Atelier Hair Studio: Winning Bookings Before the Click

Atelier Hair Studio is a four-chair salon on Mare Street in Hackney, East London. It has been open since 2019. The owner, Leah Okonkwo, built a loyal local following through word of mouth and Instagram, but by late 2025, she noticed something odd. New clients were arriving and saying the same thing: "An AI recommended you."

She hadn't done anything to make that happen. Or so she thought. What had actually changed was that Vendoora had structured her services for agentic commerce, and AI assistants had started reading them.

The problem: invisible to machines

Before working with Vendoora, Atelier's online presence was typical of a small salon. A single-page website with a few photos, a list of services with no prices, an Instagram grid, and a Google Business Profile that hadn't been updated since the pandemic.

For human visitors, this was fine. The photos looked good, the reviews were strong, and regulars already knew the drill. But for an AI assistant trying to answer a query like "recommend a good hair salon near Hackney for balayage", there was almost nothing to work with.

No structured service descriptions. No pricing signals. No FAQs. No machine-readable location data beyond the Google listing. No booking path an agent could follow or describe. The salon was effectively invisible to the agentic commerce discovery layer.

What was structured, and why it mattered

Vendoora's services team worked with Leah to build a structured profile for Atelier. This wasn't a website redesign. It was a data exercise, making what the salon already offered legible to reasoning systems.

The key elements:

• Service descriptions with attributes. Each service, cuts, colour, treatments, styling, was described in plain, specific language with the details an agent needs to match a query: what the service involves, who it suits, typical duration and approximate price. "Balayage (from £95, 2.5 hours, includes consultation and toner)" tells an agent vastly more than "Colour Services" with no further detail.
• Pricing signals. Leah was initially reluctant to publish prices, a common concern among salon owners. The compromise was "from" pricing: enough for an agent to filter by budget without committing to a fixed menu. This alone made Atelier viable as a recommendation where salons showing no pricing data were skipped entirely.
• Location and coverage data. A clean address, postcode, coordinates and transport references (nearest station, bus routes), structured so an agent answering "near Hackney" or "easy to reach from Liverpool Street" could reason about proximity without guessing.
• FAQs. Six questions covering the topics AI assistants are most commonly asked about salons: cancellation policy, whether consultations are free, suitability for curly or textured hair, parking, accessibility and whether the salon is vegan-friendly. These weren't written for the website. They were written for agents, and marked up so agents could read them directly.
• Booking path. A clear, described route from discovery to appointment: "Book online via the Atelier website or call 020 XXXX XXXX. Same-week availability is common on Tuesdays and Wednesdays." An agent can relay this to the user verbatim. Without it, the recommendation dead-ends.

The result: recommended before the search

Within six weeks of the structured profile going live, Leah began seeing a pattern. New clients mentioned they had asked an AI assistant, usually ChatGPT or Google's Gemini, for salon recommendations in East London, and Atelier had appeared in the response. Not as a link in a list. As a named recommendation with reasoning: price range, location, suitability for the client's hair type, and how to book.

The numbers were modest but directionally clear. Atelier went from zero AI-referred bookings to roughly eight per month within two months. For a four-chair salon, that is meaningful, it represents new clients arriving without any advertising spend, discovered through the agentic commerce channel that didn't exist a year earlier.

More telling was what competitors were doing: nothing. Most independent salons in the area had no structured service data at all. The bar for visibility wasn't high, it was simply that nobody had cleared it. Atelier did, and the result was disproportionate.

What this tells other service businesses

Atelier's story is not about technology. Leah didn't install anything, learn a new platform or change how she runs her salon. The shift was entirely in how her business was described to machines.

Three lessons apply broadly:

Structured data is the new shopfront. For product retailers, this means clean feeds. For service businesses, it means described, attributed, machine-readable service profiles. If an AI assistant cannot parse what you offer, your pricing and your location, you will not be recommended, regardless of how good your reviews are.

Early movers win disproportionately. In a category where almost nobody is prepared for agentic commerce, the first business to provide structured data captures most of the AI-referred demand. This advantage is temporary, competitors will catch up, but the window is open now and it is wide.

You don't need to build this yourself. Leah spent a single afternoon on the phone with Vendoora's team. The structuring, markup and optimisation were handled for her. This is the model Vendoora is scaling for service businesses: the provider focuses on the work; we make the work legible to AI assistants.

If your business depends on local discovery, and you are not yet structured for agentic commerce, the question is not whether to start. It is how long you can afford to wait while the salon, trainer or tradesperson down the road gets there first. Read the complete guide to agentic commerce to understand the full landscape, or see how the same principles apply to product retailers in the future of marketplace discovery.

Read next