Status : ActiveLat 45.5231Lng 122.6765Ref BS_RECORD
2.0.26London / Remote
← The LedgerLEDGER-010
DeclassifiedStandaloneProducta multi-tenant voice SaaS

Answers the phone and books the appointment

A voice agent that books the appointment

A voice agent answers the phone for small service businesses, holds a natural conversation, and books the appointment — one engine quietly serving many businesses at once.

The problem

For a service business, a missed call is usually a lost job — but staffing the phone reliably, including nights and weekends, costs more than small operators can justify.

The solution

A voice agent that answers the call, holds a natural conversation, and books the appointment — with one engine serving many businesses at once.

ClaudeVoice (speech-to-speech)TelephonyPython
BeforeAfter
A missed callusually a lost job — the caller moves onanswered live and turned into a booking
Phone coveragea receptionist or answering service per businessone shared engine serving many at once
Evenings and weekendsvoicemail, and the work goes elsewhereanswered the same as any other hour
Cost to cover the phonea fixed salary or service fee per locationa shared system that scales across all of them
What the caller experiencesa recording, or no answera natural conversation that books their slot
The delta

Phone coverage stops being a cost each business carries alone and becomes a shared capability. The same engine answers for any number of businesses at the same time, turning what used to be a receptionist's salary per location into one system that scales across all of them — and never sleeps.

What I built

A multi-tenant voice agent for small-business appointment booking. "Multi-tenant" means one shared system serves many businesses at once — each set up as its own separate "tenant" with its own configuration — rather than building a new system for each.

  • Answers the phone. It picks up live calls at any hour, for whichever business the number belongs to.
  • Holds a real conversation. It listens and speaks back naturally — "speech-to-speech," meaning it works in spoken voice both ways rather than making the caller press buttons or wait on a recording.
  • Understands what the caller needs. It works out the request and asks only the questions required to schedule it.
  • Books the appointment. The call ends with a confirmed slot, not a promise to call back.
  • Serves many businesses at once. The same engine handles all of this across multiple operators simultaneously, each with their own setup.

The pieces under it: telephony (the phone-line plumbing that connects calls in and out), the voice layer that turns speech into understanding and back into speech, and Claude — the AI — running the conversation and the booking logic.

Why it matters

Every call gets answered and turned into a booking — including the evening and weekend calls that small operators used to lose entirely. For a business where the phone is the funnel, that's the difference between winning the job and watching it go to the next name on the list, now available around the clock.

Normally, "answer every call and book it, day and night" means a receptionist or an answering service for each business — a fixed cost per location. Here it's one engine serving all of them at once, which turns reliable phone coverage from a luxury small operators couldn't justify into something they can simply switch on.

The hard part

Voice is unforgiving. A caller hears every pause, talks over the agent, and loses patience fast — so the conversation has to feel natural in real time, not like a phone tree. The other hard part is the booking itself: it's only worth anything if the appointment actually lands, with the right details, in the right business's calendar. So the work went into making the spoken exchange feel human despite the split-second timing voice demands, and into a booking handoff that doesn't drop or scramble information. Get either wrong and the caller hangs up; get both right and they don't realize they weren't talking to a person.

The bottom line

Voice agents have crossed from novelty to genuinely usable, and that flips the economics of answering the phone: a fixed cost per business becomes one engine serving all of them, around the clock. The small operator who used to lose every after-hours call now answers all of them.

Where the edges are

The hard part is making the conversation feel natural in real time, and never dropping a detail when it hands the booking off. That's what earns a caller's trust — and where the real work is.

Start a conversation