Saygents is not a self-serve template builder. We cover the intake live on a 20-minute demo call, then a real human launch lead structures your knowledge, tunes the agent to your voice, wires your channels, and runs real-conversation QA — all before your agent ever talks to a customer. Here is exactly what happens between hour zero and hour 72.
Most "AI chatbot" platforms are self-serve — you build the prompt, you wire the integrations, you debug it when it goes sideways. Saygents inverts that. A 20-minute demo call is the only thing on your side of the line.
Every Saygents agent goes through the same six steps. The clock starts the moment we wrap your 20-minute demo call.
The build starts with a 20-minute demo call — not a long form to fill out. We show you a live agent in the first few minutes, then walk through the questions we need to build yours: what you sell, who you sell to, how you price, what hours you work, what makes you different, and the questions your customers ask most often.
No prep on your side. We ask, you answer in plain English, and we capture the same information a sales team would walk a new hire through on day one. That call is the entire intake — there's no separate onboarding form afterwards.
Right after the call, a real human on our team turns the conversation into a structured agent profile — your services, offers, pricing tiers, qualifying questions, escalation rules, tone of voice, and product knowledge.
If anything material is missing — pricing for a service you mentioned but didn't price, an hours-of-operation gap, an unclear booking flow — we email you within hours, not days, with the specific gaps to close. Usually a one-line reply is all it takes.
We build the agent specifically for your business. This is not a template, not a prompt-engineering exercise on top of a generic chatbot, and not a 'configure this dropdown' self-serve flow. The agent is wired to your services, your pricing, your booking rules, and your tone.
The same Sonnet 4.6 model powers every agent on the platform, but the system prompt, knowledge base, escalation logic, and channel-specific behavior are tuned for what your business actually needs to do — qualify a roofing lead, book a med spa consult, route an HVAC emergency.
While the agent is being built, we wire it to whichever channels you've chosen — Web Chat (a 2-minute embed snippet), Instagram DM, and Facebook Messenger. Each channel has its own integration path; we handle all of it for you.
If you're on Solo, you pick one channel to launch with. Duo covers two channels of your choice, and Trio covers all three from day one. You can move up a tier later — it's a configuration change, not a rebuild.
Before your agent ever talks to a customer, our team runs it through a battery of real conversations. We pose as a price-shopping homeowner, a fully qualified buyer, a tire-kicker, an angry customer, and a person trying to break the agent. We confirm it qualifies correctly, books correctly, escalates correctly, and stays in your voice.
Anything that doesn't pass QA gets fixed before you see a single message. By the time we hand off, the agent is conversation-ready — not a starting point you have to refine.
Your agent goes live across the channels you chose. You get login credentials for your customer dashboard, where you can see every conversation in real time, take over any chat, retrain on edge cases, and adjust tone or rules without touching code.
From hour 72 onward, the agent is yours. Updates are included as you grow — adding services, changing prices, adding seasonal offers, or onboarding a new location. You don't pay extra for changes, and you don't wait days for them.
A live agent is not just a chat widget. Here is everything that lands in your tenant on the day you launch.
Wired to your services, pricing, hours, and tone. Powered by Sonnet 4.6, never a generic ChatGPT bolt-on.
Web Chat, Instagram DM, Messenger — pick the ones your customers actually use.
Every conversation from every channel in one screen. Take control on any chat with one click.
Connect Google Calendar, Microsoft 365 / Outlook, Cal.com, or Acuity. The agent offers your actual availability and writes the event inside the conversation. Calendly works as a share-link drop-in.
Every conversation is auto-classified as Booked, Helped, Chatted, Confused, or Failed — so you can see what is working and what needs retraining without reading every transcript.
When the conversation goes outside the agent's training, your team gets pinged with a full thread summary — not just a "user wants to talk" notification.
Every member of your team gets a login. Edit services, prices, hours, and FAQs without filing a change request.
Add a new service. Change a price. Open a new location. We update the agent for you — no per-change fees.
Solo: 500. Duo: 1,000 pooled across 2 agents. Trio: 1,500 pooled across 3 agents. Overage at $50 per 250 — the agent never stops talking.
No. The 20-minute demo call is the intake. We walk through the questions we need to build your agent — live, on the call — so there's no long onboarding form to grind through afterwards. If something material is missing after the call, we email you a quick question; usually a one-line reply closes it.
The call is a conversation, not a checkbox quiz — anything that doesn't fit a standard answer, we dig into right there. If something material comes up later, our launch lead emails you within hours to close the gap. Complex integrations like custom CRMs can extend the timeline up to a week — we tell you that on day one, not at hour 72.
Yes. After internal QA passes, we give you preview access to the dashboard so you can run your own test conversations. Anything you flag is fixed before public launch.
Updates are included for the life of your subscription. Add a new service, change pricing, swap a phone number, retrain on a tough conversation — no change-order fees, no rebuild charges. The dashboard handles most edits without us; bigger changes ship within 12 hours. Mark a request urgent for a faster fix.
A 24-hour build forces a template. The 72 hours is the time it actually takes to do the work — process what we heard on the call, structure your knowledge base, tune the system prompt to your voice, wire your channels, and run real-conversation QA. We would rather ship something correct on day three than something generic on day one.
Three plans. Solo is $150/month for 1 agent on 1 channel with 500 conversations included. Duo is $250/month flat for exactly 2 agents — two channels of your choice — with 1,000 conversations pooled across the tenant ($125/agent). Trio is $300/month flat for exactly 3 agents — one for each channel (Web Chat, Instagram, Messenger) — with 1,500 conversations pooled across the tenant ($100/agent). All three tiers share the same overage rate ($50 per extra 250 conversations). The first month is charged in full at signup — no proration, no refunds — because the 72-hour build happens before you ever serve a conversation. Cancel any month from the dashboard.
Book a 20-minute demo and see a live agent handle real conversations for your business. Then we build yours in 72 hours. No template. No DIY prompt engineering.