From cold list to booked demo, without a human in the loop
Autonomous SDR that replaced a 6-person outbound team for a Series B SaaS
A vertical SaaS operator was burning ~$52k/month on outbound SDRs with a 3.1% meeting rate. We shipped an autonomous sales-qualification agent that enriches, scores, personalizes and books — end-to-end, no touch.
What was breaking
- Six SDRs cranking through Apollo lists at a 3.1% meeting-book rate — most of their day spent on manual research and copy-paste.
- Response SLA on inbound demo requests was creeping past 6 hours; leaders were losing meetings to faster competitors.
- Territory coverage was uneven — high-fit accounts in EMEA sat untouched for weeks between rep rotations.
How we built it
- Built a research + scoring layer that pulls firmographic, tech-stack and hiring signals for every inbound and outbound lead in under 8 seconds.
- Trained a fit model on the client's last 24 months of closed-won and closed-lost deals to score every lead on a 0-100 ICP index.
- Composed personalized first-touch sequences with an evaluator that rejects generic openers — every draft must pass a rubric before it sends.
- Wired direct calendar handoff via Cal.com so qualified prospects self-book into the correct AE's calendar without a rep in the loop.
Under the hood
- Ingestion: inbound webhook + outbound Apollo pull
- Enrichment: Clearbit, Apollo, LinkedIn, hiring-signal scraper
- Scoring: fit model + intent model, cached in Postgres
- Reasoning: GPT-class LLM with rejection-sampling evaluator
- Delivery: HubSpot Sequences + Cal.com handoff
- Observability: every run traced, replayable, scored
What changed for the business
How it shipped
Mapped the existing outbound stack, pulled 18 months of CRM data and defined the ICP fit rubric with the sales leader.
Shipped the enrichment + scoring pipeline and a shadow-mode agent that drafted (but did not send) replies for human review.
Turned on autonomous send with guardrails, closed-loop evals on every draft, and Cal.com handoff for booked meetings.
Transferred prompts, evals and the operator dashboard. The client's RevOps lead now owns tuning; we run the SLA.
What operators said
“It's replacing an entire outbound team — but the part I actually love is the traceability. Every meeting on my calendar, I can see exactly why the agent thought this lead was worth my time.”
More case studies
What people are saying
Turn your business into an AI-powered machine
30 minutes. No pitch. We map the highest-leverage AI plays in your business — yours to keep.