Start by expanding the initial ICP.

• Identify companies similar to the ICP.
• Use multiple sources to maximize coverage.
• Tools to use:
  • DiscoLike
  • AI Ark
  • Clay
  • Ocean.io

‍

Centralize all sources into a single system.

• Create one table in Clay.
• Aggregate results from all tools into this table.
• Dedupe records to remove overlaps.
• Standardize fields by creating standard columns:
  • Company Name
  • Domain
  • Description
  • Company LinkedIn Profile

‍

Filter the list down to relevant accounts.

• Use the ‘Use AI’ feature in Clay.
• Select any ChatGPT model (ChatGPT 4o mini recommended).
• Prompt the model to classify companies as:
  • Qualified
  • Not qualified
• Remove not qualified companies from the table.

‍

Segment companies by quality.

• Use Clay formulas to assign a score.
• Assign tiering logic based on the score:
  • Tier 1: Best fit
  • Tier 2: Good fit
  • Tier 3: Low fit

‍

Identify the right contacts inside qualified companies.

• Define target personas:
  • Example: CEO, VP of Sales, Head of Growth
• Use multiple tools for maximum coverage:
  • AI Ark
  • Ocean.io
  • Apollo
  • Clay

‍

Add context required for outreach.

• Enrich each contact with:
  • LinkedIn profile
  • Work email address
  • Any additional relevant firmographic or role data

‍

Apply personalisation to the best-fit accounts.

• Research Tier 1 contacts using:
  • Warm connection discovery via The Swarm
  • Custom decks and presentations via Gamma
  • Personalized landing pages or microsites via HubSpot
• Use these information in:
  • Manual cold calling
  • LinkedIn outreach via HeyReach

‍

Handle remaining leads using:

• Automated email campaigns via Instantly
• Lighter personalization based on enriched fields.

‍

Keep CRM data clean and synchronized:

• Create an automation in HubSpot.
• Logic:
  • If a deal already exists:
    • Associate the contact and message reply with the existing deal
  • If no deal exists:
    • Create a new deal
    • Associate the contact with it

‍

Set up Slack notifications for replies.

• Use insights to:
  • Refine AI qualification logic
  • Adjust tiering criteria
  • Improve future TAM and lookalike builds

‍