Market Segmentation and Personas for B2B Outbound Teams

    Short answer

    Cut the market into segments first, size each one, then write one persona per segment and turn both into list filters

    Artem Smirnov
    Artem Smirnov

    Last updated · 11 min read

    Artem Smirnov in a dark suit against a charcoal studio backdrop, next to the line 'SEGMENT FIRST. Personas come second'

    Market segmentation divides a market into groups of companies that share traits you can filter on, such as industry, size, location and the way they buy. A buyer persona describes one person inside a segment who makes or pushes the buying decision. Build the segment first and the persona second. A persona written before its segment exists describes someone you cannot find in a list tool.

    For a B2B company that sells through outbound, the work runs in four steps:

    1. Cut the market into segments, starting from the companies that already pay you.
    2. Count each segment and drop any that cannot fill a campaign.
    3. Write one persona per segment, built around the title that actually signs.
    4. Turn the segment and the persona into list filters: industry codes, size band, geography, titles to include and titles to exclude.

    You can do all four without a survey budget. Your own deal history is the best research you have. If you have no closed deals yet, list the 20 companies you most want to win and treat the segment as an assumption the first campaign tests.

    How do market segments and buyer personas differ?

    A segment is a group of companies, and a persona is one person working in them. The segment is closely tied to your ideal customer profile, and the persona is the B2B buyer persona inside it.

    The ideal customer profile (ICP) sits between the two. It describes the best-fit company inside a segment: the kind your closed deals show closes and pays best.

    SegmentICPPersona
    UnitA group of companiesOne best-fit company profileOne person at that company
    Question it answersWhich companies share a problem and a buying pattern?Which of them are worth chasing first?Who decides or pushes the deal, and what do they care about?
    Where the data comes fromIndustry codes, company size, location, your deal historyYour closed and lost dealsPast projects, sales call notes, campaign replies
    What it does in outboundSets the industry, size and geography filtersSets priority and the offerSets the title filter and the message angle

    Why segment a B2B market before writing a single email?

    Teams change their target market more often than they admit. In a 2023 Gartner Peer Community survey of 200 sales, marketing and product professionals, 56% of the 116 changing their go-to-market strategy planned to identify or adjust their target market (Gartner Peer Community).

    A common mistake is writing the copy before cutting the market. A message written for five kinds of buyer has to stay vague enough for all five. So it lands with none.

    My order of work rarely changes. I start from the companies that already paid the client, then my team finds lookalike companies that share the problem those paying clients had.

    Then I check who really signed off on each of their past projects, because that person, more than the org chart, tells me which title to target. Only after that does anyone open a list tool. Good copy cannot rescue a list built on the wrong segment.

    What are the 5 methods of market segmentation, and which work for outbound?

    The five common bases are firmographic, geographic, behavioral, needs-based and psychographic. For outbound, ask one question of each: can I turn this into a filter in a list tool today?

    MethodWhat it groups byUse for outboundFilter or data field it maps to
    FirmographicIndustry, headcount, revenueFilterable nowUK SIC or NAICS code, employee band, revenue band
    GeographicCountry, region, city, time zoneFilterable nowCountry and region fields, split per rep
    BehavioralBuying history, hiring, recent activity, research signalsPartly filterableActivity filters such as a recent job change or a hiring spike; intent data in large markets
    Needs-basedWhat the company values most when buyingFilterable by proxyHeadcount, purchase volume, buyer role
    PsychographicHow the company buys: risk appetite, build or buy, attitude to priceMessaging onlyOffer angle and the proof you lead with

    Firmographics are the backbone, and the industry part works best when it follows an official code list.

    In the UK that is UK SIC 2007, the standard for classifying businesses by economic activity. The Office for National Statistics says non-government bodies may use it as a common structure (ONS). ONS has since published UK SIC 2026 as the current version, so check which one your list tool uses.

    In the US, NAICS is the federal standard for classifying business establishments. It was adopted in 1997 to replace SIC and can be searched by 2 to 6 digit code (US Census Bureau).

    Needs-based segments are the hardest to filter. B2B International bases its view on more than 2,500 B2B studies. It says needs-based work keeps producing the same few groups: companies focused on price, on quality and brand, on service, or on partnership (B2B International).

    Nobody tags their company "service-focused" in a database. The same source adds that easy-to-see drivers like headcount, purchase volume or the buyer's role often signal needs, so outbound reaches those segments through the proxies.

    Psychographics in B2B have nothing to do with hobbies or lifestyle. They describe how a company buys. Does it take risks on new suppliers? Does it build things in-house or buy them? Does price or certainty win the internal argument? You rarely filter on this. It picks the offer angle and the proof in the first message.

    How do you do market segmentation research without a survey budget?

    The US Small Business Administration lists the direct research methods as surveys, questionnaires, focus groups and in-depth interviews. It warns that direct research can be slow and expensive, while existing sources save time but may fit your buyers less closely (SBA). For an outbound market, your existing sources hold up better than that warning suggests, because they describe your own buyers.

    Start with what you already hold:

    1. Closed deals: company, industry, headcount, country, deal value and who signed.
    2. CRM records: how each deal came in and how long it took.
    3. Sales call notes: the problem each buyer described in their own words.
    4. Lost deals: where you lost, at what stage, and to whom.
    5. Public company registers and industry code lists: to find lookalike companies and tag them by code.
    6. Campaign replies, once outreach starts: who answers, who objects and why.

    A small sample is fine. B2B International points out that even large B2B companies often rely on a hundred customers or fewer for most of their sales movement.

    Digital.gov suggests the same move for personas. Sketch a proto-persona from what the organization already knows, treat it as a set of assumptions, then verify it. In outbound, the campaign does the verifying. Replies confirm the title, and a stream of "not my area" answers tells you the persona sits somewhere else.

    Surveys earn their cost when a decision is expensive to undo, such as launching a product line for a market you have never sold to, and your own records cannot answer the question.

    Is each segment big enough to be worth a campaign?

    Run the sizing check before you write a persona. A segment that fails it does not need one.

    B2B International gives four tests. Segments must be clearly different from each other, big enough to matter, exclusive (each company sits in only one) and easy to identify. For outbound, "easy to identify" means you can pull the segment with codes, size and geography.

    Then count, in two passes:

    1. Companies: industry codes plus size band plus geography. Write the number down per segment. For UK companies, UK Business Counts on Nomis gives the number of live enterprises by employment size band and 5-digit SIC 2007 code. So you can pull civil engineering contractors band by band instead of guessing.
    2. Contacts: run the persona titles across those companies. This is the number your campaign actually spends.

    Compare that count with your monthly reach and the length of a full sequence. If a segment would be used up before you have learned anything from it, merge it with a neighbor or widen the geography.

    Here is the rough rule I work from. To land 30 to 35 booked B2B sales calls in a month, plan to contact at least 10,000 people. If a segment holds fewer contacts, treat the first sequence as a test and judge it on replies.

    Sending to 50 to 100 perfect-fit leads a month does not count as a strategy either. With a list that small, every result depends on whether those few people are busy, ready to buy or interested at all. That is a coin flip. You need quantity and quality.

    Small markets need their own rule. Take a market of roughly 2,000 target companies where just about 10 show intent signals. In a case like that my team skips intent data, because ten accounts are too few to support a campaign.

    The better plan is one offer that fits the whole market, so every company on the list has a reason to reply.

    How do you build one buyer persona per segment?

    Start with one persona per segment. If you think you need two, check first whether the segment is really two segments.

    Build the persona from past projects. Look at who signed, who pushed the deal internally and who could have stopped it. Then name the title the niche really uses. On the lists my team builds, I never target a bare "manager" title, because in most niches the person who signs carries a more specific one.

    One person rarely buys alone. Gartner's 2025 survey of 632 B2B buyers described buying groups of anywhere from 5 to 16 people, drawn from as many as 4 functions. So the persona closes with a decision-power line: who signs, who pushes and who can stop it.

    Here is an illustrative persona for one segment, mid-size civil engineering contractors in the UK and US. It shows the format; the details are invented.

    • Role: operations director or project director.
    • Responsibilities: delivery across several sites, and supplier and subcontractor choices for live projects.
    • Goals: projects on schedule, fewer site incidents, suppliers who do not need chasing.
    • Pains: late deliveries, suppliers who cannot document safety compliance, too many vendors to manage.
    • Decision power: approves suppliers for their own projects; larger contracts also go past the commercial director.

    How do segments and personas turn into list filters?

    A persona only helps outbound once it becomes a search, much like segmenting a cold email list by filters. This template turns one into filters.

    FieldWhat goes in itExample for the civil engineering segment
    Segment nameA plain description anyone on the team understandsMid-size civil engineering contractors, UK and US
    Industry codesCodes from the official national listUK SIC 2007 division 42 (civil engineering); NAICS 237 (heavy and civil engineering construction)
    Size bandEmployee or revenue range taken from your closed deals50 to 500 employees
    GeographyCountries, each assigned to one repUK to one rep, US to another
    Include titlesThe titles that signed or pushed past dealsManaging Director, Operations Director, Project Director, Head of Operations, Commercial Director
    Exclude titlesGeneric titles that cannot signAssistant, Associate, a bare Manager
    OfferOne line written for this segmentA short call about the one problem this segment feels most

    My team strips Assistant, Associate and bare Manager titles out of every list we build, because they pad the count without adding buyers. A specific title such as contracts manager can stay if your past deals show that person signs.

    Then build the list in this order:

    1. Fill one template row per segment.
    2. Translate the codes and titles into your list tool's search fields, using boolean strings (title searches with AND, OR and NOT) to catch title variants.
    3. Pull the list and delete companies outside the size band.
    4. Run every address through verification before the first send.
    5. Audit the spreadsheet by hand before anyone writes copy.

    Step 5 is how I take a lead list apart before sending. The difference between an ICP and a buyer persona matters most at this step.

    How many segments and personas should a B2B company run?

    B2B International puts the typical B2B study at 3 or 4 segments, against 10, 12 or more in fast-moving consumer goods. In outbound I start with one segment and open the next only once the first is working.

    Market sizeSegmentsPersonasOffer and data
    Small: a few thousand target companies or fewer11, with a second title in the filter so you reach more than one person per companyOne broad offer for the whole market; no intent data
    Mid: room to split without any segment failing the sizing check2 or 31 per segmentOne offer per segment
    Large: intent signals show on a useful share of accounts3 or 41 per segment, a second only where a different title signs in past dealsIntent data optional

    Even in a large market, run segments in sequence, with account tiers deciding who gets the heaviest research. I ask the founders I work with to commit a full year to a single offer and a single audience before we open the next segment. The reasoning is that one offer and one audience give you clean data. No single count fits every business. Let the sizing check set the number.

    This week, export your last 20 closed deals with industry code, headcount, country and the title that signed. Your first segment is usually sitting in the rows that repeat. Once the segment is set, B2B lead generation done for you covers the campaign side.

    If you have a segment in mind and are unsure the persona behind it is right, bring your last 20 closed deals to a call and we will check it.

    Questions founders ask about segments and personas

    What are good examples of psychographics in B2B?

    Treat these as hypotheses read from a company's public behavior. A risk-averse buyer may publish supplier-approval steps or ask for references first. A build-it-ourselves company is hiring for the function you sell. A price-led buyer often runs formal tenders. Store the label in a column next to the title list, then open with the proof that fits that label.

    Is there a customer persona template or PDF I can copy?

    Skip the designed PDF. Keep the persona as extra columns in the same spreadsheet as your segment template, so the title list and the message notes travel with the filters. When a campaign reply changes your view of the buyer, you edit one cell.

    How often should segments and personas be updated?

    Digital.gov advises treating personas as documents that keep getting updated. In outbound, review them after a full sequence has run on a segment. If replies keep coming from a title you did not list, add it. If one listed title never answers, test removing it. Revisit the segment itself when a new size band or country starts closing deals.

    Can one persona belong to two segments?

    The job title can, the persona usually cannot. A head of operations at a civil engineering contractor and one at a software company share a title, but not the problem, the budget or the proof that convinces them. Rule: one persona per segment, even when the title filter is identical.

    Want to see what we would build for your company?

    I am Artem Smirnov. I help established B2B companies build and run the go-to-market system behind their pipeline, from outbound to AI search visibility. Book a call and we will see what fits yours.

    Artem Smirnov
    Artem Smirnov

    I help established B2B companies build and run their go-to-market systems. 9 years in outbound, 10,000+ campaigns.