·Adiamo Studio
Case study: Le Velvet Restaurant, from Google Maps to a live site
How Le Velvet Restaurant’s Google Maps listing in Nantes became a live one-page site — what came from Google, what the owner edited, how long each step took, and what the process does not prove.
Answer first
Le Velvet Restaurant (Nantes) went from a Google Maps URL to a published one-page site by generating a draft from the listing and its reviews, then editing the facts a guest acts on: menu highlights, photos, hours, and the reservation contact. Generation took minutes; editing took the time it takes to be honest about your menu and photos. The result is a URL the venue controls, with the listing’s proof carried onto a page built around one Reserve action.
Le Velvet Restaurant in Nantes is a live example of the Adiamo workflow: start from the Google Maps listing guests already trust, draft a one-page site around reviews and place facts, then edit until the booking path matches how the room works.
This case study is practical, not theatrical. It walks through what we started with, what the draft included, what changed before we called it done, and what this process does not solve.
How this differs from a traditional rebuild
A traditional rebuild starts with a brief, wireframes, and weeks of back-and-forth before anything is public. The Adiamo path starts with a Maps URL and a draft guests could almost use — then concentrates human time on truth: menu, photos, hours, reservation contact. That order matters. Owners often stall on perfect wording while the basics remain wrong. Here the basics arrive first; wording is edited on top.
| Agency rebuild | Maps-first draft | |
|---|---|---|
| First public version | After design sign-off | Minutes after pasting the link |
| Where human time goes | Briefs, wireframes, revisions | Verifying hours, menu, photos, booking channel |
| Source of proof | Written testimonials, if any | Existing Google reviews, selected |
| Risk of stalling | High — waiting on copy and photography | Low — publish once facts are right |
| Ongoing edits | Tickets to the agency | Owner edits in the dashboard |
Input
Google Maps listing + reviews
Output
One-page site, editable
Goal
Trust + one clear Reserve path
What came from Google and what the owner changed
| Element | From the listing | Owner decision | Note |
|---|---|---|---|
| Name, address, phone | Yes | Verified | Must match the door and Google exactly |
| Review quotes | Yes | Selected the specific ones | Vague praise cut; dish- and service-specific lines kept |
| Rating context | Yes | Kept consistent with Maps | No rounding up |
| Hours | Yes | Verified and corrected | The step most owners skip |
| Photos | Partly | Replaced with the venue’s own | Stock-looking images removed |
| Menu highlights | Weak signal | Rewritten by the owner | Reviewers mention old dishes |
| Reservation channel | No | Set by the owner | Only the floor knows what gets answered |
| Language | French listing | French + English | Local language first, English for visitors |
The workflow, step by step
Maps URL
Copied from the listing
Draft
Reviews + facts into a one-pager
Edits
Menu, photos, hours, WhatsApp
Publish
le-velvet-restaurant.adiamo.ai
Share
Profile website field, Instagram
- 1
Generate from the Maps link
The listing URL seeded the structure: hero, proof from reviews, menu highlights, location and hours, and a call to action. The point of generation is speed to a complete skeleton — not a finished brand book. The general method is in how to turn Google reviews into a restaurant website.
- 2
Keep guest language where it is stronger than brochure copy
Specific praise survived editing. Vague filler was cut. The page should sound like people who ate there, not like a template about “culinary journeys”.
- 3
Edit operational truth
Menu highlights, photos, hours, and the reservation channel are owner decisions. This is where most of the real work sat — and where Adiamo acted as an editor, not an autopilot publish button. The checklist is the pass/fail list we used.
- 4
Set the reservation path the floor answers
The Reserve button points at the channel the team monitors during service, following the setup in WhatsApp and email reservations that actually get answered.
- 5
Publish on a clean URL and share it
Once facts were right, the site went live on its subdomain. Then the URL had to appear where guests already look: the profile’s website field, social bios, messages. A live site nobody shares is still a private draft — see why the Maps listing is not enough.
What owners should expect on timing
| Step | Who | Typical effort | What slows it down |
|---|---|---|---|
| Generate draft | Adiamo | Minutes | An incomplete Google listing |
| Verify facts | Owner | Under an hour | Unsure hours or multiple phone numbers |
| Menu highlights | Owner | Under an hour | Menu not yet decided for the season |
| Photos | Owner | Minutes to hours | Waiting for a professional shoot |
| Reservation channel | Owner | Minutes | No one assigned to answer |
| Publish and share | Owner | Minutes | Forgetting the profile website field |
What this case study does and does not prove
It shows
- A Maps-first draft can reach a credible public URL without an agency timeline.
- Existing reviews make a stronger proof layer than written testimonials.
- Most real work is verifying facts, not writing copy.
- A single Reserve path can be wired to the channel the floor answers.
It does not show
- That every cuisine, city, or multi-location group should use the same layout forever.
- Booking numbers — we are not publishing conversion figures for a friend’s venue.
- That Google Business Profile work is optional; the listing still needs accurate data.
- That a site stays good without weekly edits.
Frequently asked questions
Is Le Velvet a paid customer case study?
It is a live example site built with friends of the product to show real output. Treat it as a product demonstration with a real venue, not a guarantee of identical results for every listing.
Will my site look identical to Le Velvet?
Structure is similar; content, photos, language, and reservation channel follow your venue. The point is the workflow, not cloning one restaurant’s identity.
Can I change the site after publishing?
Yes. Edit menu, copy, images, and contacts as the room changes. Publishing is a starting line, not a freeze.
Do I need perfect photos before generating?
No. Generate first, then replace weak images. Waiting for a full photo shoot is how sites stay unpublished for months.
Why French and English?
Local guests read French; visitors and hotel referrals often read English. The site keeps the local language first and offers English without a separate build. The decision behind that is in when a restaurant website needs a second language.
How do I start the same workflow?
Paste your Google Maps link into Adiamo, create an account when prompted, edit the draft, and publish when hours and booking contacts are correct.