How to Define Your Ideal B2B Customer: A Practical Guide to ICP

Many B2B companies start prospecting too early.

They pick a country, an industry and a company size, and then start building a list of potential customers. The result is often hundreds of companies that match a few database filters but have no real reason to buy the product or service in question.

That is why a well-defined ICP – Ideal Customer Profile – matters so much. A good ICP helps you answer two fundamental questions:

Which companies are actually worth our time?

and

Which companies do we not want to target?

In this guide, we walk through a practical approach using the example of a manufacturing company looking for new customers in Germany.

What is an Ideal Customer Profile?

An ICP does not describe a specific person. It describes a type of company that is highly likely to be a strong commercial fit for your offer.

A typical target market such as “engineering companies in Germany” only tells you where to look. An ICP tells you how to recognise that a company genuinely belongs in your targeting. In B2B, it can cover, for example:

• industry,
• company size,
• region,
• type of production,
• technologies,
• purchased components,
• business model,
• purchasing structure,
• typical needs,
• how the company works with external suppliers.

An ICP is not a list of companies. It is the logic you use to decide which companies belong on that list.

A list of companies goes out of date within a few months. A clearly described selection logic keeps working for you – for a new market, a new database or a new salesperson joining the team.

ICP vs. buyer persona

ICPs and buyer personas are often confused, but they answer different questions.

ICP: What type of company do we want as a customer?

Buyer persona: Which person inside that company do we need to reach?

ICP example: a manufacturer of industrial machinery in Germany that sources machined components from external suppliers.

Example buyer personas in such a company:

• Head of Procurement,
• Technical Buyer,
• Head of Production,
• or a Supply Chain Manager.

The order matters: company first, person second. Even the best-crafted message to a Head of Procurement will not help if their company has no need for your offer.

1. Start with your best existing customers

Your own data is the best foundation for an ICP. But do not automatically pick your largest customers – look for the ones where the relationship genuinely works well.

Ask yourself questions such as:

• Who orders repeatedly?
• Where do we achieve healthy margins?
• Where does our solution solve an important problem?
• Who values our flexibility, capacity or know-how?
• Which customer relationships work efficiently over the long term?
• Where is there a high probability of future business?

Then look for the characteristics these customers have in common – industry, type of production, size, purchasing approach, technologies.

2. Define basic firmographic criteria

Firmographic criteria describe a company from the outside. They include, for example:

• country,
• region,
• industry,
• number of employees,
• revenue,
• number of locations,
• ownership,
• local company vs. international group.

But this is only the first layer. A combination such as

Germany + manufacturing + 50–500 employees

is not a strong ICP. It is just a basic database filter.

Two companies of similar size in the same industry can have completely different needs and buying behaviour. One makes everything in its own facilities; the other buys in most of its parts.

3. Add operational and technological criteria

This is where the real value of an ICP lies. Operational and technological criteria describe how a company works on the inside – and therefore whether it could realistically need your offer.

Take a Czech manufacturer of precision CNC parts as an example. Relevant questions might be:

• Does the company have its own assembly operations?
• Does it use precision-machined metal components?
• Does it work with external manufacturing suppliers?
• Does it manufacture machinery or equipment where similar components are repeatedly required?
• Does it have its own technical purchasing or procurement department?
• Does it purchase according to technical drawings?
• Does it require small, medium or serial production volumes?
• Does it need a flexible external manufacturing partner?

These are the criteria that often decide real commercial fit – they separate a genuinely relevant company from one that merely ticks the boxes of a database filter.

4. Define a negative ICP

Exclusion criteria are just as important as inclusion criteria. Together they form what is known as a negative ICP.

For a manufacturer of CNC parts, a negative ICP might include:

• pure distributors,
• trading companies with no manufacturing of their own,
• companies with extensive in-house CNC capacity,
• direct competitors,
• companies outside the supplier's technological capabilities,
• companies requiring production volumes the supplier cannot handle efficiently,
• companies where purchasing decisions are fully controlled from another country and the local plant has no supplier-selection authority.

5. Separate fit from timing

An ICP tells you whether a company matches your ideal customer profile. It does not tell you whether that company needs your offer today.

Company A is an excellent ICP fit, but has no current need – all its supplier capacity is already covered.

Company B is a similar ICP fit and is currently:

• expanding production,
• opening a new facility,
• recruiting buyers or technical staff,
• introducing a new product,
• expanding into new countries,
• looking for new suppliers.

At this moment, Company B may deserve a higher priority. Company A does not drop out of your targeting – its turn simply comes later.

6. One company can have several ICPs

Do not try to create a single universal ICP for your entire business. Different types of customers buy for different reasons and make decisions differently. A company may have, for example:

• an ICP for direct customers,
• an ICP for OEM manufacturers,
• an ICP for distributors,
• an ICP for system integrators,
• an ICP for different product lines,
• an ICP for different countries.

Several precise ICPs are better than one overly broad ICP.

Manufacturing ICP example

Here is the difference between a weak and a stronger ICP for a CNC parts manufacturer looking for customers in Germany:

A simple ICP template

You can fill in this template separately for each of your ICPs:

If you cannot explain a criterion with a clear commercial reason, it probably does not belong in the ICP.

A simple ICP quality test

How to prioritize companies using an ICP

An ICP does not have to be a simple “fits” or “does not fit”. You can build a straightforward scoring model with four layers:

A company that fails a must-have or falls under exclusion drops off the list. The rest you rank by fit and timing. Instead of a large database of companies that all carry the same priority, you end up with a prioritized company list your sales team can work with straight away.

How ICP relates to market segmentation

Market segmentation divides the broader market into groups of companies with similar characteristics. ICP goes one step further and defines which companies within those segments are truly ideal for your offer.

We describe how to split a market into segments in our guide: How to do market segmentation

From ICP to prospecting

The ICP sits at the start of the whole sequence:

ICP → segment → specific companies → decision makers → outreach

First you need to know which company is the right one. Only then can you efficiently search for specific companies and the relevant decision makers inside them.

How to move from an ICP to a list of companies and contacts is covered in our guide: How to do prospecting

The entire sales process, from choosing a market to opening a new opportunity, is summarised in: How to Do B2B Business Development

How AI can help build and use an ICP

AI does not know your ideal customer profile on its own. But it can significantly speed up work that would otherwise take days:

• analyse the common characteristics of your existing customers,
• suggest potential segments and find similar businesses,
• compare companies,
• identify technological and operational characteristics,
• prioritize companies against defined criteria.

The commercial logic, however, still has to be defined by people – you know why your customers buy from you and where your offer does not make sense.

AI can make the search much faster, but it needs to know why a particular customer is good for your business.

How Intedat fits into the process

In Intedat, you describe in natural language what your company offers, which customers you are looking for and which market you want to target. Based on that brief, Intedat can help you:

• suggest relevant segments,
• identify suitable companies,
• evaluate and compare their relevance,
• prioritize the companies worth approaching,
• identify relevant decision makers,
• support systematic outreach.

The more precisely you can describe your ICP, the better the resulting selection of companies can be.

Learn more about the platform: See how Intedat works

Fewer companies. Better companies.

The goal of a strong ICP is not to build the largest possible database of potential customers.

The goal is to find companies where you can clearly explain why there is a high probability of a good commercial fit.

If you want to improve your B2B prospecting, don't start with a bigger database. Start with a more precise definition of your ideal customer.

A well-defined ICP can help you search for fewer companies – but much better ones.