How to Build B2B Personas That Don’t Suck (and Actually Help You Sell)

Rewritten September 2026. Same argument, same examples, fewer dashes.

Most B2B personas are bollocks. You’ve seen them. “SaaS Steve is a 35-year-old who enjoys cycling and artisan coffee. He works in a fast-paced tech company and values innovation.” Cheers, Steve. That tells me nothing I can use on a landing page or in a sales call, and it’s the format of roughly every persona deck I’ve inherited.

It was never meant to help, to be fair. Most personas get built because someone senior said “we need personas”, so somebody in marketing put stock photos in a deck, guessed at some goals and filed it under strategy. Nobody in sales has opened it since. I’ve done this myself, early on, and I’ve found it waiting for me at every company I’ve joined, so this isn’t a lecture from a great height. It’s a description of the default.

A persona that earns its keep is a different object. It’s a record of how a specific kind of buyer decides to buy your thing, and what stops them. Build that and the messaging gets tighter and sales calls open on the right problem. This is how to do it: what to research, what to include and what to leave out, what sales wants from the document, two worked examples, and how to lay it out so it gets used. There’s some swearing, it’s B2B, and it’s written with software companies in mind, because that’s where I’ve spent the last ten-odd years.

What a good B2B persona is built on

A persona is a decision guide. The character-sheet version, whether they use Slack or Teams, whether they’ve got a dog called Bingo, whether they do woodworking at the weekend, doesn’t tell you anything about the deal, so none of it goes in. What does go in is five things.

1. Their role and influence in the buying process

“Job title: Head of Marketing” is a LinkedIn search, not a persona. You need what they do day to day, what they’re accountable for, and who they have to convince when it comes to spending money. Are they the decision-maker, the gatekeeper, or the poor sod who’s been asked to “do a comparison of tools” with no budget authority at all? That last one is more common than any deck admits, and it changes how you talk to them.

2. Their goals

“Grow the business” doesn’t count. You want the goals they’re asked about in performance reviews and the targets they’re chasing before quarter end so they don’t get grilled in Monday’s standup. A hundred qualified leads a month. Churn down 15% by December. That sort of thing. The more specific the better, and measurable if you can get it.

3. Their pain

This is where the money is. If you can describe someone’s problem better than they can, you have their attention and probably their trust. You’re looking for what frustrates them, slows them down, and costs them budget, time or credibility: constant firefighting with no time for anything strategic, a stack that breaks every other week, leads handed to sales and never heard of again, a budget that gets trimmed every time revenue wobbles. When you know the pain properly you can show where the product fits, and it stops sounding like a feature list.

4. What makes them say yes, and what makes them say no

Most personas fall over here. They record what the buyer wants and skip what gets in the way, and you need both. What tips them into booking a demo or starting the trial, and what makes them bin you after the second email. They won’t look at you without a Salesforce integration. They’re frightened of a six-month implementation. They’re comparing you with a better-known competitor and don’t want to be the one who picked wrong. Their boss used your competitor at the last job. Knowing this lets marketing and sales deal with the objection before the buyer raises it.

5. Where they go for answers

You have to meet people where they already are. If your buyer spends evenings on Stack Overflow and Reddit, LinkedIn whitepapers and webinars will bounce off. If they want reviews, a demo and a comparison matrix, give them those. What content do they trust, who influences them, where do they go when something breaks. Not everyone wants a newsletter. Work out what they do read and plan the content around that.

How to build one

Time to get practical. If a persona is going to be worth more than the paper it’s printed on, the guesswork has to go and the research has to come in.

Step 1: do the research

No shortcuts. If your persona research starts and ends with the marketing team in a room saying “I think our buyer cares about efficiency”, you’ll end up with a tidy pile of assumptions. You need real data, which means going to the people who have it.

• Customer interviews. Your best-fit customers. Why they bought, what nearly stopped them, what’s changed since.
• Churned customers. Brutal and useful. What didn’t work and what made them leave.
• Sales and support. They’re on the front line. They know the questions, the objections and the last-minute wobbles better than anyone.
• CRM data. Who actually buys. Roles, industries, deal sizes, how long the cycle ran.
• Web analytics. Which content gets traffic, where people arrive from, what they bounce on.
• Review sites and forums. G2, Reddit, the comments under LinkedIn posts. All of it is insight if you’re reading it.

And don’t stop at job titles. Your CFO persona might carry the same title across five companies and have five different sets of problems, because a CFO at a 40-person startup and a CFO at a 2,000-person company aren’t doing the same job.

Step 2: find the patterns

Once you’ve got a pile of notes, you’re looking for clusters: groups of buyers who behave the same way for the same reasons. Startup CTOs who hate complicated onboarding. Mid-market marketing managers who are sick of leads that don’t convert. Enterprise IT directors who lose sleep over compliance. Finance leads at 200-person companies who’ve been told to consolidate tools. That sort of thing.

The aim isn’t 15 personas with tiny differences. It’s three to five that cover the segments that matter without making the campaign planning meeting miserable. If you find yourself arguing about “Freelancer Freddy” versus “Solo Consultant Simon”, you’ve gone too far in. Every set I’ve built started with six or seven and ended with three, and the first version was wrong every time in the same way: too many of them, and too polite about the pain. It usually took a quarter of sales ignoring it before anyone noticed.

Step 3: build the profile

This is where the research becomes a document, and everything in it should help someone sell.

A quote helps if it shows how the person thinks. “I’m tired of wasting hours in status meetings when I could be automating deployments” is worth including. “I love walking the Labrador at the weekend” is not. Both have appeared in decks I’ve inherited.

Step 4: check it with the people who’ll use it

Before anyone designs slides, take the draft to sales and to the customer success team. Ask three things: does this sound like our real buyers, would it help you on a call, what’s missing. If the answer is a shrug, you’ve missed. If someone says “that’s exactly what Sarah at Acme was like”, you’re close.

Step 5: put it to work and keep it alive

A persona in a SharePoint folder called “Q2 Strategy Ideas” isn’t helping anyone. Use it when writing copy (would this land with this person), when prioritising content (what does this buyer need at this stage), and when training sales (this is what they care about, lead with it). Then update it. Market shifts, a product change, a new ICP, all of it means the persona changes too. It’s a working document. The one thing I still don’t have a clean answer to is how many to keep when sales wants one per vertical and marketing wants three in total, we usually land somewhere untidy in the middle.

What sales wants from a persona

Most marketers build personas for other marketers. Cute names, vague goals, a mood board if someone’s feeling brave, and then confusion when the sales team never looks at it. If you want it used on calls, it needs six things.

1. Pain that sounds like a person

“Struggles with digital transformation” is a phrase from a slide. Sales wants “I’m sick of running campaigns across six tools that don’t talk to each other” and “I’ve got a CEO asking about pipeline and I don’t know what’s working”. Real phrasing, lifted from interviews. When a prospect says something that’s already written on the persona, the rep knows how to frame the next ten minutes.

2. Buying triggers

If you want to turn up at the right moment, the persona has to say what starts the search. New Head of Ops hired, probably reviewing tooling. Series B closed, there’s budget and a board asking what it’s for. That sort of thing, written down where a rep can see it.

3. Decision criteria

This is where deals are won and lost. What does this buyer care about most, which features are non-negotiable, what gets you disqualified. If the persona says “must integrate with HubSpot” and your product doesn’t, sales knows not to spend three weeks finding out. If they’re fixated on security or scale or local support, that’s where the pitch goes.

4. Objections and deal killers

Every buyer has a reason to hesitate. Worried about lock-in. Burnt by a vendor who overpromised. Doesn’t want a long implementation. Has a procurement team that takes eight weeks. Surface them in the persona and marketing can deal with them early, with a case study or a pricing page that doesn’t hide the number, and sales preps an answer before the call rather than on it.

5. Who else is in the room

B2B buyers are rarely alone. Sales wants to know whether this person signs off, who else has to, and who’s whispering in their ear. “Needs buy-in from the CFO and the security team” means they’re selling to a group, which means a budget sheet for finance, a compliance checklist for IT, and probably a security questionnaire that takes three weeks to come back.

6. How to talk to them and where to find them

If your buyer reads G2 reviews, goes to the odd webinar and lives on LinkedIn, sending a 48-page whitepaper as the follow-up is a waste of everyone’s afternoon. Tell sales what content this buyer likes, what channels they’re on and what tone works. If they hate buzzwords, say so. If they watch YouTube demos, give sales a 90-second walkthrough to send. If they’re in a Slack community, get marketing in there being useful rather than promotional.

A persona isn’t finished until a salesperson says they’d use it. Ask them, and write down what they’d change.

Two examples

Enough theory. This is roughly what a finished one looks like.

CTO Charlie

“If I have to sit through another ‘scalable, cloud-native’ buzzword bonanza, I’m out.”

• Title: CTO
• Company: Mid-sized SaaS, about 150 employees, Series B
• Decision role: Final decision-maker, consults the DevOps lead and the CFO

Goals

• Cut infrastructure cost without adding downtime
• Faster deployments, more productive developers
• A stack that survives the next stage of growth

Challenges

• Legacy tools that don’t play together
• Developers spending time on manual work
• Feels locked in and wants options

Buying triggers

• Recent downtime that annoyed users
• Growing team, growing pains
• CFO just asked for a cost reduction plan

Decision criteria

• Fits the existing stack (AWS, GitHub and so on)
• Transparent pricing, no lock-in
• Hands-on support and proper documentation

Common objections

• “We’ve already got five tools that do half of this”
• “What’s the switching cost?”
• “I’ve been burned by bad support before”

Preferred content

• Technical docs
• Developer opinion (Reddit, Hacker News, Stack Overflow)
• Product walkthroughs rather than polished decks

Note for the team: Charlie wants to see the thing working. If the pitch is marketing smoke he leaves, politely, and doesn’t come back.

Marketing Lead Mia

“If I don’t show pipeline growth this quarter, I’m toast.”

• Title: Head of Marketing
• Company: B2B SaaS, about 50 employees, scaling fast
• Decision role: Budget holder for marketing tools, consults the CEO

Goals

• More inbound demo requests
• Better lead-to-opportunity conversion
• Attribution and ROI she can put on an exec dashboard

Challenges

• CRM setup is a mess
• Leads slipping through the cracks
• Sales keeps saying the leads are rubbish

Buying triggers

• CEO pushing for results
• SDR team complaining about lead quality
• Ran a big campaign and can’t tell what it did

Decision criteria

• Easy to roll out to a small team
• Integrates with HubSpot and Slack
• Reporting good enough for the board deck

Common objections

• “I don’t have time for a complicated rollout”
• “Looks expensive for what it does”
• “Will sales actually use it?”

Preferred content

• Case studies from companies her size
• ROI calculators
• Short webinars or demo videos

Note for the team: Mia isn’t technical but she’s sharp. Show her how this makes her look good in front of the CEO and she’s listening.

Neither of these is a real person. Both are stitched together from interviews with several real people, which is how all of them get made.

How to lay them out so people use them

You can have the best research in the building and lose it in a 12-tab spreadsheet. Format matters more than anyone likes to admit.

One page each. PDF, slide, Notion card, whatever the team already opens. If it runs onto a second page, cut something.

Blocks with headings. Top: name, title, quote, company size. Middle: goals, challenges, triggers. Bottom: objections, decision criteria, preferred content, stakeholders. Same order on every persona so people learn where to look.

Make it skimmable. A headshot (stock is fine), an icon per section, bold on the lines that matter. A rep will glance at it two minutes before a call, and that’s the use case to design for.

A callout with the takeaway. A sidebar per persona: “Messaging hook: lead with ROI and the HubSpot integration.” “CTA to test: see how fast you can launch.” It gives the reader the implication without making them work it out from the goals list.

The bit that’s changed since I first wrote this

You can now generate a plausible set of personas in about a minute by pasting your website into a chatbot, and plenty of teams do. They look fine, they’re also worth less than nothing, because they carry the authority of research without any of it having happened, and a persona nobody interviewed for is a guess dressed up as evidence. My view is that within a couple of years the word “persona” will be mildly embarrassing in most PMM teams, and what replaces it will be the interview library: tagged, searchable, quoted directly in briefs, with the summary regenerated whenever you add ten more calls. The research step was always the hard part and the tools haven’t changed that. I could be wrong about the timing.

Where to start

If a line in your persona doesn’t help you understand the buyer, speak their language, name their pain or move a deal, bin it. Then pick five customers who bought in the last six months, book 20 minutes with each, and ask what was going on the week they started looking. Write the pain down in their words, not yours. Show it to a salesperson and ask if they’d use it. That’s the first persona. The other two or three come the same way, at about a month each.