How to Build an AI Receptionist with Synthflow (Beginner Tutorial)
Beginner step-by-step tutorial to build an AI receptionist on Synthflow (no-code): account setup, agent build, system prompt, calendar booking, phone number, testing, going live, real 2026 costs, and the white-label reselling angle for agencies.
Why Build on Synthflow Instead of Retell or Vapi
If you have read our Retell AI tutorial or the Vapi tutorial, you have seen the two ends of the spectrum. Retell is the fast, bundled, one-price path. Vapi is the modular, bring-your-own-provider path for builders who want control. Synthflow sits in a useful third spot, and for a lot of aspiring agency owners it is the most comfortable place to start.
Synthflow is a no-code AI voice agent builder aimed squarely at non-technical users and small businesses. The whole experience happens in a visual dashboard. You build the agent, write the prompt or design a conversation flow, attach a phone number, and go live without touching an API or a server. What makes it interesting for resellers specifically is that its agency layer is one of the most complete in the market: client sub-accounts, per-feature visibility toggles, custom branding, custom domains, and built-in Stripe rebilling are all native rather than bolted on. Our white-label platform guide compares that layer against Retell and Vapi for resellers.
We will use a dental office as the running example, the same one we used in the Retell and Vapi walkthroughs, so you can compare the three builds directly. If you want the deeper feature-by-feature breakdown first, the Synthflow review covers it.
What You Need Before You Start
You can gather all of this in about 20 minutes:
- A Synthflow account. It is free to create an account and build your agent; usage charges apply only when calls actually happen.
- A phone number. Synthflow can provision one for you (about $1.50 each, billed through its managed Twilio), or you can bring your own Twilio account and route through it.
- A calendar for booking, such as Cal.com or Google Calendar, which connects through Synthflow's integrations.
- No separate LLM key is required on the standard pay-as-you-go path. Synthflow lets you pick the model in the dashboard and bills the model usage as part of your per-minute cost.
That is the entire stack. No code, no middleware. Everything below happens inside the Synthflow dashboard.
Step 1: Create the Agent
Inside Synthflow, go to the Agents tab and click + New Agent. Choose Voice Agent, then pick the agent type:
- Inbound handles incoming calls. This is your receptionist. It is also the compliance-safe choice, because inbound calls do not trigger the consent rules that outbound dialing does.
- Outbound initiates calls for things like reminders and lead follow-up. Useful later, but not where you start.
- Voice Widget is an embeddable agent for a website. Skip it for a phone receptionist.
Pick Inbound. Then you choose how to define the agent's behavior, and Synthflow gives you two paths:
| Build method | Best for | When to use it |
|---|---|---|
| Prompt Builder | Lightweight setups where one good prompt defines everything | Your first build, and most single-purpose receptionists |
| Flow Designer | Structured, multi-step conversations with explicit branching and reusable logic | Multi-department routing, strict scripts, predictable paths |
For a first receptionist, start with the Prompt Builder. It is faster, and a well-written prompt handles a dental front desk just fine. Move to the Flow Designer later when a client needs hard branching, like "press-or-say routing" to three departments. Name the agent something you can find across many clients, such as "Bright Smiles Dental — Reception." When you run ten clients in one account, naming discipline saves you.
Step 2: Pick the Voice and Model
Synthflow lets you choose the voice from a curated set of recommended ElevenLabs voices, with the full ElevenLabs library available, and Cartesia is also offered as a provider for clear, conversational, low-latency voices. For a receptionist, pick a natural everyday voice and do not reach for voice cloning until a client specifically asks and pays for it.
On the model side, choose a fast, inexpensive option for high call volume. Synthflow exposes a range of LLM choices in the dashboard, and on the pay-as-you-go plan the lighter models cost a fraction of a cent per minute more than the heavier ones. Start light; upgrade only if the agent's reasoning is visibly weak in your test calls.
Step 3: Write the System Prompt
This is where 90 percent of the quality lives, and it is identical in importance whether you build on Synthflow, Retell, or Vapi. A good receptionist prompt has five parts: identity, scope, knowledge, behavior rules, and a booking instruction.
Identity. You are the front desk assistant for Bright Smiles Dental in Austin, Texas. You are warm, brief, and efficient. If asked, you confirm you are an AI assistant for the practice.
Scope. You answer questions about hours, location, services, insurance, and new-patient onboarding. You book, reschedule, and cancel appointments. You do not give medical or clinical advice.
Knowledge. Hours: Monday to Friday, 8am to 5pm. New patient exams are 297 dollars or covered by most PPO plans. We do not currently accept Medicaid. Parking is free in the rear lot.
Behavior rules. Keep replies under two sentences. Ask one question at a time. If the caller describes a dental emergency with severe pain or bleeding, say you will connect them to staff and trigger a transfer. Never invent prices or insurance details. If you do not know, offer to take a message.
Booking instruction. When the caller wants an appointment, collect their full name, phone number, and preferred day, then check availability and confirm.
Keep the whole prompt under 500 words. Long prompts slow the agent down and make it hallucinate. Tight scope plus real knowledge beats a wall of text every time. This scaffold is the single biggest time sink when you onboard clients, which is exactly why a ready-made prompt library and setup blueprint pays for itself on the first build. Our prompt templates guide has more ready-to-edit examples.
For richer fact handling, Synthflow also supports a knowledge base: you can import help-center articles and documents so the agent retrieves answers from your client's real content instead of relying on the prompt alone. For a simple dental front desk the prompt is usually enough; lean on the knowledge base when a client has a long service menu or detailed policies.
Step 4: Set the Greeting and Interruption Behavior
Set the first line the agent speaks. Disclose the business and invite a request in one breath:
Thank you for calling Bright Smiles Dental, this is the front desk assistant — how can I help you today?
That opener sounds professional and doubles as a light, upfront disclosure that the caller is talking to an assistant — good practice everywhere, and the kind of transparency California's bot-disclosure rules expect when a bot is used in sales or transactions. For inbound calls it is a friendly disclosure, not a legal warning. For the full picture on disclosure and consent, read our TCPA compliance guide.
Then tune the speaking behavior so the agent stops talking the instant the caller jumps in, and keep the response delay short so there is no awkward pause. A receptionist that talks over people or hangs for two seconds sounds robotic immediately. Leave these near default at first, then tighten after your test calls reveal how it actually feels on the phone.
Step 5: Connect Calendar Booking
This is the feature that turns a toy into a paid service. A receptionist that only answers questions is worth maybe 99 dollars a month. One that books appointments straight into the client's calendar is worth 297 to 497.
In Synthflow, calendar booking is wired through an action or integration on the agent:
- In Cal.com, create an API key and grab the event type ID for "New Patient Exam."
- In Synthflow, connect the calendar integration and add your key.
- Tell the agent in the prompt when to use it, for example after collecting name, phone, and preferred time.
- Add an availability check so the agent never offers a slot that is already full.
Now when a caller says "can I come in Thursday morning," the agent checks real availability, offers two open slots, and writes the booking. That live booking is the demo that closes clients. Our guide on how to demo an AI receptionist and close the client walks through turning that moment into a signed retainer.
Step 6: Attach a Phone Number and Test
You can have Synthflow provision a number for you, or bring your own Twilio number and route it in. For a client deployment, pick a local number that matches their area code; callers trust local numbers. Assign the number to your inbound agent.
Before you ever go live, call the number yourself and run these scenarios:
- A normal booking ("I want a cleaning next week")
- A pricing question ("how much is a new patient exam")
- An out-of-scope question ("can you tell me if my tooth needs a root canal")
- An interruption (start talking over the agent mid-sentence)
- An emergency phrase ("my mouth is bleeding and I'm in a lot of pain")
You are listening for three things: low latency, no invented facts, and a clean handoff or message when the agent hits its limits. Open the call log after each test. Every transcript shows you where the prompt is weak. Expect to revise two or three times. The first version is never the deployed version.
Step 7: Go Live and Monitor
Point the client's existing business line to the Synthflow number using call forwarding, or set the agent to handle overflow and after-hours only if the client is nervous. After-hours-only is a great low-risk way to land a skeptical first client.
Watch the first 20 to 30 calls in the call log. Confirm every booking action returns a success the agent can read back to the caller, and that no transcript shows the agent guessing on a price or an insurance question. Within the first week you will have a tight, accurate agent.
What This Actually Costs You to Run
Synthflow's standard plan is now pay-as-you-go: free to build, and you pay for usage when calls happen. The per-minute cost is assembled from a few line items rather than one bundled rate, so it pays to understand the parts. Based on Synthflow's published pricing, a typical configuration lands in the $0.15 to $0.24 per minute range, built from a voice engine fee, the model you chose, and telephony (lower if you bring your own Twilio). Five concurrent calls are included, with additional concurrency available as an add-on, and provisioned numbers run about $1.50 each.
Here is the honest math, per client, per month, at roughly 500 minutes of calls:
| Item | Cost |
|---|---|
| Synthflow usage (about $0.15 to $0.24 per minute, all-in) | about $75 to $120 |
| Phone number | about $1.50 |
| Cal.com | Free tier |
| Your cost | roughly $77 to $122 |
Charge the client 297 dollars a month and your gross margin is comfortably above 50 percent even at the high end of usage, and well above 70 percent for a typical 250 to 350 minute month. Synthflow's per-minute cost runs higher than Retell's bundled rate or Vapi's stacked rate, so price with the upper figure in mind and you will not get surprised. For the full cross-platform breakdown, see our AI receptionist cost guide and the page on how to price AI receptionist services.
Common Mistakes That Sink Beginner Synthflow Builds
- Pricing off the wrong per-minute number. Synthflow's usage stacks a voice fee, the model, and telephony. Quote off the higher all-in figure, not the headline voice-engine rate.
- Bloated prompts. Over 500 words and the agent slows down and starts inventing answers. Cut ruthlessly, and push detailed facts into the knowledge base instead.
- No "I don't know" path. Always give the agent a fallback (take a message, offer a callback) so it never guesses on prices or insurance.
- Skipping test calls. Going live without running edge cases is how you lose a client in week one.
- Premium voice you don't need. A recommended ElevenLabs or Cartesia voice is plenty. Do not pay for cloning until a client asks.
- Outbound calling without consent. Inbound is the safe zone. Outbound triggers TCPA rules, so do not freelance there.
The White-Label and Reselling Angle
This is where Synthflow earns its place for agency builders. Its agency dashboard lets you create and fully control client sub-accounts, enable or disable specific features per client, apply your own branding and a custom domain, and set your own prices with built-in Stripe rebilling. In plain terms: the client logs into something that looks like your software, on your domain, and you charge them directly while Synthflow runs invisibly underneath.
Be honest with yourself about the cost ladder. Synthflow's full White-Label and Reseller Toolkit is a substantial monthly add-on (publicly listed at $2,000 per month at the time of writing), with white-label included at the enterprise tier. That is not a day-one expense. The right sequence is: land two or three paying clients on the standard pay-as-you-go plan first, prove the retainer sticks, then turn on white-label once the recurring revenue clearly covers it. Adding the branded layer before you have clients is a classic way to burn cash before you have proven the offer. For the reseller economics across every platform, the white-label platform guide lays it out.
Synthflow, Retell, or Vapi — Which Should You Resell?
All three end at the same place: a working AI receptionist you charge 297 to 497 a month for. The difference is the path and the agency layer.
- Choose Retell for the fastest no-code build and a bundled price that is easy to quote. Start with the Retell tutorial.
- Choose Vapi if you have light dev capability and want to swap providers and build custom infrastructure. See the Vapi tutorial.
- Choose Synthflow if you want a no-code build with the most complete native white-label and sub-account system, and you plan to put your own brand in front of clients as you scale.
Where to Go From Here
You now have a working AI receptionist on Synthflow and the real cost math to price it. The slow part is doing it again for the next client: rewriting prompts, rebuilding the booking flow, and packaging the offer so a local business says yes.
If you want the shortcut, the AI Receptionist Agency Launch System gives you the done-for-you version of this build: guides covering client acquisition, pricing, the sales call, and onboarding, plus the templates you reuse on every deployment so your second build takes 20 minutes instead of an hour. Pick a vertical to start with, like dental or home services, build one agent, demo the live booking, and charge for it. The platform is ready. The only thing between you and your first paying client is the offer.
Related guides
- How to Build an AI Receptionist with Retell AI (Beginner Tutorial)
Beginner Retell AI build tutorial for compare.getneurobyte.com blog — agency/freelancer audience, matches existing data-forward no-hype voice, real 2026 platform numbers, soft CTA to /start, /shop, /tools/retell-ai and /tools/claude-prompt-library. No backticks, no dollar-brace. Body-only markdown.
- How to Build an AI Receptionist with Vapi (Beginner Tutorial)
Beginner Vapi (vapi.ai) build tutorial for the compare.getneurobyte.com blog — agency/freelancer audience, honest 2026 platform numbers, the bring-your-own-provider stack explained, calendar booking via tools, and a soft CTA to /start, /shop, and /tools. Pairs with the Retell tutorial.
- White-Label AI Receptionist Platforms: The 2026 Agency Buyer's Guide
SEO buyer's-guide blog post comparing white-label/reseller fitness of Retell, VAPI, Synthflow, and Bland for agencies, with reseller economics and a soft CTA to /start, /shop, and /tools.
Ready to launch your own AI receptionist agency?
The AI Receptionist Agency Launch System gives you the done-for-you toolkit, templates, and support to land your first paying client — backed by a 60-day money-back guarantee.
See the Launch System →The NeuroByte Signal
Get weekly AI tools intelligence, new comparisons, and revenue strategies for agencies. Every Friday.
Subscribe Free →