Skip to content
Work with Shashwat
Menu
ICP Intelligence

ICP Deep Dive

Find the patterns your best customers share

For founders and demand gen leads. The decision: which customers should we go after next?

Who should we sell to next? List your best customers with industry, size, contract value, sales cycle, trigger and champion. ICP Deep Dive shows what they share, names ties as ties, and writes a draft ideal customer profile to check against your CRM. It looks nothing up and invents no customer.

Try it with an exampleUse your own details

Patterns into one profilePatterns into one profile: Your customer description sorted into Likely Company Size, Likely Industry, Likely Stage.Patterns into one profileYour customer descriptionLikely CompanySizeLikely IndustryLikely Stage
How the answer is laid out, using the sections the tool prints.

How do I find ICP patterns from our best B2B customers and CRM data?

Enter your best customers in ICP Deep Dive as a plain-text description or as structured customer rows with name, industry, size, ACV, and champion title, plus optional sales cycle days, buying trigger, and tech stack. The tool extracts shared patterns across industry, company size, deal economics (average ACV, ACV range, and average sales cycle), tech stack signals, champion roles, and buying triggers, listing tied values when there is no single leader and adding sector notes when your inputs name a supported sector.

What you get

This sample is a made-up example, built from the inputs behind Fill in an example. It shows part of one section of the answer.

Detected Patterns

Likely Company Size

300 to 3,000 employees (as you typed it)

Likely Sector

fintech, spend and expense

  • Buyer roles this sector usually involves: Chief Financial Officer, Finance Controller, Head of Accounts Payable, Head of Treasury, Internal Audit Lead, Compliance Officer, Head of IT (your text names: Finance Controller)
  • Metrics to ask about: days to close the books, reconciliation effort, policy breach rate, approval cycle time

See the full example answer or run it with your own details.

Run ICP Deep Dive

Every field is optional. You can add: Your best customers, Customer descriptions, Product category.

At least 5 values per line, separated by commas; ACV is a number in dollars. After them you can add the sales cycle in days, the buying trigger and then each tool in their tech stack. Leave it empty if you describe your customers in words below. For example: a manufacturing company, Manufacturing, 300 to 3,000 employees, 24000, Finance Controller

Alternative: Describe your best customers in text format

Up to 6,000 characters in this form. For a longer document, use ICP Intelligence in Claude, which takes up to 100,000.

What type of product you sell

Optional: your company or product name, so the answer can name it

Optional: how you charge (software subscription, services, connectivity, per transaction, marketplace, hardware plus software, or investment management). Read from your other inputs when left out

Worked example: Darwinbox (illustrative inputs)

About this example. Darwinbox is a real company. It has not reviewed this page, and Helix GTM Consulting has no relationship with it. Darwinbox sells an HR platform covering recruiting, onboarding, payroll and employee administration (TechCrunch, 5 March 2025). Every input below is illustrative, written by Helix to show how the tool works. None of it is Darwinbox's data, and the output is an exercise result, not Darwinbox's strategy.

We ran ICP Deep Dive on 19 November 2026 through the ICP Intelligence MCP endpoint (server version 1.2.28); the excerpt is cut word for word from its answer.

Show the inputs
  • Company: Darwinbox
  • Product category: HR software (HCM): recruiting, onboarding, payroll and employee administration
  • Business model: saas
  • Customers: name: an illustrative Indian private bank, industry: banking, size: 5,000 to 10,000 employees, acv: 180,000, sales cycle days: 150, tech stack: legacy on-premise payroll, buying trigger: payroll consolidation across states, champion title: CHRO | name: an illustrative quick-commerce company, industry: retail and ecommerce, size: 2,000 to 5,000 employees, acv: 90,000, sales cycle days: 90, tech stack: spreadsheets, buying trigger: frontline hiring surge, champion title: Head of HR operations | name: an illustrative Middle East conglomerate, industry: diversified group, size: 10,000+ employees, acv: 250,000, sales cycle days: 210, tech stack: legacy on-premise payroll, buying trigger: payroll in six countries, champion title: Group CHRO | name: an illustrative hospital chain, industry: healthcare, size: 2,000 to 5,000 employees, acv: 110,000, sales cycle days: 120, tech stack: spreadsheets, buying trigger: shift rostering errors, champion title: CHRO

Exercise output from illustrative inputs. It is not Darwinbox's data, result or strategy.

Synthesized ICP

Based on pattern analysis (a value is named only where it leads; a tie lists every tied value):

Ideal Customer Profile:

  • Industry: banking; retail and ecommerce; diversified group; and healthcare (tied: 4 industries equally common)
  • Size: 2,000 to 5,000 employees
  • Budget: $90,000 to $250,000 ACV seen in your customers (average $157,500)
  • Tech Stack: legacy on-premise payroll and spreadsheets (tied)
  • Buying Trigger: "payroll consolidation across states", "frontline hiring surge", "payroll in six countries" and "shift rostering errors" (tied)
  • Champion: CHRO
  • Sales Cycle: ~143 days (average of your deals)

Run it with your own details

Example answer

Generated on 12 November 2026 from this branch, with the inputs behind Fill in an example. Your own inputs give a different result.

ICP Intelligence report

Example: ICP Deep Dive

Made with ICP Deep Dive by Helix GTM ConsultingBased on the made-up example inputshttps://gtmhelix.com/tools/icp-intelligence/icp-deep-dive/

Show the full result

ICP Pattern Analysis (from Description)

  • Company: a spend management software company in India
  • Product category: spend management software

Sector: read from your inputs as fintech, spend and expense. Business model: software subscription (from your answer).

Input Analyzed
Mid-size manufacturers with 300 to 3,000 employees and large field sales teams; the finance controller buys and the pain is a slow month-end close. Also IT services firms with heavy travel spend, and distributors with many branches.

Detected Patterns
Likely Company Size

300 to 3,000 employees (as you typed it)

Likely Sector

fintech, spend and expense

  • Buyer roles this sector usually involves: Chief Financial Officer, Finance Controller, Head of Accounts Payable, Head of Treasury, Internal Audit Lead, Compliance Officer, Head of IT (your text names: Finance Controller)
  • Metrics to ask about: days to close the books, reconciliation effort, policy breach rate, approval cycle time
Likely Stage

not stated in your text

Sector notes: fintech, spend and expense
  • Who usually decides: The CFO signs; the Finance Controller or the head of the affected finance team champions; internal audit and compliance review controls; IT checks ERP integration and security; HR joins when employees are affected.
  • What this sector measures: days to close the books, reconciliation effort, policy breach rate, approval cycle time, audit findings, cost per transaction, payment success rate.
  • Words this sector's buyers use: reconciliation, month end close, policy controls, audit trail, accounts payable, ERP posting, compliance review, approval workflow, data residency.
  • A proof point that lands: Close time or reconciliation effort before and after for one entity, signed off by the finance controller.

Recommendations

For more precise ICP analysis, provide structured customer data in this format (the customer below is an example, not taken from your input):

Example figures: replace with your own.

{
  "customers": [
    {
      "name": "Example Manufacturing Co",
      "industry": "Manufacturing",
      "size": "300-1000 employees",
      "acv": 24000,
      "sales cycle days": 60,
      "tech stack": ["SAP"],
      "buying trigger": "Audit finding",
      "champion title": "Finance Controller"
    }
  ]
}

Next Step: Collect structured data from your CRM, then re-run this analysis

Reading the answer

For a spend management software company in India: This profile describes 0 supplied customer records for spend management software. A common trait in this sample does not establish why customers bought. Evidence that could change this conclusion: a comparison with lost or unsuitable accounts, verified deal economics and contrary examples to the proposed traits.

Decisions by role

Proposed owners. Confirm the named person before acting.

Founder: For a spend management software company in India, compare the buying reasons in the supplied customer records with at least one lost or unsuitable account before approving the profile; record any trait shared by both groups.

CEO: For a spend management software company in India, ask finance to verify contract value and sales effort for the supplied customer records before funding expansion; do not use customer count as a profitability check.

CMO: For a spend management software company in India, check whether the concentration in the recorded industries reflects the supplied sample or an observed customer need before writing the audience brief.

CRO: For a spend management software company in India, ask the account owners for the supplied customer records to confirm the recorded champion and buying trigger before using those roles to qualify new deals.

Demand generation head: For a spend management software company in India, test the proposed criteria on a fresh account sample outside the supplied customer records before widening outreach; record false matches as well as matches.

Region context: Which region applies: India, Southeast Asia, Middle East, UK, Western Europe or North America? Add it to your description and rerun.

Built by Shashwat Ghoshhttps://gtmhelix.com/tools/icp-intelligence/icp-deep-dive/

When a tool is not enough: if the decision needs more than one answer, talk it through in a call. Book 45 minutes with Shashwat.

How to read the answer

When you enter a plain-text description of your best customers, the tool extracts Detected Patterns (Likely Company Size, Likely Sector, Likely Stage, usual buyer roles, and sector metrics) and prompts you under Recommendations to supply structured customer rows (name, industry, size, ACV, champion title, plus optional sales cycle, buying trigger, and tech stack). When you enter structured customer rows, it calculates the shared patterns across your list. It never looks up external companies; it only summarizes patterns in the data you provide and adds sector notes where your product category is clear.

What to do next

  • Founder: Export your top 10 to 20 closed-won customers from your CRM in the 5-column format (name, industry, size, ACV, champion title) and re-run the tool to replace text impressions with hard counts.
  • CEO: Use the Detected Patterns and Sector notes to align leadership on which company size and buyer role represent your core market today.
  • CMO: Check your website copy and case studies against the Words this sector's buyers use and A proof point that lands printed in the sector notes.
  • CRO: Focus outbound prospecting on the company size range and champion title that appear most often in your best customer rows.
  • Demand generation head: Use the detected company size, sector, and buyer roles to tighten audience filters on LinkedIn and in your contact database.

Why come back

Come back as soon as you pull your structured CRM export (including sales cycle days, buying triggers, and tech stack) to get a data-backed pattern analysis, and re-run it every six months as new customers close. The tool applies the same fixed rules every time. It does not call an AI model, and this site does not keep what you enter.

Questions about ICP Deep Dive

How many customers do I need?

There is no fixed minimum. With only a few records, more values come back as ties, and the answer names them as ties instead of picking one. Add customers until the patterns separate.

Does it find companies that match my ICP?

No. It describes the pattern in the customers you give and lists the data gaps to fill. To turn the profile into search criteria, use the Lookalike Signal Generator; to rank accounts you already have, use Account Prioritization.

What does the budget line mean?

It is the range of contract values seen in your own customers, with the average, not a market benchmark. The four illustrative customers in the worked example give $90,000 to $250,000, with an average of $157,500.

Is my customer list stored?

No. The tool needs no sign-in and stores no inputs or outputs. The hosted service records usage metadata only.

Who is the ICP Deep Dive tool not ideal for?

  • Teams looking for a list of companies to target: the tools describe and score profiles from your data, but they do not find companies.

How do I use the ICP Deep Dive tool with an AI assistant?

Add ICP Intelligence once (see Use it in Claude, ChatGPT or Gemini), then ask in plain words. Here is a prompt built from the tested example:

My role is not stated. Ask me for my role first. Use ICP Intelligence's ICP Deep Dive tool. customer descriptions: Mid-size manufacturers with 300 to 3,000 employees and large field sales teams; the finance controller buys and the pain is a slow month-end close. Also IT services firms with heavy travel spend, and distributors with many branches; product category: spend management software. If this tool is unavailable, ask me to open its web page; do not invent its output. In Gemini CLI or Antigravity, use the Helix GTM Suite skills package for related guidance, without claiming to run this connector.

How does the ICP Deep Dive tool work?

You do not need to sign in. A web browser is enough, and the answer comes back in a second or two.

  1. Fill in the form above, or press Fill in an example.
  2. Press Run ICP Deep Dive.
  3. Read the result on this page, then copy it and edit it before you use it.

It applies fixed rules in the ICP Intelligence code to the inputs above and returns text. It does not call an AI model, does not look anything up, and does not store what you type. The web form and the AI assistant connector run the same tool code, so the same input gives the same answer in both.

Next: ICP Scoring Model

More ICP Intelligence tools

ICP Intelligence home · All Helix GTM tools · Tools by GTM job

For developers

MCP tool name: icp_deep_dive. REST: POST https://icp-intelligence.gtmhelix.com/api/tools/icp_deep_dive with this JSON body (see How it works):

{
  "input": {
    "customer_descriptions": "Mid-size manufacturers with 300 to 3,000 employees and large field sales teams; the finance controller buys and the pain is a slow month-end close. Also IT services firms with heavy travel spend, and distributors with many branches.",
    "product_category": "spend management software"
  }
}

Choose your app

Run a Helix tool in your browser, then take the result to ChatGPT, Claude or Gemini. Connectors work with supported plans in Claude, ChatGPT, Cursor, VS Code and Windsurf. The Helix GTM Suite skills package supports Claude Code, Gemini CLI, Antigravity, Cursor and VS Code, and is available through skills.sh. Each plugin has its own installation steps, below or on its own page.

Connector setup guides · Skills package setup guide · Find the skills on skills.sh