Most B2B organisations have some version of an Ideal Customer Profile. The problem is that most of those ICPs describe an average customer rather than an ideal one - a broad demographic profile built from gut feel rather than data analysis, and defined loosely enough to include nearly everyone the company has ever sold to.
That kind of ICP produces diffuse targeting, generic messaging, and wasted spend. A sharply defined ICP - built from patterns in your best customers, not your broadest - concentrates every resource on the accounts most likely to buy, derive maximum value, and remain long-term. It's the foundational input into every channel decision, content strategy, and sales motion you run.
What an ICP is - and what it isn't
An Ideal Customer Profile is a data-driven description of the type of company most likely to: buy your solution, derive significant value from it, remain a customer for the long term, and expand their investment over time. It is defined primarily by firmographic attributes - industry, company size, revenue, geography, technology stack - layered with behavioural and situational signals that further sharpen fit.
What it isn't:
- A description of your average customer - Your average customer includes a mix of good-fit and poor-fit accounts. Your ICP should reflect only the former
- A wishlist of your largest potential deals - Target market aspiration and ICP are different. Your ICP must reflect what you can actually win and retain, not what you'd like to win
- A static document - ICPs drift as your product evolves, your market changes, and your customer base grows. Review it at least annually
- The same as a buyer persona - Your ICP describes a company type; your buyer personas describe the individuals within those companies (see below)
Your Ideal Customer Profile answers the question: "What type of company is most likely to buy our solution, get substantial value from it, stay with us long-term, and expand?" It is built from analysis of your best existing customers - not from assumptions about who should buy, or from a broad description of your total addressable market.
ICP vs buyer persona: the distinction that matters
The ICP and buyer persona operate at different levels and answer different questions. Conflating them leads to targeting errors that are expensive to diagnose and fix.
Your ICP describes a company. It tells you which accounts to target in your ABM programmes, which accounts to enrich in your CRM, which companies to build account lists from, and which industries to prioritise in your paid media targeting. It operates at the account level.
Your buyer personas describe individuals within your ICP accounts. They tell you what content each stakeholder needs at each stage, what objections they raise, how they research independently, and what evidence moves them to a decision. They operate at the person level.
"Your ICP tells you which doors to knock on. Your buyer personas tell you who answers each one - and what they need to hear to let you in."
In practice: build your ICP first, then build buyer personas for each meaningful buying committee role within those ICP accounts. A SaaS company targeting CFOs at mid-market professional services firms has one ICP (mid-market professional services, specific firmographic parameters) and multiple personas within it - the CFO who controls the budget, the IT lead who evaluates technical fit, the operations manager who will live with the product.
The five dimensions of a strong ICP
A complete ICP combines five categories of firmographic and behavioural attributes. Each dimension narrows the target universe and improves the relevance of everything that follows - from ad targeting to content themes to sales qualification criteria.
The most fundamental ICP filter. Defines the regulatory environment, buying culture, language, and competitive landscape your solution will operate in. Resist the temptation to list every industry - the narrower, the more targeted your positioning and content.
Determines deal complexity, budget authority, and decision-making structure. SMBs and enterprises buy in fundamentally different ways - the same solution often needs a different motion, different content, and different pricing for each. Pick the tier your solution genuinely fits best.
Determines language, regulatory context, sales coverage model, and time zone requirements. For UK-based agencies, geography also affects GDPR compliance in outreach. Be specific - "EMEA" is not a target geography, it's a region containing dozens of distinct markets.
Tech stack signals integration readiness, operational maturity, and budget availability. It's also highly targetable - data providers like Bombora, ZoomInfo, and Clearbit can identify accounts by installed technology. A strong tech stack signal is one of the most reliable ICP qualifiers available.
The most powerful ICP dimension for prioritisation. Situational signals indicate that an account is currently in a buying window - not just theoretically a good fit. Intent data platforms surface these signals at scale, enabling marketing and sales to concentrate effort on accounts most likely to convert now.
Defining who you're not targeting is as valuable as defining who you are. Negative ICP attributes prevent sales and marketing from wasting effort on accounts that look superficially attractive but consistently churn, require disproportionate support, or never expand. Build this from analysis of your worst customers as rigorously as you build your best.
Not confident your ICP reflects your best customers rather than your broadest market? Our GTM Audit includes an ICP analysis - comparing your current targeting to patterns in your best accounts.
Explore Demand Generation →How to build your ICP from existing customer data
The most common mistake in ICP development is building it from assumptions - "our ideal customer is a mid-market SaaS company" - rather than from analysis of actual customer data. Assumption-based ICPs feel accurate because they confirm what the team already believes, but they rarely surface the specific attributes that actually predict success.
What a completed ICP looks like
Below is an example ICP for a B2B marketing agency - illustrating the specificity that produces meaningful targeting improvements. Note how many generic "B2B marketing agency clients" this excludes. That exclusion is the point.
Example ICP: B2B Marketing Agency
Mid-market B2B SaaS - UK and DACH - Series B to D
- B2B SaaS or tech-enabled services
- 50-500 employees
- £5M–£50M ARR or equivalent revenue
- UK, Germany, or Austria HQ
- Series B, C, or D funding stage
- Dedicated marketing function (1+ FTE)
- CRM in place (HubSpot or Salesforce)
- Hiring for Head of Marketing or CMO
- Recently raised Series B/C round
- Entering a new geographic market
- Intent data: researching demand gen, ABM, or pipeline tools
- Flat or declining MQL volume
- Sales team growing faster than marketing
- Consumer-facing or B2C companies
- Under 30 employees or pre-Series A
- No internal marketing function
- ACV under £5k (economics don't work)
- Highly regulated (pharma, financial advice) without compliance resource
Using your ICP for account fit scoring
Once your ICP is defined, it becomes the foundation for account fit scoring - a systematic way to prioritise which accounts marketing and sales should focus on. Fit scoring assigns weighted scores to each ICP dimension, producing a ranked list of accounts by their probability of being a good-fit customer.
Tier 1 accounts (highest fit score + active intent signal) receive the most concentrated resources: personalised outreach, direct mail, bespoke content, and senior sales attention. Tier 2 accounts receive programmatic ABM - targeted display, LinkedIn advertising, and content nurture. Tier 3 accounts receive only broad marketing touch until an intent signal elevates their priority.
Activating your ICP across channels
An ICP only creates value when it's consistently applied across every function that touches a prospect. The most common failure is building a sharp ICP and then only using it in sales qualification - while marketing continues to target broad audiences and spend budget on accounts that would never convert.
Paid media targeting
LinkedIn's targeting capabilities allow direct application of firmographic ICP criteria: industry, company size, job seniority, function, and - for Matched Audiences - specific company lists. Your ICP should be the direct source of your LinkedIn campaign targeting parameters. If your ICP says "Series B-D B2B SaaS, 50-500 employees," your LinkedIn targeting should reflect that with company size and industry filters - not a broad "technology professionals" audience.
ABM account selection
Your account selection for ABM programmes should flow directly from your ICP fit scoring - Tier 1 and Tier 2 accounts only. ABM resources are too expensive to deploy against low-fit accounts; the economics only work when you're targeting accounts with high probability of converting to high-LTV customers.
Content strategy
Knowing your ICP's industry, size, and situational triggers tells you which pain points to address in content, which terminology to use, and which use cases to lead with. A content strategy built around a generic ICP produces generic content; one built around a specific ICP produces content that resonates precisely with the accounts you're trying to reach - improving both organic search relevance and paid media engagement.
Sales qualification
Every inbound lead and outbound prospect should be scored against the ICP before entering the sales process. Using ICP fit scores in CRM qualification (BANT, MEDDIC, or equivalent) prevents sales time being consumed by low-fit accounts that feel like opportunities but rarely convert - or convert to customers that churn.
A completed ICP: what it actually looks like
Most ICP documents are either too vague ("mid-market B2B SaaS companies") or too long (a 40-slide deck nobody reads). A working ICP fits on one page and is specific enough that any salesperson can immediately tell whether a prospect qualifies or not.
Tiering your ICP: not all ideal customers are equal
A single-tier ICP treats every qualifying company as equally worth pursuing. In practice, there is significant variation in conversion probability, sales cycle length, and LTV even within a well-defined ICP. Tiering allows you to allocate resources proportionally to opportunity quality.
Accounts that match every dimension of the ICP with multiple confirmed trigger events and no disqualifiers. Typically 50-150 named accounts. Receive full ABM treatment: personalised content, direct mail, executive engagement programme, sales-marketing joint pursuit. Marketing spend per account: £500-£2,000+.
Accounts that meet core ICP criteria with one or two trigger events. Typically 300-500 named accounts. Receive targeted digital advertising, persona-level content, SDR sequence with modest personalisation. Sales and marketing coordinate on timing but not on full account strategy.
All remaining accounts that match core ICP criteria without specific trigger events. Could be 1,000-5,000 accounts. Receive programmatic advertising and demand generation content at scale. No personalisation beyond persona-level. The purpose is building brand awareness so that when a trigger event occurs, your brand is already known.
Companies that don't meet ICP criteria. The discipline to turn away these opportunities — especially when they approach inbound — is one of the most valuable and difficult things a revenue team can develop. Every out-of-ICP deal costs disproportionate sales time and produces below-average retention and expansion revenue.
Keeping your ICP current: the quarterly review
An ICP defined once and never revisited is a liability, not an asset. As your product evolves, as you win deals in new verticals, and as market conditions shift, the ICP should evolve with it. The quarterly ICP review takes 2 hours and keeps the targeting sharp.
What firmographic, technographic, and trigger event patterns appear most frequently in closed-won deals? Are there new verticals or company types converting that weren't on the original ICP? Are any original ICP segments underperforming? Let the data update the hypothesis.
If you're losing deals predominantly to a specific competitor in a specific segment, that segment may be less ICP-fit than it appeared. If loss reasons cluster around pricing in a specific company size band, the ACV model may not work for that tier. Closed-lost data is one of the most underused ICP calibration inputs in B2B.
Which customer segments are expanding? Which are churning? High churn in a specific firmographic segment is a signal that the ICP is wrong for that segment — even if those companies are converting at a reasonable rate. Sustainable ICP targeting requires optimising for LTV, not just acquisition.
Need help building or refining your ICP? Our GTM Audit includes a structured ICP workshop that analyses your existing customer data and builds a tiered account model for your ABM and demand generation programmes.
Explore ABM Services →- Your ICP describes the type of company most likely to buy, get value, stay, and expand - built from analysis of your best customers, not assumptions about who should buy
- ICP and buyer persona are different: ICP is the company; buyer persona is the individual. Build the ICP first, then personas for each meaningful buying committee role within ICP accounts
- A strong ICP covers five dimensions: industry, company size, geography, technology stack, and situational/behavioural signals - plus a negative ICP of disqualifying attributes
- Build your ICP from your top 20% of customers by commercial value - not your average customer, and not wishlist aspiration
- Use your ICP for account fit scoring to tier your total addressable market and concentrate resources on the accounts with the highest probability of converting
- Apply the ICP consistently across paid media targeting, ABM account selection, content strategy, and sales qualification - not just in one function
Our account-based marketing programmes start with ICP analysis and fit scoring - building the account universe that concentrates your investment where the probability of success is highest.
Explore Account Selection & ICP →