What problem does it solve?
In small businesses that take work by phone, calls go unanswered while staff are busy with customers, after hours or at peak times. The caller often does not call back and goes to a competitor instead. Not every business can afford to hire a receptionist, and call center services are expensive and inflexible at small scale. On top of that, appointments and notes taken over the phone end up on paper or in someone's memory and cannot be followed up later. The owner has no idea why each call came in or how many turned into business. The result is lost revenue and a damaged customer experience.
Target customers
- Dental, veterinary and aesthetic clinics
- Hair salons, barbershops and beauty salons
- Auto repair shops, technical services and repair workshops
- Law, accounting and consulting offices
- Restaurants and food delivery businesses
- Real estate agencies and small hotels
How the product works
The business forwards its existing number to a virtual PBX during busy hours or outside working hours. The AI assistant answers the call: it understands speech through speech-to-text, generates replies with a language model trained on the business's service list, opening hours and FAQs, and speaks in a natural voice through text-to-speech. The assistant checks free slots in the calendar and books appointments, escalates urgent matters to a staff member, and records details from callers who want to leave a message. Every call is summarized and sent to the owner in the dashboard and via WhatsApp/email notifications. Scenarios, voice selection and transfer rules are configured in the dashboard without code.
What the first version (MVP) needs
- Virtual number and forwarding: Route calls from the existing line to the assistant
- Voice dialogue: Turkish conversation via speech-to-text, language model and text-to-speech
- Business knowledge base: Enter services, pricing policy, hours and FAQs
- Appointment booking: Free-slot checks and booking via calendar integration
- Messages and transfers: Take notes and connect urgent calls to staff
- Call summary: Transcript, summary and notification for every call
- Admin dashboard: Call history, appointments and scenario settings
- KVKK module: Privacy notice announcement at call start and recording retention settings
Revenue model
- Monthly subscription: Tiered plans based on call volume
- Extra minute/call bundles: Usage-based fees for overage
- Setup and scenario design: Custom knowledge base and voice configuration service
- Industry white label: Licensing to software resellers and vertical software vendors
Why now?
Speech recognition and synthetic voice technologies have reached the point where they can deliver a natural, low-latency experience in Turkish, and language model costs now make a subscription affordable for small businesses. Most SMEs in Turkey still win work by phone and cannot carry the cost of a receptionist. Similar services are starting to spread in Europe, but local solutions that understand Turkish accents, local scheduling habits and industry jargon are limited. Starting with ready-made scenarios for specific industries allows a fast market entry.
End-to-end roadmap
- Validate the idea (1 week): Interviews with target customers, competitor and pricing research; we sharpen the idea together in a free pre-analysis.
- Scope and design (1–2 weeks): MVP feature list, screen designs and technical architecture; written scope and quote.
- Build the MVP (8–12 weeks): Staged development with weekly demos and a test environment; the source code belongs to you.
- Pilot (2–4 weeks): Real use with the first customers, feedback and fixes.
- Launch and grow (ongoing): App Store and Google Play release, website and SEO, payments and subscriptions, new modules.
Timelines are estimates; they are fixed in a written quote once the scope is clear.
Risks and how to handle them
- Misunderstandings and wrong answers: Restrict the assistant to the business's knowledge base and have it transfer to a human or take a message when unsure
- KVKK and call recording: Play a privacy notice at the start of the call, set retention periods for recordings and transcripts, and process data in Turkey or with appropriate safeguards
- Voice latency and cost: Use streaming speech-to-text and fast models, cap long calls and price per minute
Technologies
Frequently asked questions
How well does an AI phone receptionist understand Turkish?
Current speech-to-text models understand everyday Turkish speech well; still, confirmation questions and a human-transfer rule should be added for noisy environments and proper names.
Does the business have to change its existing phone number?
No. The existing line is forwarded through the carrier to the virtual number when busy, unanswered or outside working hours.
Is recording calls a problem under KVKK, the Turkish data protection law?
Callers must be informed at the start of the call about the recording and the purpose of processing, a retention period must be set, and explicit consent obtained where required. Getting advice from a legal counsel before going live is recommended.
Can the assistant answer every question?
The assistant should be designed to answer only with information the business has defined. On topics it does not know, it takes a message or transfers the call to a staff member.