Sailing Saugatuck
A family sailing charter out of Saugatuck, Michigan — #1 outdoor thing to do on TripAdvisor, 440 five-star reviews, with a web presence designed like an Airbnb listing rather than a tour-operator template.
- StatusLive · sailingsaugatuck.com
- ScopeWebsite, branding, booking, video
- TimelineRunning since 2019 · current site launched 2024
- Use25 guests × 4 sails/day · Saugatuck, MI
Full product walkthrough coming soon.
Tour-operator websites mostly look the same. Stock sunset, a gallery, a price list, a booking widget — all generated from whatever template the booking platform hands out. We did not want to look like a tour operator.
The business is a 65-foot schooner out of Saugatuck, Michigan — 25 guests at a time, four sails a day through the summer. A family business, running since 2019, with a smaller charter out of South Haven before this one. There's a real story here — crew, lake, light, a specific harbour town. A templated site flattens all of that into the same sunset-and-price-list everyone else ships. The site had to do the job the brand deserved.
A consumer-brand website with a booking engine and a video reel behind it — all of which I made.
Prototyped in Figma against Airbnb's visual language as the reference. Built on Framer so we could ship a production site in weeks without a bespoke codebase the business didn't need yet. FareHarbor wired at the page layer for bookings — no hand-off to a generic booking domain. Video shot and edited in-house (I have a video production background), so the hero content is the actual schooner, the actual crew, the actual Michigan summer light — not a stock clip. Analytics instrumented end-to-end, because at 25 guests across 4 sails across a short season, a one-percent conversion lift is real money.
Built with- Figma (prototyping)
- Framer (production build)
- FareHarbor (booking engine)
- Self-shot + self-edited video
- GA4 + platform analytics
- Agentic AI inquiry layer (in build)
Preview panels will live here once the full case study is authored.
Three decisions shaped the site more than anything else — and a fourth one is in build.
Design against the category.
Look at ten tour-operator sites in a row and you'll see the same page ten times. Stock sunset at the top, a photo gallery, a price list, a booking widget at the bottom. Nearly all of them were built from templates provided by the booking platform's CMS. They convert, sure — at roughly the same rate as everyone else.
The reference class I set for Sailing Saugatuck was Airbnb, not other charters. Hero imagery that does the work of a thousand bullet points. A typographic hierarchy that treats the experience like a product, not a SKU. The booking engine lives underneath the brand instead of the other way around. When the builder is inside the business — not a contractor being handed a template — that decision is yours to make, and nobody is going to talk you out of it by citing “what the competitors do.”
The conversion engine, assembled in pieces.
A small business doesn't have a web team. It has whoever the owner is willing to become. For this site that meant prototyping in Figma, building in Framer, integrating FareHarbor at the page layer, and shooting the video myself. Each piece is cheap on its own — the effect compounds.
The three decisions below are the ones that made the site do the work. None of them is clever technology. All of them are about putting the right thing in the right place.
- 1 In-house video instead of stock. I shot and edited every frame — the boat, the crew, the lake, the light late in August. Stock clips of “adventure” read as stock in about three seconds; real footage of the actual schooner is the single biggest lever for the “this is a real thing” signal that carries the rest of the page.
- 2 FareHarbor wired at the page, not siloed on /book. Bookings happen where the intent is — inline with the content a visitor is already reading. The category default is to bolt a booking page on and hope people click through. Embedding the widget next to the experience content means the decision and the action live in the same place.
- 3 Analytics as infrastructure, not a Monday-morning report. Every decision about layout, copy, and CTA placement is made against numbers. End-to-end funnel instrumented on day one. At our scale a 1% lift across a season is a meaningful amount of money — which is the argument for treating analytics like wiring rather than reporting.
The next chapter — agentic AI with a human in the middle.
The constraint at a small business isn't the volume of inquiries, it's the voice. Every reply back to a guest has to sound like us — specific about the dock, the safety talk, what to bring, the way the light changes in late August. A generic autoresponder would flatten the brand we just spent years earning, and a typical off-the-shelf chatbot would be worse.
So the next build is an agentic AI trained on our own past correspondence, drafting replies to inquiries. Every draft goes through a human for approval before it sends — literally a person in the middle. Corrections feed back into calibration, and over time the model's first drafts get closer to final. The pattern is <em>agentic but accountable</em>: the operator keeps their voice, the AI takes the grind off the inbox, and the feedback loop tightens every week. That's the shape I think small-business AI has to take to actually work.
Live at sailingsaugatuck.com. #1 outdoor thing to do in Saugatuck on TripAdvisor. 440 five-star Google reviews. Running since 2019 with a crew, a schooner, and four sails a day through a short Michigan summer. The agentic-AI inquiry layer is next on the board.