SubstackGitHubBook a call →
← Back to side projects
Prototype

Alice ABM Prototype

Built an end-to-end account-based workflow prototype showing how 11x Alice could move from target accounts to pipeline.

Product rationale for account-based workflows

ABM isn't a channel — it's a coordination problem. The gap in most AI SDR tools is that they're built to maximise outbound volume, not account-level precision. Enterprise deals involve 5–10 decision-makers, long sales cycles, and brand risk from bad outreach. A truly enterprise-grade AI SDR needs to think in accounts, not contacts — and that requires fundamentally different workflows: committee coverage, coordinated sequencing, and measurement that tracks pipeline, not just replies.

What I built

A fully interactive prototype demonstrating how an AI SDR like 11x Alice could be redesigned for enterprise ABM — moving from single-lead outbound to coordinated account-level campaigns with measurable outcomes.

Why it matters

Most AI SDR tools are built around high-volume, single-lead outbound. Enterprise buyers don't buy that way. They buy through committees. This prototype shows what an ABM-native workflow looks like — account selection, committee mapping, coordinated sequencing, and human-in-the-loop review at every high-risk step.

How it works

  1. 1Account selection: upload a target list, apply ICP filters, and score accounts by fit and intent signal
  2. 2Committee mapping: discover key personas per account and assign coverage responsibilities
  3. 3Sequence design: build multi-persona, multi-channel campaigns with branching logic and review gates
  4. 4Execution: run campaigns with guardrails, flag anomalies, and log all activity at the account level
  5. 5Measurement: track meetings booked, opportunities created, and pipeline by cohort vs control

What it demonstrates

  • Account selection and prioritisation based on ICP and intent signals
  • Buying committee discovery, persona mapping, and account coverage
  • Coordinated multi-persona messaging and campaign execution with review points

Trust and control

  • ·Guardrails for deliverability and brand risk
  • ·Human-in-the-loop approval for high-risk sends and account-level changes
  • ·Measurement design: lift vs control for meetings, opportunities, and pipeline

Stack and tools

Next.jsTypeScriptTailwind CSSVercel

What I'd improve next

  • Reporting layer: meetings, opportunities, pipeline, and multi-touch attribution
  • Integration with Salesforce for two-way sync on account and contact data
  • Enrichment layer (Clay, Apollo) for real-time contact and signal data
  • Sequence tool integrations (Outreach, Salesloft) for execution