The ring builder
The real configurator, production code, embedded here on mobile. Pick a diamond, a setting and a metal, and watch the ring resolve. The same engine runs inside the Shopify storefront and the showroom tablet.
- My role
- Design lead for the product line. I set product & design strategy, hired and led a designer for each surface, and owned the cross-surface system and the configurator UX, end to end
- Context
- Nivoda, a global B2B jewellery marketplace, productising its catalogue as SaaS for the jewellers it supplies
- Duration
- First year, $0 to ~$1M ARR
- Stack
- Shopify app · web ring builder · iPad showroom app · Nivoda catalogue API (1M+ stones)
- What shipped
- Back-office configurator · Shopify app with embedded ring builder · showroom tablet consultation app · one shared configurator engine
A million stones is a curse at the point of sale
Nivoda's edge is supply: over a million diamonds and gemstones, natural and lab-grown, white and fancy colour, plus settings, metals and melee. For a jeweller, that's inventory they could never hold. For a consumer at the buying moment, it's paralysis. And jewellers had no good way to put that catalogue in front of their customers, online or in store, without rebuilding it themselves.
- Overwhelming choice. A million SKUs is a spreadsheet, not a shopping experience.
- No shared surface. Each jeweller's website, and their showroom, was a separate, manual build.
- Two opposite buying moments. Self-serve online and assisted in-store want very different things from the same catalogue.
Configure once, sell the ring everywhere
One back office, owned by the jeweller, curates the catalogue: the stones, the metals they'll sell, the settings, the markups. Everything downstream inherits it. Consumers never meet a million options; they meet the jeweller's curated shop, with a guided build on top of it.
And the build itself is one engine, embedded in two consumer surfaces, so the experience is the same whether someone configures a ring at midnight on their phone or beside a sales consultant on a tablet.
Three surfaces, one system
1 · The back office, where it all starts
The jeweller-facing tool, on Nivoda's core platform. It's where a jeweller curates their digital shop: which stones and jewellery they'll carry, which metals and variants they'll offer, and the markups they'll charge. Set it here and it flows, unedited, into every consumer surface. This is the surface that makes the other two possible, and the reason the catalogue's size never reaches the customer as noise.
2 · The Shopify app and ring builder
The online surface. The Shopify app pipes the jeweller's curated diamonds, gemstones and jewellery straight into their own website, and the ring builder lets an end consumer build a fully custom engagement ring: any stone from the million-plus catalogue, white or fancy colour, natural or lab-grown, plus gemstones and melee. It was designed mobile-first, because that's where most people first meet the ring.
3 · The showroom tablet, assisted sales
The in-store surface, a tablet app built as a consultation tool for assisted sales. It runs the same configurator, but the experience is more immersive: a guided question flow, big imagery and video, the kind of glossy, shared screen a consultant and a customer lean over together. It augments a human sale; it doesn't replace it. Same engine as the web, opposite posture.
The decisions that shaped it
A handful of calls set the whole line, the opinion is the artifact:
- One configurator engine, two surfaces. I resisted building separate web and tablet flows. Divergence is a cost you pay forever; one engine, themed per context, is paid once.
- Curation over search. The back office decides what a shop sells; the consumer gets a curated build, never a million-row search. The catalogue's scale is a backstage asset, not a front-stage burden.
- Mobile-first for the storefront. Most consumers meet the ring on a phone, so the web builder was designed mobile-first and scaled up, not the reverse.
- Assisted, not automated, in the showroom. The tablet is glossy and immersive on purpose, big stones and video, so consultant and customer look at it together. The goal is a better conversation, not a self-checkout.
A designer to each surface, one direction
Beyond the screens, what the role actually was:
- Led the line. Design lead for the product line, I hired and led a designer for each surface and held a single direction across all three.
- Strategy to pixels. Product and design strategy at the top, hands-on configurator UX at the bottom, and the system that connected them in between.
- Built the spine, not three silos. Shared patterns and one configurator engine kept the surfaces feeling like a single product, not three apps wearing the same logo.
- Anchored to the commercials. I designed against ARR and GMV, not just usability. The configurator is the revenue surface, so its conversion was a design metric, not an afterthought.
What it became
- $0 to ~$1M ARR in year one, a new product line stood up from nothing.
- ~$800K of recurring SaaS, plus marketplace GMV and commission on the stones worth multiples more.
- Three live surfaces from one back office, back-office configurator, Shopify app + ring builder, showroom tablet.
- A configurator real enough to ship, and to play with, embedded live at the top of this page.
And the subscription line undersells it. The SaaS is the visible tip; the business underneath is the marketplace GMV and the commission earned on every stone the configurator sells:
Configure once. Sell the ring everywhere.