From Etsy to ChatGPT: How Customers Are Finding Sivan Lotan Jewelry
One customer found the Israeli fine jewelry designer while searching ChatGPT for a Star of David necklace. The sale is a small but revealing example of how jewelry discovery is changing.
By Dana Roth · Found & Form·

Reporting
From Etsy to ChatGPT
For more than a decade, Sivan Lotan has been building a jewelry business around something that is difficult to manufacture at scale: work that looks unmistakably like her own.
Lotan is an Israeli jewelry designer and trained goldsmith who graduated from Shenkar College of Engineering and Design in 2010. Her work moves between sculptural, organic fine jewelry and pieces rooted in identity, memory and Jewish symbolism.
She has also been selling jewelry online long enough to have watched the mechanics of discovery change around her.
Her Etsy shop dates to 2014. Today, she also operates an extensive independent ecommerce site, maintains a presence across social and visual discovery platforms, and receives customers through search, recommendations, referrals and repeat business.
And recently, a customer found her somewhere else:
ChatGPT.
Reporting
A jewelry search that became a sale
Lotan shared the story with Found & Form after a customer contacted her following a purchase.
The customer had been searching ChatGPT for the right Star of David necklace. During that search, Sivan Lotan Jewelry surfaced as a recommendation.
He explored her work, found a design that stood out to him and bought it.
Afterward, he told Lotan that he was happy with the purchase, that he had found something special and that her designs stood out from the alternatives he had considered.
It is only one customer journey.
It would be a mistake to turn one sale into a sweeping claim that AI has become a major acquisition channel for Sivan Lotan Jewelry.
But that is precisely what makes the example interesting.
It is a documented instance of something jewelry brands increasingly need to understand: a potential customer can now begin a product search without opening Google, Etsy, Instagram or a retailer at all.
The recommendation itself can happen inside an AI conversation.
And the brands that appear there do not necessarily arrive from nowhere.
Reporting
Before ChatGPT, there was Etsy
When Lotan began selling through Etsy, the marketplace solved a fundamental problem for a small independent designer: audience.
She did not have to first build a large following and then convince that audience to visit a standalone store. Etsy already had shoppers searching for handmade and distinctive jewelry.
Lotan told Found & Form that Etsy was crucial in the early years because it exposed her work to international customers who were already looking to buy.
Her Etsy shop remains active today and has been on the marketplace since 2014.
The experience of making her first online sale clearly mattered. On her Etsy profile, Lotan recalls the feeling of realizing that something created in her small studio had traveled across the world to a customer.
That is the original marketplace promise: the maker brings the product; the marketplace brings discovery.
But as the business developed, Lotan wanted something Etsy could not completely provide.
Ownership.
Reporting
Building somewhere she could control
Lotan eventually developed her own ecommerce site alongside Etsy.
She told Found & Form that the decision was about more than marketplace fees. An independent site gave her greater control over how the work was presented, helped reduce dependence on a marketplace, created a more direct relationship with customers and offered more protection around a brand she had spent years developing.
That distinction matters.
On a marketplace, the product lives inside someone else's discovery environment.
On an owned website, the brand controls the environment, but it also becomes responsible for getting people there.
“Creating jewelry and creating visibility for that jewelry are two completely different jobs.”
Her current website shows what that second job can eventually produce.
Sivan Lotan Jewelry now has hundreds of product pages spanning engagement jewelry, symbolic pieces, nature-inspired designs, personalized work and Jewish heritage jewelry.
Its Star of David collection has grown into a substantial body of work rather than a single product optimized around a search term.
That depth becomes particularly interesting when we return to the ChatGPT customer.
The work
A selection from Sivan Lotan Jewelry
Photography: Sivan Lotan Jewelry. Used with permission.
Reporting
Discovery is no longer one channel
Ask where Sivan Lotan Jewelry gets customers today and the answer is not simply Etsy.
Lotan told Found & Form that customers now find her through a mixture of Etsy, Google, Instagram, Pinterest, recommendations, word of mouth and returning customers.
The paths overlap.
Someone might first see a piece on Pinterest and later search the designer's name on Google. Someone else might discover a product on Etsy before eventually buying directly. A past customer might recommend the brand to a friend.
And now an AI recommendation can enter that same system.
This makes the traditional question, “Which channel brought the customer?” increasingly incomplete.
“What made this brand discoverable in the first place?”
Found & Form analysis
Why did ChatGPT find Sivan Lotan?
We cannot answer that definitively from a single customer interaction.
Found & Form does not have access to the exact internal process that produced that recommendation, and AI recommendations can change depending on the model, prompt, date and context.
But we can examine the public discovery footprint surrounding the brand.
Sivan Lotan Jewelry is not represented online by a single homepage.
There is more than a decade of Etsy history associated with Sivan Lotan and her work.
There is an independent website with hundreds of product pages.
There are dedicated collections built around specific product categories and concepts.
There are detailed pages about Star of David jewelry, Magen David jewelry, Jewish heritage pieces, engagement rings, symbolic jewelry and other distinct areas of the collection.
The brand maintains identifiable presences across multiple discovery platforms in addition to its owned site.
The same designer, brand name, product categories and themes therefore exist across multiple parts of the public web.
None of that proves why an AI system selected Sivan Lotan Jewelry for this particular customer.
It does show that the recommendation happened against a much larger existing digital footprint.
That distinction is important.
Found & Form visualization
Sivan Lotan's Found Footprint
Eight discovery surfaces around one brand. Each tile names the surface and the kind of evidence behind it.
The brand
Sivan Lotan Jewelry
Marketplace
Etsy
Observed: shop active on Etsy since 2014
Owned
Sivan Lotan Jewelry website
Observed: 391 product pages
Search
Google and organic search
Observed: Semrush US ranking data
Social
Observed: linked from the official website
Visual Discovery
Observed: linked from the official website
Community
Word of mouth and recommendations
Reported by Sivan Lotan
Retention
Returning customers
Reported by Sivan Lotan
AI
Documented ChatGPT customer discovery
Reported by Sivan Lotan: one customer journey
Methodology note. Found Footprint maps observable discovery surfaces. It is not a performance score and does not imply equal contribution from each channel.
Found & Form analysis
AI discovery may be an outcome, not a channel you build in isolation
There is an understandable temptation for brands to treat AI visibility as another optimization project:
- How do I rank in ChatGPT?
- What do I need to add to my website?
- Which keywords will make AI recommend me?
Sivan Lotan's example suggests a more complicated possibility.
The customer who found her through ChatGPT encountered the latest layer of a discovery system Lotan had been building for years.
Etsy created marketplace visibility.
Her independent site created an owned source of information about the brand and its products.
Search made those pages discoverable outside the marketplace.
Social and Pinterest created additional paths into the work.
Customers, reviews, recommendations and repeat purchases created another layer around the business.
Then an AI system became another interface through which a customer could potentially encounter all of it.
The AI recommendation did not necessarily replace Etsy, Google or the independent website.
It sat on top of that ecosystem.
Found & Form analysis
The move beyond Etsy does not have to mean leaving Etsy
Lotan's story also complicates another common narrative among independent sellers: that graduating from Etsy means abandoning it.
She hasn't.
Etsy still provides something valuable: an established marketplace with built-in consumer intent.
Her website provides something different: control.
The two can coexist.
For independent jewelry brands, the more durable strategy may not be choosing between marketplace discovery and owned ecommerce. It may be building enough independent visibility that no single platform determines whether the brand can be found.
That becomes particularly important as discovery fragments further.
- A customer might begin with Google.
- Or Etsy.
- Or Instagram.
- Or Pinterest.
- Or a YouTube creator.
- Or Reddit.
- Or a recommendation from a friend.
- Or, increasingly, a question asked to an AI assistant.
Found & Form analysis
What Sivan Lotan's ChatGPT sale actually tells us
One sale cannot tell us how frequently AI recommends independent jewelry brands.
It cannot tell us what percentage of Sivan Lotan Jewelry's revenue comes from AI discovery.
And it cannot prove which elements of her web presence caused ChatGPT to surface the brand.
What it can establish is much narrower, and much more useful:
“A customer looking for a particular kind of fine jewelry asked an AI system for help, discovered an independent designer through that interaction, and purchased from her.”
That happened.
For an industry in which discovery has historically been dominated by retailers, marketplaces, search engines, magazines and social platforms, that is worth paying attention to.
Especially because the recommendation was not simply a famous jewelry house.
It was an independent designer working from Tel Aviv whose digital footprint had been accumulating for more than a decade.
The Found & Form Takeaway
AnalysisSivan Lotan Jewelry did not become discoverable the day a customer typed a question into ChatGPT.
The brand had already spent years becoming discoverable.
Etsy helped customers find the work.
An independent website gave the brand a home it controlled.
Search, social, Pinterest, referrals and returning customers created additional routes back to it.
And somewhere along that path, AI became another door.
For independent jewelry brands, that may be the more useful way to think about AI visibility.
- Not as a replacement for search.
- Not as a replacement for marketplaces.
- Not as a trick involving a handful of new keywords.
But as another discovery layer built on top of everything the web already knows about a brand.
The emerging challenge is not simply to “rank in AI.” It is to build a brand with enough clear, consistent and distinctive evidence across the web that it can be found wherever the next jewelry search begins.
The brand in this study
Sivan Lotan Jewelry
- Category
- Independent fine jewelry
- Based in
- Israel
- Official website
- sivanlotan-jewelry.com
- Etsy
- etsy.com/shop/SivanLotan





