Lead Generation & Outreach System
Workroom Asia's own outbound engine: a browser-agent research process that finds and qualifies leads across the US, UK, Australia, and Canada, hands them into a Notion CRM with an AI-drafted first message, and routes every reply through a VA-run pipeline instead of a single crowded inbox.
- ROLE
- Research agent design, CRM architecture, outreach templates, VA workflow
- TOOLS
- Claude Co-Work · Notion · Apollo.io
Sourcing leads by hand doesn't hold up across four markets and four platforms.
Finding small business owners, coaches, clinics, and ecommerce brands worth reaching meant checking Google, LinkedIn, and Instagram by hand, market by market, then writing a first message from scratch for each one. Outreach was happening on LinkedIn, Instagram, Twitter/X, and email with no shared record of who'd been contacted or what was said, so messaging drifted from platform to platform and handoffs between whoever picked up a lead went undocumented.
Research, CRM, and outreach chained into one repeatable pipeline.
A morning research agent — a standing prompt run through the Claude Co-Work browser agent — searches Google, LinkedIn, and Instagram against a defined ideal-customer profile and populates a structured Notion "Lead Pipeline" database directly: business details, platform, an A/B/C lead score, a pain-point note, and a personalized draft message. Every lead moves through a defined multi-stage pipeline, from first contact through to closed or disqualified, with saved views so a VA always knows what to work next. A VA quality-checks the AI-drafted message before anything sends, works from a library of reply templates once a lead responds, and logs a simple end-of-shift report. A parallel three-email cold sequence runs through Apollo.io's automated cadence, roughly 20–40 sends a day, reaching the same ideal-customer profile across all four markets.
Built to work inside platform rules, not around them.
Some forms of LinkedIn automation were deliberately scaled back once they ran up against platform policy — the system was adjusted to operate within what LinkedIn actually allows rather than automating everything that was technically possible. Apollo's cold email cadence became the channel built to scale instead, since it doesn't carry the account-ban risk that heavier social-platform automation does. The result is dozens of qualified leads a day without gambling the accounts the business depends on.
Leads and follow-up living in separate systems?
Tell us what's manual today, and we'll map what a practical implementation could look like.