What a five-star review taught us about how people actually shop for jewelry
Nayiri Jewelry told us their customers search by feeling, not by filter. That single line explains why so many jewelry and accessories brands are invisible in AI search, and what to do about it.
There’s a line in one of our Shopify App Store reviews that we keep coming back to internally. It’s from Nayiri Jewelry, a UK jewelry brand that installed Veristyle earlier this year, and it says something we hadn’t quite put into words ourselves.
“Veristyle focuses on how AI models interpret product attributes: metal type, design style, gemstone characteristics, symbolic meaning, and purchase intent. For jewelry, where customers search by feeling, having this level of semantic context makes a meaningful difference.”
Search by feeling. Not by carat weight as a standalone filter. Not by metal purity alone. By feeling.
Why jewelry breaks the usual search box
Most on-site search and most AI search optimization advice is written with categories in mind that behave like electronics. A shopper wants a 55 inch TV, types “55 inch TV,” and gets a clean list of matches. Specs map neatly to intent.
Jewelry doesn’t work that way, and neither does most of fashion, but jewelry makes the gap obvious faster than anything else in the catalog. Someone isn’t searching for “14k gold ring, size 7.” They’re searching for “a ring that feels like it means something” or “understated gold pieces for a meaningful gift” or “something that doesn’t look like it’s trying too hard.” None of that maps to a spec sheet. All of it maps to how a piece is designed, what it symbolizes, and who tends to buy it for what occasion.
That’s the gap Nayiri was describing. A product feed that lists “925 sterling silver, 18mm” is technically accurate and functionally useless for the way people actually decide.
What “semantic context” looks like in practice
This is where computer vision earns its keep. When Veristyle reads a jewelry catalog, it isn’t just logging metal type and stone. It’s tagging design style (minimalist, statement, vintage-inspired), symbolic associations (celebration, remembrance, self-purchase, anniversary), and the kind of purchase intent a piece tends to satisfy. That gets structured into a format AI engines can actually use, so when someone asks ChatGPT or Perplexity for “rings for people who prefer understated gold” or “jewelry for a meaningful gift,” there’s something specific for the model to match against.
Nayiri’s team noticed this directly. After enrichment, their pieces started surfacing in exactly those kinds of searches: gifting queries, sentiment-driven queries, the ones a keyword-matching system would have missed entirely because none of the right words were sitting in the product title.
It’s not only a jewelry problem
We picked jewelry for this post because it’s the clearest example, but the same pattern shows up across the brands we work with. PeachLilly, a fashion brand also on our reviews page, described it slightly differently: “Veristyle takes the content we already have and structures it in a way that AI systems can actually understand. Since using it, we have noticed better visibility and more relevant traffic.” Same mechanism, different vertical.
The common thread is that a lot of genuinely good product knowledge already exists inside these brands. A merchandiser knows a necklace is bought mostly for anniversaries. A stylist knows which silhouette flatters which body type. That knowledge just usually lives in a person’s head, or in the way products get merchandised on a landing page, rather than anywhere a language model can read it.
What to take from this if you sell jewelry, or anything people buy for how it makes them feel
If your catalog is described only in specs (metal, size, material), you’re leaving the actual decision-making language on the table. The fix isn’t a rewrite of your product pages. It’s adding a layer of structured, semantic attributes underneath what’s already there, the kind that map to how your customer actually thinks and searches, so that both your on-site search and AI assistants have something real to work with.
If you want to see what this looks like for your own catalog, an AI visibility audit takes about five minutes and shows you where the gaps are between how your products are tagged and how your customers actually search. Or if you’d rather talk through what this would look like for a jewelry, accessories, or fashion catalog specifically, book a demo and we’ll walk through it together.