Build an AI-recommended lookalike account list in ChatGPT

Built by: Lusha
Time to build: 1 min
Difficulty: Easy
Tools: ChatGPT ▪ Lusha
Type: Prompt

Example outputs in this play are illustrative — they reflect the structure, fields, and format of real Lusha connector output, but were not pulled from a live session. Run the prompt with your own seed customers to see live results.

Your best customers are not just proof that your product works. They are a pattern.

They show which company types convert, which teams feel the pain, which markets respond, and which signals often appear before a strong opportunity. The problem is that most teams turn that pattern into a static ICP and stop there.

A static ICP tells you what to search for. A lookalike workflow helps you find more companies that resemble the accounts already working.

This prompt uses Lusha in ChatGPT to analyze your best customer accounts, find lookalike companies, enrich each account, check recent buying signals, and create an explainable recommendation score based on verified Lusha data. Instead of building a list from broad filters alone, you start from proven customers and let Lusha help you find the next best accounts to target.

How to start

Open in Claude Open in ChatGPT
1

Open Lusha in ChatGPT

Go to Lusha in ChatGPT and click “Start chat.” Every conversation started this way is automatically Lusha-enabled.

2

Or invoke Lusha in any existing conversation

Type @Lusha in the prompt bar and select Lusha from the dropdown. Unlike Claude, Lusha does not activate automatically in every ChatGPT conversation. You must invoke it every time.

3

Add your best customers and send

Copy the prompt below, add at least five seed customer domains or LinkedIn company URLs, include your ICP context and exclusions, and send. Lusha finds lookalike accounts, and ChatGPT recommends which ones to target first based on the returned data.

The prompt

Start from Lusha in ChatGPT or type @Lusha before sending.

@Lusha Build an AI-recommended lookalike account list
based on my best customers.

SEED CUSTOMERS:
Add 5-100 customer company domains or LinkedIn company URLs
that represent strong-fit accounts.

1. [customer domain or LinkedIn company URL]
2. [customer domain or LinkedIn company URL]
3. [customer domain or LinkedIn company URL]
4. [customer domain or LinkedIn company URL]
5. [customer domain or LinkedIn company URL]

BEST CUSTOMER CONTEXT:
Why these customers are good examples:
[high ACV / fast sales cycle / strong retention /
high expansion potential / successful use case /
strategic segment / other]

ICP:
Best-fit industries: [industries]
Best-fit company size: [employee range]
Best-fit regions: [regions]
Target personas: [titles or personas]
Relevant departments: [Sales / Marketing / RevOps / IT /
Operations / Customer Success / Finance / HR / other]

EXCLUDE:
Do not include:
- Existing customers: [company domains, if any]
- Competitors: [company domains, if any]
- Open opportunities: [company domains, if any]
- Poor-fit segments: [industries, company sizes, regions,
  or business models to exclude]

MY PRODUCT:
[One sentence describing what you sell and the problem
it solves]

Using Lusha, do the following:

1. VALIDATE THE SEED ACCOUNTS
   Confirm that each seed company can be matched to a
   real Lusha company profile.

   Return:
   - Company name
   - Domain
   - Industry
   - Employee count
   - HQ location
   - Match status

   If fewer than 5 seed accounts can be verified, ask me
   for more seeds before running the lookalike search.

2. FIND LOOKALIKE ACCOUNTS
   Use the verified seed customers to find lookalike
   companies that resemble my best accounts.

   Prioritize companies that appear similar across:
   - Industry
   - Company size
   - Region
   - Business model
   - Market category
   - ICP relevance

   Do not include companies listed in the exclusion section.

3. ENRICH EACH LOOKALIKE ACCOUNT
   For each lookalike company, return:
   - Company name
   - Domain
   - Industry
   - Employee count
   - HQ location
   - Revenue range if available
   - Company LinkedIn if available

4. CHECK RECENT BUYING SIGNALS
   For each lookalike account, check recent signals from
   the last 6 months, if available.

   Prioritize:
   - Hiring surges
   - Hiring surges by relevant department
   - Hiring surges by location
   - Headcount increases or decreases
   - IT spend changes
   - Website traffic changes
   - Commercial activity news
   - Corporate strategy news
   - Financial events news
   - People news
   - Product activity news
   - Risk news
   - Intent topics related to my product, if available
   - Promotion or company-change signals for relevant contacts

5. CREATE AN EXPLAINABLE RECOMMENDATION SCORE
   Create a recommendation score from 1-100 using only
   the Lusha data returned and the ICP context I provided.

   Break the score into:
   - Lookalike match: 35 points
     How similar the company appears to the seed customers.

   - ICP fit: 30 points
     How well the company matches my target market.

   - Timing: 20 points
     Whether recent signals make the account more timely.

   - Actionability: 15 points
     Whether relevant contacts and contact data are available.

   Make the score explainable. Do not present it as a
   prediction that the account will convert or buy.

   If there is not enough verified data to support a score,
   mark the score as low-confidence and explain what is missing.

6. ASSIGN A RECOMMENDATION
   Assign one recommendation:

   Target now:
   Strong lookalike match + high ICP fit + relevant signal
   or strong actionability.

   Add to campaign:
   Strong lookalike match + good ICP fit, but weaker timing.

   Nurture:
   Partial fit, weak signal, or unclear urgency.

   Review:
   Interesting match, but missing data or unclear fit.

   Exclude:
   Poor fit, disqualified, competitor, existing customer,
   open opportunity, or unverifiable account.

7. FIND STARTING CONTACTS
   For each account marked Target now, find 1-2 relevant
   contacts matching the target persona or department.

   Return:
   - Name
   - Current title
   - Department
   - Seniority
   - Location
   - LinkedIn profile if available
   - Verified business email availability
   - Direct or mobile phone availability
   - DNC status if available

   If contact data is not available or cannot be revealed,
   say so clearly rather than guessing.

8. CREATE AI RECOMMENDATIONS
   For each Target now account, write:
   - Why this account looks like our best customers
   - Why now may be a good time to reach out, if supported
     by signals
   - Who to contact first
   - What angle to lead with
   - One subject line under 7 words
   - One opening line under 30 words
   - One discovery question

   Do not:
   Invent signals, tools, vendors, internal projects,
   contract status, or buying intent.
   Claim the account is ready to buy.
   Present the score as a conversion prediction.
   Include excluded companies.
   Force personalization when the data is weak.

9. OUTPUT FORMAT
   Return:
   - Seed account validation
   - Lookalike account table
   - Company enrichment
   - Recent signals, if any
   - Explainable recommendation score with breakdown
   - Confidence level
   - Recommendation
   - Starting contacts, if available
   - Recommended outreach angle
   - Excluded accounts and why

Do not invent companies, contacts, emails, phone numbers,
signals, or recommendations. If Lusha cannot verify a
company, contact, or signal, mark it clearly.

What you’ll get back

 

An AI-recommended lookalike account list with verified company data, explainable scoring, buying signals, and starting contacts when available. Here’s what the output looks like:

AI lookalike account recommendations — Lusha

FieldValue
Seed accounts5 verified customer accounts used as lookalike seeds
Lookalikes found25 companies · 8 Target now · 11 Add to campaign · 6 Review
Top recommendation[Company A] · recommendation score 87/100 · high confidence
Score breakdownLookalike match 32/35 · ICP fit 25/30 · timing 18/20 · actionability 12/15
Strongest signalHiring surge in RevOps · supports workflow and data quality angle
Starting contactR.M. · VP Revenue Operations · verified email available · mobile available

Example outputs in this play are illustrative — they reflect the structure, fields, and format of real Lusha connector output, but were not pulled from a live session. Run the prompt with your own seed customers to see live results.

 

Why use Lusha in ChatGPT for AI lookalike recommendations

 

Lookalikes are useful because they start from proof. Instead of guessing what your ICP should look like, you use the accounts that already converted, retained, expanded, or moved quickly through the pipeline.

Lusha helps turn those seed accounts into a recommended target list. The prompt finds similar companies, enriches them with verified company data, checks recent buying signals when available, and identifies relevant contacts when available. ChatGPT then adds the recommendation layer: why the account looks similar, why the timing may matter, and what action to take next.

The scoring step matters because a lookalike match alone is not enough. A company may resemble your best customers but show no current movement. Another may be slightly less similar but have a stronger buying signal, a better-fit persona, or more actionable contact data. The prompt makes that tradeoff visible.

The result is a more practical prospecting workflow: start with your best customers, find similar accounts, score them transparently, and move first on the accounts with both fit and timing.

Lusha data is sourced and used in accordance with Lusha’s Privacy Policy and Terms of Use. Lusha is GDPR compliant and covers contacts across North America, EMEA, and APAC.

FAQ

  • How many seed customers do I need?

    Use at least five seed customers. Choose accounts that represent the segment you want to repeat, such as high-retention customers, high-ACV customers, fast-moving opportunities, or accounts with strong expansion potential.

  • What is the recommendation score based on?

    The score is created by ChatGPT using the Lusha data returned in the workflow. It is based on lookalike match, ICP fit, recent timing signals, and actionability. It is a prioritization aid, not a prediction that the account will convert.

  • How is this different from a normal lookalike list?

    A basic lookalike list finds similar accounts. This prompt adds enrichment, signal checks, explainable scoring, starting contacts when available, and AI recommendations so the list becomes more actionable for sales and marketing.

  • Can I exclude existing customers or competitors?

    Yes. Add existing customer domains, competitors, open opportunities, and poor-fit segments in the exclusion section. The prompt asks Lusha to keep those accounts out of the final recommendations.

  • Should sales trust the AI score automatically?

    No. The score should help prioritize, not replace judgment. The prompt keeps the score explainable so sales can see which data points drove the recommendation and decide how to act.

Related Prompts

Ready to run this?

One data connection. Works in Claude, ChatGPT, your CRM, or any agent you build.

Get free access Read the Docs