Who is the client?

Colleva operates in recruiting services with an AI angle.

As an early-stage business, their growth depended on learning which segments respond, why they respond, and how to turn those conversations into qualified pipeline, without relying on email.

‍

What was the problem?

They were still validating product-market fit and needed a structured way to test ICP and messaging while building pipeline.

They wanted a system that could generate conversations through LinkedIn-only outreach, run multiple experiments in parallel, and create reliable reporting so they could see what was actually working.

‍

The Approach we took:

• Started with positioning and ICP clarity: We aligned on the audiences and problems worth targeting so outbound would serve as a learning engine, not just a lead engine.
• Built a LinkedIn-only outbound system: We used Clay and AI agents to build segmented lists of companies and decision-makers and to support research and relevance at scale.
• Launched multiple ABM plays driven by signals: We ran ABM-style plays based on website visitors, competitor audiences, lookalike targeting, job hiring signals, and technology usage targeting, so outreach was not random, it followed intent patterns.
• Protected sender accounts while scaling: We managed pacing and volume across multiple LinkedIn senders to avoid platform restrictions while maintaining high engagement.
• Ran structured testing and scaled winners: We tested roles, value propositions, segments, and message angles, then doubled down on what produced qualified conversations.
• Connected outcomes to pipeline tracking: Qualified conversations were pushed into HubSpot so pipeline could be tracked and measured cleanly over time.

‍

Results:

‍

Over six months, Colleva generated $2.6M in qualified pipeline using a LinkedIn-only motion.

The outbound program achieved a 27.1% connection acceptance rate and a 23.8% reply rate, showing both strong targeting and strong message-market fit for the segments that performed best.