- Is it really worth pausing to define your ideal customer before launching campaigns?
- Why can't marketing and sales agree on lead quality?
- What goes into a truly useful ICP, and what sets it apart from a list of industries?
- Do you need an ICP for each product, or is one company-wide ICP enough?
- How does the 95:5 rule work, and what does it change for campaigns?
- Three real-world scenarios: how does lead management change for different Italian companies?
- How do you turn the ICP into operational data in your CRM and campaigns?
- What tools can you use to build the ICP and activate it in campaigns?
- Why it's a hot topic now: market data on marketing-sales alignment
- How to get started: the steps to build an ICP from scratch
- What are the most common mistakes when building an ICP?
- How do I know if the ICP is working and campaigns are improving?
- So, is building a shared ICP worth it for your company?
- How we see it at SHM
- Frequently Asked Questions
- Without a shared definition of the ideal customer (ICP), marketing and sales optimize different campaigns toward different goals, wasting budget.
- A useful ICP goes beyond industry and revenue: it includes adoption signals, technology compatibility and relationship history with your company.
- The ICP must become operational data in your CRM and advertising platforms, not remain a strategy document.
- Aligning marketing and sales on the ICP can more than double the revenue generated by marketing activities.
49% of sales leaders say their definition of a "qualified lead" is significantly different from marketing's. Meanwhile, half of the leads generated end up in the trash without anyone calling them.
Is it really worth pausing to define your ideal customer before launching campaigns?
Yes, and it's not a theoretical exercise: without a shared definition of the ideal customer, every campaign optimization fixes the symptoms without touching the cause. The time spent building the ICP pays back within a few weeks of more precise campaigns.
Sales wants more leads. Marketing generates more. Sales says they're not good. Marketing adjusts the targeting. The cycle starts over. This pattern repeats in almost every B2B company that has never formalized the answer to one simple question: which companies are we really best positioned to serve?
The ideal customer profile, or Ideal Customer Profile (ICP) for short, is the written answer to that question. It's not a list of industries or revenue brackets. It's an agreement between marketing and sales on where it's worth investing time and budget, and where it isn't.
| Your situation | Is building an ICP worth it? | why |
|---|---|---|
| Marketing and sales use different criteria to evaluate a lead | Yes | This is the main symptom of a missing or unshared ICP |
| You have plenty of leads, but few close or renew | Yes | You're acquiring accounts that aren't a good fit for your product |
| You sell multiple products to different segments | Yes | You need one ICP per product, not just one for the company |
| You already have a written ICP, but sales ignores it | Yes | The ICP hasn't yet become operational data in your CRM |
| You have few customers, but all are satisfied and loyal | Periodic review | The ICP probably already exists implicitly: it's worth formalizing it |
Why can't marketing and sales agree on lead quality?
The disagreement doesn't come from bad faith, but from the lack of a written, shared common criterion. Each salesperson uses their own experience as a yardstick, and marketing optimizes for metrics that don't reflect the real priorities of the people closing contracts.
One salesperson prioritizes company size. Another looks at the prospect's technology ecosystem. A third chases any account that shows interest. None of them is wrong in absolute terms, but without a shared criterion, marketing doesn't know which signal to optimize for.
The result is the classic cycle: marketing generates leads, sales criticizes them, marketing adjusts the targeting, sales stays dissatisfied. According to HubSpot data, 49% of Chief Sales Officers say their organization's definition of a qualified lead differs significantly from marketing's.
50% of marketing-generated leads are ignored by salespeople in mid-sized B2B companies, according to The Starr Conspiracy, citing Forrester data. In effect, half of the acquisition budget doesn't even produce a phone call.
What goes into a truly useful ICP, and what sets it apart from a list of industries?
A useful ICP answers one precise question: of all the companies that could buy from us, which are we best positioned to win, serve and grow over time? The answer takes more than an industry and a revenue bracket.
Saying "we sell to manufacturing SMEs with 50-200 employees" describes a segment, not an ideal customer. A company in that segment might have an incompatible technology ecosystem, require excessive customization or have a history of abandoning the product after six months. It's not a good fit, even if it falls within the parameters.
The original source of this article, published on MarTech, identifies five categories of information that make an ICP actionable.
| ICP dimension | What it includes | Practical example |
|---|---|---|
| Basic company data | Industry, revenue, number of employees, operational size | Manufacturing, €10-50M in revenue, 80-300 employees |
| Technology ecosystem | Systems in use, competitors adopted, manual processes | Uses SAP ERP, no structured CRM, order management on Excel |
| Operational compatibility | Implementation complexity, required integrations, customizations | Standard integration with SAP, no customization required |
| Relationship history | Current customer, past opportunities, cross-sell potential | Former prospect lost due to timing, not product limitations |
| Customer success indicators | Traits associated with renewal, expansion, adoption | Companies with a dedicated IT contact renew at a 90% rate |
A complete ICP also includes the list of who is not a good fit: companies in unsupported regions, accounts with incompatible technologies, organizations that require more customization than the team can handle. Knowing where not to look is as valuable as knowing where to look.
Do you need an ICP for each product, or is one company-wide ICP enough?
If you sell multiple products to different segments, a single ICP hides important differences and leads to wrong decisions. It's better to build a separate fit score for each product.
An account can be a great fit for product A, mediocre for B and completely unsuitable for C. If you use a single score for all three, you risk excluding real opportunities or sending salespeople to the wrong accounts.
The solution described in the source is to build separate ICP scores in the CRM (the customer relationship management system), one per product. The factors to consider change: company size and type, technology ecosystem, presence of a competitor already adopted, history of previous opportunities, and whether the account is already a customer of another product.
That last signal is particularly valuable. An existing customer is often a great cross-sell opportunity. An opportunity lost because of timing should be handled very differently from one lost due to a structural limitation of the product: in the first case it's worth reconnecting, in the second it isn't.
How does the 95:5 rule work, and what does it change for campaigns?
At any given time, only 5% of B2B potential customers are ready to buy. The remaining 95% have no intention of buying in the short term. Ignoring this ratio means wasting budget on audiences that won't convert any time soon.
Research by the Ehrenberg-Bass Institute for the LinkedIn B2B Institute, known as the 95:5 rule, shows that at any given moment only 5% of B2B buyers are actively looking for a solution. The other 95% aren't evaluating anything.
This doesn't mean ignoring the 95%: it means treating them differently. Accounts that are out of market today should be nurtured with content that builds brand familiarity, not with direct sales messages. When they enter the evaluation phase, your brand will already be on their minds.
The ICP helps you understand which accounts in the 95% are worth cultivating: the ones that will be a good fit when they enter the buying phase. Without an ICP, you also nurture accounts that will never convert, or that will convert but then abandon the product within six months.
Three real-world scenarios: how does lead management change for different Italian companies?
The impact of the ICP varies depending on the type of company, the sales cycle and the number of products. Here are three illustrative scenarios to see where the difference comes from.

Illustrative examples: numbers are meant to show the reasoning, they are not real customer data.
| Company type | Typical problem without an ICP | What changes with an ICP | Expected result |
|---|---|---|---|
| B2B management software (SaaS) | Sales closes deals with SMEs that abandon the product after 6 months because it's too complex | The ICP excludes companies below a certain digital maturity and without in-house IT | Higher renewal rate, lower support costs |
| Manufacturing company with two product lines | Marketing generates generic leads, sales doesn't know which product to pitch | A separate ICP per line: different criteria for industry, size and production process | More focused sales reps, more relevant offers at first contact |
| Professional consulting firm | Every prospect looks promising, the pipeline is full, but nobody closes | The ICP defines the profile of a company that already has an allocated budget and an urgent problem | Shorter pipeline, higher close rate |
How do you turn the ICP into actionable data in your CRM and campaigns?
An ICP that lives only in a strategy document changes nothing. It needs to become a field in the CRM, a segmentation criterion in campaigns and an automatic score that guides the daily priorities of marketing and sales.
The source identifies three signals that, combined, give a complete picture of an account:
- Account fit: does the company match the ideal profile for the product?
- Contact fit: is the person engaging relevant to the buying process?
- Engagement: are the people at the account showing real interest?
A contact can be very active on your content but work at a company that isn't a good fit. A perfect ICP account might not yet show enough interest to justify a sales call. Looking at all three signals together keeps you from wasting time on the wrong cases.
- 1 · Marketing + SalesDefine the ICP togetherRun a workshop with your most effective salespeople: which customers renew, expand and are easy to serve? Which ones cost too much in support?
- 2 · Marketing OpsTranslate the ICP into CRM fieldsEvery ICP criterion becomes a structured field in the CRM (e.g. Salesforce, HubSpot). The fit score is calculated automatically.
- 3 · MarketingSegment campaigns by fitPaid campaigns (Google Ads, LinkedIn, Meta) use the ICP score to exclude irrelevant audiences and increase budget on high-fit accounts.
- 4 · SalesPrioritize the pipeline by fit + engagementSalespeople can see in the CRM which accounts have high fit and are showing signs of interest: those are the first to call.
- 5 · Marketing + SalesReview the ICP every quarterThe market changes. The ICP should be updated with data from customers won and lost in the last 90 days.
What tools exist to build and activate the ICP in your campaigns?
You don't need a dedicated tool to get started: a shared spreadsheet and the CRM you already use are enough. Advanced tools add automation and real-time signals, but they come after you define your criteria.
| Tool / Approach | What it does | Suitable for | Indicative cost |
|---|---|---|---|
| Shared spreadsheet + basic CRM | You define the ICP criteria manually and the team applies them case by case | Companies with fewer than 200 accounts in the pipeline | Cost of internal time |
| HubSpot / Salesforce with scoring | Automatically calculates the ICP score for each account based on the fields filled in | SMEs with a structured pipeline and a CRM already in use | Included in the CRM's advanced plans |
| ABM platforms (e.g. 6sense, Demandbase) | Adds external intent signals: which companies are searching for solutions like yours | Mid-market companies with long sales cycles and high-ticket deals | From €2,000/month |
| LinkedIn Campaign Manager with ICP | Lets you target campaigns exactly on the ICP's characteristics (industry, role, size) | Any B2B company with a LinkedIn budget | Pay-per-click, variable |
LinkedIn campaigns targeted on the ICP achieve a 68% higher return on investment than generic targeting, according to LinkedIn data reported by CXL. That figure justifies the time spent building your criteria before you switch campaigns on. To dig deeper into how to optimize your budget on paid platforms, also read how to really measure campaigns beyond CAC.
Why it's a hot topic now: market data on marketing-sales alignment
The topic isn't new, but the data shows the problem has gone unsolved for years. 61% of B2B marketers name lead quality as their top challenge for the third year in a row, a sign that tactical fixes aren't enough.
40% of B2B marketers name lead quality and MQLs (Marketing Qualified Leads, meaning leads that marketing considers ready to hand over to sales) as their most important success metric, according to HubSpot. Yet the definition of an MQL often remains subjective and isn't shared with sales.
Companies with tightly aligned marketing and sales generate 208% more revenue from marketing activities than companies with poor alignment, according to MarketingProfs, cited by The Starr Conspiracy. That's not a marginal advantage: it's the difference between a marketing function that pays for itself and one that multiplies its cost.
To understand how automation can speed up this process, read the practical case study on how to go from 12 to 108 qualified leads a month with marketing automation. And if you're already investing in paid campaigns, also check how changes to budget and targeting on Google Ads don't restart the learning period.
How do you get started: the steps to build an ICP from scratch
The starting point isn't a template, but the customers you've already won. Your best current customers already hold all the information you need to build a realistic ICP.
- Analyze your highest-value customers: not just by revenue, but by margin, renewal rate and ease of management. What do they have in common?
- Analyze your most problematic customers: high support costs, early churn, excessive demands. What do they share? These criteria go on your "non-fit" list.
- Interview your most effective salespeople: what signals do they use to tell whether an account is worth their time? This knowledge is often implicit and has never been written down.
- Define criteria by product: if you sell multiple solutions, repeat the analysis separately for each one.
- Translate the criteria into CRM fields: every criterion must become structured data, not a text note. Only then can scoring be automated.
- Share the ICP with sales and collect feedback: your first ICP won't be perfect. You need a quarterly review cycle using data from closed-won and closed-lost deals.
- Activate the ICP in your campaigns: use the criteria to exclude irrelevant audiences and increase budget on high-fit accounts.
What mistakes are most common when building an ICP?
The most common mistakes aren't about data, but about process: the ICP is built by marketing alone, stays a static document, or is defined too broadly to be useful.

| Error | Why it is a problem | How to avoid it |
|---|---|---|
| ICP built by marketing alone, without sales | Sales doesn't see it as its own and keeps using informal criteria | A joint workshop with salespeople before writing a single line |
| ICP too broad ('Italian B2B SMEs') | It doesn't help you decide where to focus budget: it includes too many accounts | Add at least 3-4 specific criteria that exclude non-fits |
| Static ICP, updated once a year | The market changes, and fit criteria change with it | Quarterly review using data from closed-won and closed-lost deals |
| A single ICP for all products | It hides important differences between segments | Separate score per product in the CRM |
| ICP not translated into CRM data | It stays a document nobody checks in day-to-day work | Every criterion becomes a structured field with an automatic score |
| Ignoring negative signals (non-fit) | You keep chasing accounts that won't convert or will churn | Explicitly define who is NOT a good customer, not just who is |
How do I know if the ICP is working and campaigns are improving?
The metrics to track aren't just lead volumes, but pipeline quality: close rate, sales cycle speed, renewal rate and cost per closed deal.
| Indicator | What it measures | Positive signal |
|---|---|---|
| Lead acceptance rate by sales | How many marketing-generated leads are actually worked by salespeople | Above 70% (from under 50% typical without an ICP) |
| Close rate for ICP vs. non-ICP segments | Whether ICP accounts close more often | A difference of at least 15-20 percentage points |
| Sales cycle speed | How much time passes from first contact to signature | 20-30% reduction for ICP accounts |
| 12-month renewal rate | How many customers renew their contract | Increase over the historical average, especially for ICP customers |
| Cost per closed deal | Marketing + sales budget divided by signed contracts | Decreases over time as targeting improves |
| Average ICP score of the pipeline | Average quality of accounts in the funnel | Quarterly increase after activating the ICP in campaigns |
If you run paid campaigns, it's worth combining these metrics with attribution data. The article on what to ask before increasing your budget on Amazon Ads offers a framework that's useful for other channels too. You'll find more practical ideas in the Marketing and ADV performance section.
So, is building a shared ICP worth it for your company?
It's worth it if you have a B2B sales cycle with more than one salesperson involved, marketing and sales use different criteria to evaluate leads, or you have multiple products with distinct customer segments. The time invested pays off within a few weeks of more precise campaigns.
Somewhere in between: you already have an informal ICP that your salespeople use implicitly. In this case, the job is to formalize it, turn it into data in the CRM and share it with marketing. There's no need to start from scratch.
Not yet if you sell to a very small, homogeneous market, you personally know almost all your potential customers, and marketing and sales are the same person. In that case, the ICP is already in your head and the problem lies elsewhere.
How we see it at SHM
For us, the ICP isn't a marketing exercise: it's the starting point of any acquisition strategy that aims to be measurable. We believe marketing is the science of relationships, built on data and clear goals, and without a shared definition of the ideal customer, the data you collect optimizes the wrong direction.
In the projects where we supported B2B companies on the structure of their digital systems, the most frequent problem wasn't a lack of traffic or leads: it was that no one had ever put in writing which accounts were worth pursuing and why. The result was pipelines bloated with opportunities that never closed, and salespeople frustrated by leads they couldn't make sense of.
Our approach always starts by mapping the fit criteria, by product, not by company, and turning them into operational data in the systems you already use. Only then do we work on campaigns. If you're evaluating how to structure this process, we can help with the strategic integration of AI into qualification flows, with LinkedIn campaigns built on the ICP, or with data processing to build automatic fit scores in your CRM.
Frequently Asked Questions
For a company with an already structured customer portfolio, a half-day workshop with marketing and your most effective salespeople is enough to define the main criteria. Turning them into CRM fields and automatic scores takes one to three days of technical work, depending on the system in use.
Yes, though with different tools. On LinkedIn you can target directly by industry, company size and role. On Meta and Google you work more with custom audiences based on account lists or behaviors. In any case, the ICP defines the exclusion and inclusion criteria that guide campaign setup.
Disagreement is normal and useful: it means you're bringing to the surface implicit assumptions that were never discussed before. The solution is to start from the data of customers you've already won (who renewed, who left, who costs the most in support) rather than from opinions. Numbers cut debates short.
A quarterly review is the recommended rhythm for most companies. Do it with data from deals won and lost over the last 90 days: the patterns that emerge update your criteria based on facts, not impressions.
The ICP describes the ideal type of company (industry, size, technology ecosystem, relationship history). The buyer persona describes the type of person within that company who takes part in the purchase process. You need both, but the ICP comes first: there's no point building a persona if you don't yet know what type of company to look for them in.
Yes, and it's actually the easiest case to build it for. With a single product, the fit criteria are clearer and CRM scoring is easier to implement. The risk without an ICP is the same: chasing accounts that could technically buy but then don't renew or cost too much to manage.
The problem isn't finding more leads. It's to stop chasing the wrong ones.
Related articles
Discover more articles exploring similar topics, selected to offer you a more complete and stimulating perspective. Each piece of content is carefully chosen to enrich your experience.