AI-per-step · highest quality
The default mode. The AI drafts every message fresh at send time · personalized to the prospect and injected with the agent's market learnings. One LLM call per send, ideal for audiences that need real personalization.
A sequence is your multi-step outbound cadence · the ordered list of touches an agent works through on each prospect. Every sequence runs in one of two modes. AI-per-step (the default) writes each message fresh at send time, personalized per prospect and infused with what the agent has learned about your market. Template lets you write each step once with {{variables}} and merges them at send time with zero LLM cost per first-touch. Either way, inbound replies always route through the AI auto-reply engine, so the conversation stays smart no matter how the first message was sent.
The default mode. The AI drafts every message fresh at send time · personalized to the prospect and injected with the agent's market learnings. One LLM call per send, ideal for audiences that need real personalization.
Write each step body once with {{variables}} like {{firstName}} and {{company}}. At send time it is pure variable-merge · no LLM call, no per-send cost. Built for high-volume mass outreach.
Template only removes the LLM cost from first-contact sends · not from conversations. Every reply, on every channel, always runs through the AI auto-reply engine. Your cadence stays cheap to launch and smart to answer.
Let the AI build the whole cadence from the agent's context in one pass, then tweak it · or compose it step by step yourself. Both land in the same editor.
Each step picks a channel · email · LinkedIn · WhatsApp · Telegram · Instagram · Messenger · SMS · Voice. The channel router enforces per-channel daily caps and deliverability safety on every send.
Flip a sequence between AI-per-step and Template whenever you want from the editor. A warning reminds you that template bodies were authored as AI briefs · edit them before launch so they read as finished copy.
Pick AI-per-step when personalization wins the reply · founder-led sales, complex offers, audiences that smell a mail-merge. Pick Template for high-volume cold outreach where you want predictable, zero-LLM-cost first touches at scale. You can switch a sequence between the two anytime from the editor.
Yes, through merge fields. You write the body once with {{variables}} like {{firstName}} and {{company}}, and each send substitutes the prospect's values · no LLM call. It is variable-merge personalization, not the fresh per-prospect rewriting that AI-per-step does.
No. Template only removes the LLM cost from first-contact sends. Every inbound reply, on every channel, always routes through the AI auto-reply engine regardless of the sequence mode. Your conversations stay fully AI-driven either way.
Yes. The AI can build the entire cadence from your agent's context · steps, channels, timing and bodies · in one pass. You review and edit it in the same editor, or start from scratch and add steps by hand.
Each step targets one channel, and a sequence can mix all 8. At send time the channel router applies your per-channel daily caps and deliverability safety, so steps fire on the right channel without overshooting any account's limits.
Trial on request · contact our sales team.