photo production
- Increased agility in launching campaigns
While still in early testing, this initiative signals how AI could collapse the time-to-market window for fashion and retail brands.
Key takeaways for ecommerce leaders
Omoda’s AI strategy is a masterclass in how to blend innovation, operational efficiency, and brand differentiation. Here’s what to take away:
- Personalization must evolve from transactional to emotional. Use AI to inspire, not just recommend.
- Lean-back shopping experiences reduce friction and mirror high-touch boutique service online.
- Data quality is a growth lever. Automating product attributes isn’t just efficient—it boosts findability and conversion.
- Customer support should be strategic. Let AI find patterns, but keep humans for empathy.
Don’t wait on AI imagery. The brands that experiment early will be first to scale content velocity.
Closing thought: Innovation is human
Ewoud’s approach reminds us that the future of ecommerce isn’t just tech-driven—it’s experience-driven. The goal isn’t automation for its own sake. It’s to elevate the digital experience to something that feels intuitive, personal, and human—at scale.
If you're a CDO, CTO, ecommerce VP, or digital experience leader, this episode offers a real-world blueprint for applying AI with purpose and precision.
Tune in to the full episode for more insights:
.png)
