Playbooks

Clone Your Best Customers Playbook

Stop rebuilding your ICP from filters and guesses. Use the 30 to 50 deals already in your CRM as the ICP, then execute this playbook to find more of them.

Juzefin Gjecaj

Co-Founder | Head of GTM

Modern team workspace

Objective

Turn your closed won deals into a repeatable prospecting system instead of rebuilding your ICP from scratch every quarter.

Why this works

The companies already paying you are the strongest ICP signal you have. They converted, paid, and stayed. That is stronger evidence than any demographic filter. This playbook shows you how to use that data as the input for finding more accounts like them.

Tools needed

  • CRM: HubSpot, Salesforce, or Pipedrive

  • CompanyEnrich, Ocean.io etc (lookalike search, buying signals, people search)

  • Claude or Clay (account scoring)

  • Prospeo and FullEnrich (email verification)

  • lemlist or Instantly (email sequencing)

Step 1: Export your closed won domains

  1. Open your CRM.

  2. Pull your last 30 to 50 closed won deals.

  3. Export domains only. No other fields needed.

  4. Do not filter or curate this list. Every company on it already went through your full sales process and became a paying customer. That is the seed list.

Step 2: Generate lookalike accounts

  1. Open CompanyEnrich.

  2. Upload your seed domain list.

  3. Run the similar companies API. It searches across more than 30 million companies.

  4. Expect roughly 500 ranked matches from a list of 10 seed domains.

  5. Export the full lookalike list.

Step 3: Add buying signals to each account

  1. In CompanyEnrich, pull signal data for every account in your lookalike list.

  2. Track these four signals:

    • New funding round

    • New leadership in the buyer seat

    • Hiring spike in target roles

    • Tech stack change that signals intent

  3. Save the specific signal next to each account. You will use it later for both prioritization and outreach copy.

Step 4: Score and tier every account

  1. Open Claude or Clay.

  2. Build a classification prompt using your closed won data and the signal layer as the scoring rubric.

  3. Run every account through the prompt.

  4. Sort the output into three tiers:

    • Tier 1: strong fit, active signal. Move on these immediately.

    • Tier 2: strong fit, no active signal yet. Enter a standard sequence and promote to Tier 1 when a signal fires.

    • Tier 3: adjacent fit. Watch these, do not prioritize yet.

Step 5: Enrich contacts at each account

  1. In CompanyEnrich, run people search on every Tier 1 and Tier 2 account. It searches across more than 170 million profiles.

  2. Pull the titles that match your buyer persona.

  3. Waterfall the contact list through Prospeo, then FullEnrich, to verify emails.

  4. Discard any contact that fails verification at both tools.

Step 6: Build outreach by tier

Tier 1

  1. Write one custom outreach prompt per account in Claude or Clay, using the account's specific signal as context.

  2. Load the contact into lemlist or Instantly for parallel email and LinkedIn outreach.

  3. Run both channels at the same time.

Tier 2

  1. Build one templated message with a single variable swap, using the signal or a relevant firmographic detail.

  2. Load contacts into lemlist and Instantly.

  3. Run LinkedIn plus email only.

Tier 3

  1. Load contacts into Instantly only.

  2. Run a high volume email sequence.

  3. Do not add other channels until the account shows more intent and moves up a tier.

Step 7: Sync results back to the CRM

  1. Log every reply in HubSpot, Salesforce, or Pipedrive.

  2. Log every meeting booked.

  3. Log every closed won deal.

  4. Use this updated data as the new seed list the next time you run this playbook.

Repeat

Run this system every cycle using your latest closed won data as the seed. Each pass improves the lookalike matches and sharpens the scoring model, because the input data gets richer with every deal you close.

Create a free website with Framer, the website builder loved by startups, designers and agencies.