Summary
The site is structurally sound and search engines can read it. The problems are maintenance: broken links, draft text left in production, and business details missing from the parts of the page machines read.
Most of it is small and quick. Two items need decisions from the winery before anything is published, and the online store is a separate matter covered at the end.
The complete list
Thirty-six items, each verified on the live site. Filter by seriousness or by the part of the business it touches. The four with the largest effect are explained in detail below.
In detail
Evidence for the items with the largest effect on people reaching the tasting room, captured from the live site.
The most important button on the homepage — the one that says Plan Your Visit — does not lead to the visit page.
It points to an address that was retired at some point: /visiting-us.
That page is gone. What makes this hard to spot is that it doesn't fail loudly —
there's no error message, no broken-page screen. Squarespace quietly serves the
homepage again.
So a visitor clicks Plan Your Visit, the screen changes, and they're looking at the homepage they just left. Most people assume they misclicked, try once more, and give up.
Homepage content, served under a 404
There's a detail here that's almost too neat. On that same screen, the Winery Hours panel reads:
Our Retail Shop is now open 7 days, 11-5 for purchases. Please check the Visiting Us page on this website for any closings.
The site is directing people to the page whose address is broken. Anyone following that instruction ends up in the same loop.
The black bar at the top of every page is a milder version of the same thing. The words Plan your visit are correctly linked and work. The bar behind them is a second, separate link pointing at an address that no longer exists — so clicking the words is fine, and clicking anywhere else along the bar isn't.
Every visitor who does exactly what the site asks — click the main button — fails to reach hours, directions, or booking. Google has had the retired address indexed against searches for tasting room hours, so some people never even start on the homepage.
Draft text made it into production and stayed there.
The Events page — a page tied directly to people turning up and spending money — has a note-to-self where its headline should be. It's set in the same elegant serif as everything else, in white, over the photograph, which is what makes it read as deliberate for a half-second before it doesn't.
Live on the site today
Elsewhere, a page is still carrying the Latin filler text Squarespace ships with a blank template — "It all started when… Vivamus a ante congue…" — and it is listed in the site's own index of pages for search engines to find. Three more abandoned draft pages sit alongside it.
The Wines menu has a related problem. It points at an old copy of the entire wine catalogue, left over from a previous version of the site — roughly 21,000 words duplicating tasting notes that already live on their own pages. The real wine section, ninety-nine individual pages built properly, isn't linked from the homepage at all.
Placeholder text on a page about events is the kind of thing a first-time visitor notices and quietly downgrades you for. Search engines treat unfinished pages as a signal about the whole site — and the wine catalogue currently competes against its own duplicate.
Several things that work perfectly for visitors don't exist at all as far as search engines are concerned.
Behind every website is a small machine-readable card describing the business — name, address, phone, hours, social accounts. It's what Google reads to decide whether to show you in the map results, and what AI assistants read when someone asks where to go wine tasting in the Hamptons.
On this site that card exists, and the business name field is empty. So are the social accounts, the opening hours, and the map coordinates — despite all of that information being visible elsewhere on the page for humans to read.
The site's own business record, read live
"@type" : "LocalBusiness"
"name" : "" ← blank
"telephone": absent
"sameAs" : [] ← no social accounts
"openingHours": absent
"geo" : absent
The same pattern repeats with the booking button. There is one — Tock, working, on the visit page. But it's drawn onto the page by a script after loading, which means search engines and AI assistants don't see a booking option here at all. That is very likely why searches like book a tasting at Channing Daughters surface tour operators and travel marketplaces instead of the winery's own page.
Works for people · invisible to search
The tasting menu has a version of this too — the page is live and linked, but it was never added to the site's index, so it isn't eligible to appear in search at all. Four other pages in the navigation are in the same position.
One smaller thing worth mentioning because it's cheap to fix: most pages have no preview image attached, so links shared to Instagram, Facebook or a text message arrive as plain grey boxes. On the Events page the preview text is currently the photographer's note — "banner image of tasting room floor with stylized footprints."
And on a phone, the first screen is brand copy about grape varieties. No hours, no address, no booking — those come after scrolling.


A blank business record keeps you out of the map results that decide wineries near me. An invisible booking button hands that search to Expedia and the tour companies. Together they mean the winery is competing for its own name against people reselling it.
Worth raising, but honestly its own project — and not one to start this month.
First, the reassuring part: the online store and the wine club signup work. I completed a checkout through to the payment step to be certain. Nothing about this is costing sales today.
The store runs on Nexternal, a separate platform from the main website, on its own web address. It's protected by bot-blocking software, which does its job well — it also blocks Google. Every wine listing, every price, and the wine club signup page are unreadable to search engines.
Sixty links across the website point into that store. All of that recommendation value stops at the door.


There's a presentation gap too. A customer who clicks Shop arrives somewhere that looks like a different company — different typefaces, different colours, none of the character of the main site. That's not a criticism of the platform, which is built for wine compliance and shipping rules rather than looks.
The practical first step isn't a rebuild. It's a conversation with Nexternal about the bot-blocking configuration, which is theirs to adjust rather than yours. That alone would make the wines findable. Anything beyond it — closer visual alignment, or eventually consolidating the store — is a genuine project with a budget attached, and it should be decided on its own merits rather than folded into this.
Someone searching for a specific bottle by name won't find it. The press coverage and links the winery has earned over the years don't reach the store, so it starts from nothing every time.
Before anything gets published
A few facts need confirming from inside the business. I'd rather ask than guess and put something wrong on the internet under your name.
The site currently gives three different answers. The homepage says the retail shop is open 7 days, 11–5. The visit page says the tasting room is open daily 11–6, staying open until dark on Friday and Saturday. If the shop and the tasting room genuinely keep different hours, that's worth saying plainly rather than leaving people to work out which applies to them.
There's no founding date anywhere in the site's text, which is why nothing can answer when was Channing Daughters founded. The site references Walter's first planting in 1982 and Larry Perrine joining in 1996. Which date do you want to be the one people repeat?
The address used across the site includes the P.O. Box, which confuses mapping software. I'd like the tasting room entrance pinned deliberately rather than approximated from a postal address.
If a move off Nexternal is being discussed, some of the recommendations here change shape and a few become pointless. Worth knowing before work starts.
The listing that appears in maps and local results couldn't be checked from outside. It's frequently the single biggest factor in local visibility, and it's usually quick to improve once someone can get into it.
Proposed plan
Repairs happen immediately. The larger pieces follow in an order that keeps the site earning while the next stage is built.
I take care of the short items directly — the visit path, the placeholder and draft pages, the wine menu, the business record, the sitemap gaps and the link previews. Roughly two thirds of the list closes here.
By the end of it The main call to action works, the site stops looking unfinished, and you're eligible for local results.
Design work on the new marketing site, alongside the content the current one is missing — the visit page, private events, and the wine pages. This is also when the outstanding questions get answered and the facts settled.
By the end of it A site ready to build against, with the thin pages properly written.
Build and launch the new marketing site, carrying over everything fixed in phase one so none of it is lost in the move. Redirects handled so existing search positions transfer rather than reset.
By the end of it A site that reflects the winery, with the repairs built in rather than bolted on.
Two routes, and the recommendation depends on what you want the point of sale and wine club to do:
Move off Nexternal to a platform that integrates properly with the POS, wine club and mailing list, so membership and purchases live in one place.
Keep their backend and work with their developers on access to it, while I build a new storefront that matches the site — so the shop stops looking like a different company.
By the end of it A costed recommendation with a clear route, rather than a decision made by default.
Phase one is short enough that it's worth doing regardless of what's decided about the rest.
For the record
All 427 published pages were inventoried. Twelve page types were pulled apart in full — homepage, visit, events, an individual event, a wine page, the wine club, private events, contact, a staff profile, a recipe, a press article and the events index.
Every finding was confirmed against the live site on the day of writing, and the screenshots are from that session. Where something depended on how a page behaves for a real person, I opened it in a browser and used it rather than trusting an automated reading.
Page speed figures come from simulated testing rather than measurements of your actual visitors, so treat them as directional. Review counts, map listing details and the full picture of who links to the site need access I don't have — those are flagged as questions rather than findings.
An initial automated pass produced several findings that didn't survive checking — it reported no booking system, no enquiry form, and a pop-up on every page. All three were wrong. They've been corrected here rather than repeated.
A note on the overall score you may have seen quoted elsewhere: a single number out of a hundred is a reasonable temperature reading, but it isn't what to act on. The specific broken things above are.