Seb
Our own sales agent architecture — objections, pricing, and a calculator, each handled by a specialist.
The problem
A single chatbot trying to handle pricing questions, product objections, and ROI math all in one prompt tends to do all three poorly. Online course and infoproduct sellers needed a sales conversation that stays sharp on every one of those fronts at once.
The solution
Seb is our proven sales agent framework: instead of one agent trying to do everything, it delegates to specialists — a dedicated objection-handling agent and a dedicated ROI-calculator agent, each called as a tool — while Seb itself runs the 5-phase sales conversation, pulls pricing, modules, success stories, and guarantee details from a live knowledge base, and hands off to a human the moment it hits something it can't answer.
- 015-phase sales conversation: a structured flow instead of a free-for-all chat, so every lead gets the same quality of conversation.
- 02Specialist sub-agents: objection handling and ROI calculation are each their own agent, called as a tool — not crammed into one prompt.
- 03Live knowledge base: pricing, modules, success stories, guarantee terms, technical requirements, and competitor comparisons are pulled live, not hardcoded.
- 04Multimodal input: voice notes and images are normalized to text before Seb ever sees them.
- 05Natural multi-part replies: long answers are split into sequential messages, the way a real person texts, instead of one wall of text.
- 06Built-in honesty: a dedicated tool reports knowledge gaps and escalates to a human instead of guessing.
The result
This is the architecture behind our sales-agent client engagements — refined across real deployments into a reusable framework we bring to every new sales agent build.