Skip to content
1994 — Nineteen Ninety-Four
Problem Solving, Whitelabel

Feed-Driven Yacht Search

By Daniel Jones • 23rd September 2026
Feed-Driven Yacht Search
Client
Luxury yacht sales & charter brokerage, International (via partner agency)
Scope
Full bilingual website build, from partner agency designs, with live listing feed and yacht search
Technology employed
Craft CMS
Algolia
Alpine JS
Show 1 more
Tailwind CSS

The Problem

Listings that update themselves, and a search that never hides a yacht.

If you sell expensive things with long spec sheets, your stock already lives somewhere: a CRM, a back-office system, a spreadsheet. The website is a second copy of it, and second copies drift. The price changes in one place and not the other. The yacht that sold last month is still the first thing on the homepage. Somebody spends Friday afternoon retyping cabin counts.

Your buyers, meanwhile, expect to filter by what matters to them: how much, how long, how many guests, which builder, where she cruises. Then they want to send the shortlist to a partner and pick it up again on their phone that evening.

That was the thinking when 1994, working through a partner agency, rebuilt the website for a brokerage that sells and charters superyachts. The brokerage already ran its own yacht management system. The new website had to treat it as the only source of truth, in two languages, and give buyers a search worth sharing.

Specific challenges

The brokerage's yacht management system had to stay the only place a yacht is ever edited
Every change made there had to reach the website straight away, in two languages, with no overnight sync to wait for
Buyers had to be able to narrow a large, changing stock to a shortlist in seconds and pass it on
The search had to stay fast and simple however much stock was on the market

The solution

The starting point was a rule: the website never owns a yacht. Every listing belongs to the brokerage's own system, and the site only ever reflects it. So instead of the site collecting listings on a timer, the system tells the site the moment something changes. A price tweak, a new set of photos, a yacht marked sold: each is live within moments, in both languages, from one push. If a half-finished update ever arrives, the site declines it and keeps what it had, so a glitch can never quietly empty the listings.

With the listings arriving clean, the search could be built around how buyers think rather than how the data is stored. Price, length and guests are sliders, because a buyer knows "about thirty metres", not 29.4. Length switches between metres and feet with one click, because buyers do. Builder and year narrow the yachts for sale; destinations narrow the charter fleet. Every nudge of a slider updates the results at once, with no page reload and nothing to wait for.

Each search is written into the page's address as the buyer refines it. Send that link to a partner and they see the same shortlist in the same order; come back tomorrow and it is still there. The back button steps through the searches the buyer actually made, not every twitch of a slider.

Above all of that sits one search box that looks across yachts for sale, charter yachts and destinations at once, and ranks the mix by how well each result matched rather than which list it came from.

It runs on Craft CMS, with Algolia doing the searching and Alpine JS and Tailwind CSS on the front end. New yacht in the feed, new yacht on the site. Nobody typed a thing.

Broader uses

The pattern is simple: your stock system stays the source of truth, and the website mirrors it and makes it searchable. Some places we would expect it to pay off:

Property and holiday-let portals fed from a CRM
Aircraft, classic car and plant dealers with spec-led stock
Any brokerage where stock lives in a back-office system and the website must mirror it
Multi-language product catalogues driven by a PIM or ERP

Know someone who'd use this?

Pass it on in one click.