In recruitment, ICP means two things: Ideal Customer Profile and Ideal Candidate Profile. Here is how to build both, and why data quality makes the difference.
In recruitment, ICP stands for two different profiles: the Ideal Customer Profile (which clients fit your agency) and the Ideal Candidate Profile* (who is the perfect hire for this role).
* An ICP is not a persona and not a job ad. It is a data-based profile of where your success comes from, used to say yes or no faster.
* You build an ICP by looking at your best existing clients or placements, finding the patterns, and splitting them into must-haves, nice-to-haves and knock-outs.
* An ICP goes stale. B2B databases lose roughly 22.5% of their value per year, so without a refresh your profile works against you within a year.
* The quality of your ICP depends entirely on the quality of the data underneath it. Gut feeling in, gut feeling out.
In recruitment, ICP means two things: the Ideal Customer Profile, the type of client that fits your agency best, and the Ideal Candidate Profile, the perfect candidate for a specific role. The abbreviation is borrowed from B2B sales, and in both cases it describes a data-based ideal profile. Only the subject changes.
The Ideal Customer Profile describes the type of client your agency serves best: which sector, which company size, which type of roles, which way of working. You use it to focus your business development and to decide which leads to chase and which to let go.
The Ideal Candidate Profile describes the perfect candidate for a specific role: which hard skills, which experience, which context (availability, salary, location) and which soft qualities. You use it to source, screen and match.
Put simply: one profile is about who you serve, the other about who you place. An agency that wants to grow needs both. Do not confuse them, because they are used at different moments and by different people.
The Ideal Customer Profile and the Ideal Candidate Profile share the same abbreviation but serve a different purpose. One focuses your business development on the right clients, the other focuses your sourcing and matching on the right candidate. They are used by different people and at different moments. The table below sets them side by side.
| Dimension | Ideal Customer Profile | Ideal Candidate Profile |
|---|---|---|
| Who or what you describe | The type of client | The perfect candidate for a role |
| Purpose | Focus business development | Focus sourcing and matching |
| Used by | Sales, account management, leadership | Recruiters, sourcers, consultants |
| Typical fields | Sector, headcount, region, role volume, buyers | Skills, experience, salary, availability, motivation |
| When you build it | Per market segment or service | Per role or job family |
| When you refresh it | Every 6 to 12 months | At every new intake |
Notice that the two overlap. A strong Customer Profile tells you which roles you will fill most often, and that feeds your Candidate Profiles. In reverse, your best placements reveal which clients are worth the most. They are two sides of the same coin.
An Ideal Customer Profile answers one question: with which type of client do you earn the most, with the least friction? Not the client who shouts loudest, but the client who fits your specialism, your fee model and your way of working.
Start with your own numbers, not a brainstorm. Pull your ten most profitable assignments from the past year out of your CRM and look for the overlap. You often spot patterns you did not expect: a specific company size, a particular sector, or clients who just went through a funding round or a reorganization.
The fields you usually want to capture:
Firmographics.* Sector, headcount, revenue band, region, and whether it is a corporate or a scale-up.
Recruitment signal.* Annual role volume, type of roles (scarce versus volume), and whether they hire structurally or occasionally.
Buying behavior.* Do they work with a preferred supplier list, how long does the decision take, and who actually decides. Gartner shows that a complex B2B purchase involves six to ten decision makers on average, so “the contact” is rarely the only voice.
Fit signal.* Do they value quality over the lowest price, and does their way of working match yours.
Once this profile is sharp, new business gets a lot simpler. You know which leads to prioritize and which to politely pass on. And your intake with a new client improves, because you know exactly which signals to ask about. The data from those intake conversations is worth its weight in gold, provided you capture it instead of keeping it in your head. More on that below.
The Ideal Candidate Profile is the data-based profile of the candidate who will genuinely do a specific role well, and stay. Where the Customer Profile lives at agency level, this one lives at role level. You split attributes into must-haves, nice-to-haves and knock-outs, and you use it to source, screen and match.
That last part matters. Quality of hire is the metric recruiters worldwide rate as the most important, and yet almost every vacancy starts with a job description full of generalities. A Candidate Profile is sharper than that. It splits attributes into three categories:
The difference between a Candidate Profile and a job ad is that the profile is honest. The job ad is written for the candidate and sells the role. The profile is written for you and separates candidates from suspects. A good intake with the hiring manager is where that profile is born, and where most failed placements become predictable in advance.
Beyond hard criteria, context belongs in there: desired salary or rate, availability, willingness to travel, and why earlier candidates did or did not work in this role. Exactly that kind of information rarely comes from a CV. It comes from conversations. See our pillar guide on recruitment intelligence for how to get that conversation data into your CRM in a structured way.
You build an ICP in five steps: gather your best cases, find the patterns, split them into must-haves, nice-to-haves and knock-outs, put it on one page, and refresh every six months. The method is the same for both profiles. Only the input differs: clients for the Customer Profile, placements for the Candidate Profile.
The most common mistakes are a wish list of nothing but must-haves, a profile too broad to filter anything, confusing it with the persona or the job ad, a profile you never update, and underneath it all: building on gut feeling instead of data. The same five show up at nearly every agency.
The wish-list trap. The profile describes the client or candidate you want, not the one you have proven success with. Recognizable by ten must-haves and not a single knock-out.
Too broad to be useful. “Companies in the region that are hiring” is not an ICP. If your profile describes 80% of the market, it helps you with no decision at all.
Confusing it with the persona or the job ad. A buyer persona describes the individual decision maker; an ICP describes the organization. A job ad sells; a Candidate Profile filters. Different documents, different purpose.
Never updating it. This is the silent killer. A bad hire quickly costs around 30% of first-year salary according to widely cited research, and stale profiles cause exactly those misses because you keep filtering on old assumptions.
Built on gut feeling instead of data. And this is the most important one, because it sits underneath all the others.
Data quality makes or breaks your ICP, because the profile is nothing more than a summary of your data. If that data is messy, incomplete or stale, so is your profile. Three things decide the outcome: completeness, reliability and traceability of every data point. Gut feeling in, gut feeling out.
That is not a theoretical risk. B2B contact databases decay by roughly 22.5% per year: people change jobs, companies merge, rates shift. Build your Customer Profile on numbers from eighteen months ago and you aim at a market that no longer exists. And with candidates it is no different: a profile based on half-filled CV fields and loose notes misses exactly the context that makes the difference.
This is where the quality of your underlying system comes in. Three things decide whether your data produces a reliable ICP:
Completeness.* Do the data points from your conversations actually land in your CRM, or do they stay in someone’s head and in loose notes? Our CV parsing and smart CRM data entry make sure conversation data lands in a structured way on the right fields.
Reliability.* Do you know which fields are correct and which are a guess? A validation system that flags per field whether something is certain (green) or doubtful (orange) stops you building an ICP on wrong assumptions.
Traceability.* Can you see where a data point came from? If your profile says a client has “many scarce roles,” you want to trace that assumption back to a real intake conversation, not a vague memory.
This is exactly where recruitment-native tooling separates itself from a generic database. Simply captures conversations across every channel, validates the data per field, and keeps traceable where each detail came from. That keeps the foundation under your ICP clean. On how fragmented data wrecks your productivity, read more in ending fragmented recruitment data, and on the conversation side in conversation intelligence for recruitment.
An ICP is not a document you make once and drop in a drawer. It is a filter you use every day: when picking up a lead, when writing a vacancy, when assessing a candidate. Two profiles, both data-based, both short enough to remember, and both refreshed periodically.
The Ideal Customer Profile focuses your business development. The Ideal Candidate Profile focuses your delivery. And the quality of both stands or falls with the data underneath. So do not start with the profile, but with the question of whether your conversation data sits complete, reliable and traceable in your system.
Simply is the AI-native ATS that makes that possible, because the conversation and the record are the same system: meeting bots for Meet and Teams, a desktop and mobile app, and VOIP for phone calls, with per-field validation and traceability back to the source. ISO 27001 certified, GDPR compliant, and without using your data to train models. Want to see how clean data keeps your ICP sharp? See how Simply works.
About the author
Remo Vloet is a co-founder of Simply, the AI Operating System for recruitment agencies: inbox, meetings, sourcing, CV parsing, search and matching, documents and automation in one system. With a background in building complex software, he contributes to the technical vision behind Simply.
LinkedInArticles by Remo VloetICP has two meanings in recruitment. The Ideal Customer Profile describes the type of client that fits your agency best and steers your business development, while the Ideal Candidate Profile describes the perfect candidate for a specific role and steers your sourcing and matching. Which of the two is meant depends on the context: sales or delivery.
ICP stands for Ideal Customer Profile or Ideal Candidate Profile. The abbreviation originally comes from B2B sales, where ICP always means Ideal Customer Profile. In recruitment the candidate variant was added because recruiters apply the same principle (a data-based ideal profile) to the candidate side.
An ICP describes the organization (or, on the candidate side, the role and the profile). A buyer persona describes the individual within that organization: the hiring manager, the HR director, the procurement lead, each with their own drivers and objections. You use the ICP to decide which companies to approach, and the persona to decide how to speak to the people inside them.
You refresh an Ideal Customer Profile every six to twelve months, because markets and your own client base shift. B2B data decays by roughly 22.5% per year, so standing still means falling behind. You sharpen an Ideal Candidate Profile at every new intake, because every role has slightly different requirements.
No, but they are connected: the intake is the conversation where you gather the information, and the Candidate Profile is the structured result of that conversation. A good intake without a profile stays stuck in someone's head, and a profile without a good intake is a guess. The two reinforce each other only when the conversation data actually lands validated in your system.
As a rule of thumb: at least ten real cases (your best clients or your best placements) to see reliable patterns. Fewer than that and you build on chance. More important than the quantity is the quality: ten fully filled, validated profiles say more than a hundred half-empty records where you do not know which fields are correct.
A QBR (Quarterly Business Review) is the quarterly review with your client. How to run one that retains and grows the account, and why data quality decides it.
Remo Vloet9 min.
AI screening tools read and rank candidates in minutes. Here is how the main types work, where bias and compliance risk creep in, and how to choose one.
Remo Vloet9 min.
CV parsing or CV formatting to house style? The difference, which tools actually do it, and why house-style formatting saves agencies time and trust.
Remo Vloet9 min.
Find out how Simply can completely evolve your workflow.No slides, just product.