Teams often use ICP and persona as if they mean the same thing, then wonder why their targeting is vague. They answer two different questions. The ideal customer profile answers which companies are worth pursuing. The buyer persona answers who inside those companies you need to convince and what they care about. Get the ICP wrong and you spend money on accounts that will never buy or will churn. Get the personas wrong and you reach the right companies with the wrong message. This article gives you a fill-in template for each, with blanks in square brackets, and shows how I turn them into CRM fields so they drive routing, scoring and reporting rather than sitting in a slide deck.
ICP vs persona: what each one is for
The ICP is a company-level filter. It should be specific enough that two people applying it to the same list would make the same calls. It covers industry, size, geography, technology, the business trigger that creates demand, the economics of a good deal and the reasons to walk away. It is used for account selection, list building, ad targeting, lead scoring on fit and territory design. The persona is a person-level description. HubSpot's own knowledge base defines personas as fictional, generalised characters that capture the needs, goals and behaviour patterns of your customers. A persona is used for messaging, content planning, sales discovery and deciding who belongs in a sequence. The order matters. Write the ICP first, from data, because personas only make sense inside companies that fit. A CFO persona at a 20-person company and a CFO persona at a 2,000-person company are different people with different problems. If you sell into one segment you may need two to four personas. If you sell into several segments, write personas per segment rather than one generic set.
Where the inputs come from
Both templates should be filled from evidence. For the ICP, export your closed-won deals from the last 12 to 24 months with company properties attached, and sort by deal size, sales cycle length and, if you have it, retention or expansion. The top segment by those measures is your starting ICP. Then export closed-lost and churned accounts and note what they have in common; those patterns become disqualifiers. For personas, the best inputs are recorded sales calls, win and loss interviews, support tickets and the exact phrases customers use in reviews. Five to eight conversations per persona usually surface the repeated themes. Survey data helps with scale, but interviews give you the language. If your CRM does not capture industry, employee range or the trigger that started the deal, the exercise will expose that gap. That is useful in itself. My CRM data dictionary template covers how to define those fields so the next version of this exercise is a report, not a research project.
The ICP template
Template text. Segment name: [short label, e.g. Mid-market logistics SaaS buyers]. Industry: [included industries]; excluded: [industries]. Company size: [employee range] and/or [revenue range]. Geography: [countries or regions you can sell and support in]. Technology: [systems they must use or have replaced, e.g. uses Salesforce, outgrew spreadsheets]. Trigger events: [events that start a buying process, e.g. new head of sales, funding round, failed audit, contract renewal with competitor]. Business problem: [the problem in the customer's words, one sentence]. Deal economics: typical deal size [range], typical sales cycle [range], buying group size [number of roles]. Signs of a strong fit: [three observable signals]. Disqualifiers: [three to five reasons to walk away, e.g. no budget owner, below minimum seats, regulated region you cannot serve]. Evidence: [number of closed-won deals this is based on, date range]. Owner and review date: [name], [date]. Keep it to one page. If a field cannot be filled from evidence, write 'unknown' rather than guessing; the unknowns become your research list.
The buyer persona template
Template text. Persona name and role: [e.g. Revenue leader, VP Sales or CRO]. Segment: [which ICP segment this persona belongs to]. Role in the deal: [decision maker, budget holder, champion, evaluator, user or blocker]. Reports to: [title]. Measured on: [the two or three metrics this person is judged by]. Goals this year: [in their words]. Problems that trigger a search: [specific situations, not categories]. What they have tried already: [alternatives, including doing nothing and spreadsheets]. Objections and risks they worry about: [e.g. implementation time, adoption, switching cost, internal politics]. Information sources: [communities, publications, peers, analysts, LinkedIn voices they follow]. How they evaluate: [what proof they want, who else they involve, what a trial or pilot looks like]. Phrases they use: [three to five direct quotes from calls]. Phrases to avoid: [jargon that lost deals]. Content that helps at each stage: [first touch, evaluation, decision]. Evidence: [number of interviews and calls this is based on]. The quotes are the most valuable line. They are what copywriters and SDRs will actually reuse.
Turning the templates into CRM fields
A template that lives only in a document stops being used within a quarter. I convert both into fields. For the ICP, create a company-level fit field. In HubSpot, the default Ideal customer profile tier property on companies works, and HubSpot provides a workflow template that sets the tier from annual revenue, which you can extend with industry and employee range. In Salesforce or Zoho CRM, create an ICP tier picklist on the Account object and set it with a flow or workflow rule. For personas, HubSpot has a default Persona contact property you can edit in property settings; Salesforce and Zoho users can add a persona picklist on Contact or Lead. Set persona from job title rules at creation, and let reps correct it. Add a Buying role field as well, because persona and role in a specific deal are not the same thing. Once these fields exist, you can route by tier, score on fit separately from engagement, and report win rate and deal size by ICP tier and persona. That report is how you know whether the ICP is right.
Keeping the ICP and personas current
Review both on a fixed schedule, quarterly at this stage, and after any significant change such as a new pricing tier, a new product or a move into a new region. The review takes an hour if the fields above exist. Pull win rate, deal size, sales cycle and early retention by ICP tier and persona for the last quarter. If Tier 2 accounts are winning at a similar rate to Tier 1, your tier boundaries are wrong. If one persona keeps appearing in lost deals as a blocker, that persona needs its own content and discovery questions. Watch for drift in disqualifiers too; a reason that was valid at $2M ARR, such as no support for a region, may no longer apply at $8M. Record each change with a date in the template so you can explain shifts in reporting later. The goal is not a perfect document. It is a shared definition that sales, marketing and customer success all apply the same way, and that changes when the data says it should.
Sources
HubSpot Knowledge Base, Create and edit personas: https://knowledge.hubspot.com/properties/create-and-edit-personas HubSpot Knowledge Base, Get started with account-based marketing in HubSpot (Ideal customer profile tier and Buying role properties): https://knowledge.hubspot.com/branding/get-started-with-account-based-marketing-in-hubspot
FAQ
An ICP describes the type of company worth selling to, using firmographics, triggers, deal economics and disqualifiers. A buyer persona describes a person inside those companies: their role, goals, problems, objections and how they evaluate. The ICP decides which accounts; the persona decides what you say and to whom.
Usually two to four per ICP segment: typically the economic buyer, the champion who runs the evaluation and the main user. Add a persona only when it needs different messaging or content. More personas than your team can write for just dilutes the work.
Start from current customers, specifically the closed-won deals with the best size, speed and retention. A target market you have not yet won can be a hypothesis to test as a separate segment, but it should not replace the evidence-based ICP that drives your pipeline today.