The question isn't whether to use AI. It's whether you have the design judgment to use it without losing what makes your brand worth finding in the first place.
There's a version of ecommerce design that works. It converts. It follows a well-known playbook with sticky headers, urgency nudges, social proof stacked above the fold, hero images cropped to show product fast. It's optimized. It's also, almost always, forgettable.
AI is making it worse. As AI tools become standard in ecommerce workflows, they reference the same sources, surface the same best practices, and nudge teams toward the same decisions. The result is a kind of design convergence: sites that are technically sound, comprehensively optimized, and increasingly indistinguishable from each other.
The initial reaction to the above is usually to question whether AI is helpful or hurtful in this context. But that's the wrong question. The question isn't whether to use AI. It's whether you have the design judgment to use it without losing what makes your brand worth finding in the first place.
"Our job is to keep the experience clear while finding the moments that make a brand feel like itself instead of everyone else."
Marine Layer: Inside the brief
Marine Layer was founded in 2009, built on West Coast optimism, whimsy, and materials described as absurdly soft. By the time Pattern came in, the brand had 55+ stores, mailed over a million catalogs every six to eight weeks, and had a loyal customer base that knew exactly what Marine Layer felt like.
The brief was direct: make the site feel more like us. Conversion and AOV mattered but so did brand fidelity, with both weighted equally. That's a harder brief than it sounds. Most ecommerce engagements optimize for one or the other. Marine Layer wanted both.
The risk vs. the reward
The risk Pattern identified going in was the same risk facing the broader industry. AI-powered personalization tools are widely available, increasingly powerful, and when deployed without restraint, tend to make every site feel like every other site. Feature-first implementations like adding recommendation engines, chatbots, and behavioral triggers because they're available, can actively undercut brand differentiation in a market where differentiation is one of the most powerful tools a brand has.
Marine Layer, with its analog warmth and catalog heritage, would have been particularly exposed to that kind of damage. An AI chatbot doesn't fit a brand built on handwritten catalog copy and tactile softness. Personalization that announces itself doesn't fit a brand whose whole proposition is ease and warmth.
The standard had to be brand fit, not just feature completeness.
"When deployed without restraint, AI-powered personalization tools tend to make every site feel like every other site."
Conversion follows experience, not the other way around
Pattern deployed AI as an intelligence layer, not a creative one. Personalization worked in the background, surfacing the right products to the right customers without announcing itself. Rather than maximizing the number of AI touchpoints, our team identified the highest-impact placements and left the rest alone.
Product pages, for example, needed only two recommendation types rather than the three that most implementations default to. Outfitting and personalized. That's it. Every element was tested against a single question: does this feel like Marine Layer? If the answer was no, it didn't ship.
In an industry that prioritizes squeezing every last drop of conversion out of any given feature, that level of discipline is much harder to maintain than it sounds. The temptation with any powerful tool is to use more of it. The judgment to use less - to treat restraint as a design decision - is what separates true long-term brand-building from feature deployment.
The results
The post-launch numbers for Marine Layer reflected how customers actually respond to design that actually trusts them:
- Revenue per session: 3x higher
- Conversion rate: 2x higher
- Conversion + AOV lift: Mid double-digit
Customers who engaged with the personalization layer generated 3x higher revenue per session and converted at 2x the rate of those who didn't. Mid double-digit increases in both conversion rate and AOV followed, with AOV becoming the primary north star, because conversion was already strong before the project began.
That last point is a really important one, because it flies in the face of the rationale in most site redesign undertakings. The goal from the outset wasn't necessarily to fix a broken experience. It was to make a good one even better by making it look, feel and behave more like itself. That's a much harder design problem and technology problem, and a much more valuable one to solve.
What this means for ecommerce
AI will keep raising the floor. Competent ecommerce is getting cheaper and easier to produce. Which means the ceiling - design with a genuine point of view, experiences that earn loyalty rather than just transactions - is where the real differentiation lives.
The brands that will define the next decade of ecommerce aren't the ones that will have deployed every available tool. They're the ones that had the taste and judgement to know which tools served the brand and which ones didn't. That judgment doesn't come from AI. It comes from the team applying it.
Marine Layer already knew who they were. Pattern's job was to make sure the site knew it too.
About Pattern
Pattern helps brands unlock their next catalyst moment through experience-driven eCommerce. We design modern brands, eCommerce experiences, products & platforms that help founders, disruptors and visionaries find their place at the intersection of culture, community & commerce.
No Comments.