Most B2B personas are decoration. A slide with a stock photo, an alliterative name like "Marketing Mary," an age, a few bullet points about goals and frustrations. The team builds them in a workshop, everyone nods, and the deck goes into a drawer. Nobody opens it again, because there is nothing in it you can actually use.
The persona was never the problem. The vagueness was. "Marketing Mary, 34, values efficiency" cannot tell you which competitor wins a deal, which objection kills it, or what content would change the outcome. It is too thin to drive a decision.
That thinness has a cost, and it compounds. A vague persona corrupts everything downstream: the buying journeys you run to test your AI search performance, the content you produce to win buyers, the positioning you build your whole funnel around. Garbage persona in, garbage strategy out.
A note on the examples in this post: the personas shown come from a demo project we built on Oxide, a real infrastructure company, to illustrate the feature against a realistic B2B buyer set. Oxide is not an Ansehn customer.
A persona is a model of a decision, not a portrait of a person
The old persona answered "who is this person?" The useful question is "how does this person decide?"
That shift changes what goes into a persona. Demographics barely matter. What matters is the buyer's pain points, their decision criteria, the order they weigh those criteria in, their tolerance for risk and complexity, and the actual questions they would type into an AI when researching a purchase. A persona built this way is not a portrait. It is a model you can run.
Build a persona, or let Ansehn build one for you
Personas in Ansehn are built across steps that force depth: buying behavior, narrative and pain, and the queries the buyer would actually bring to an AI.
Ansehn generates each persona from your brand, market, and competitive context, and you can edit any field or add custom personas as your buyer mix shifts. The build flow itself is the discipline. Every persona has to carry pain points, decision criteria, a risk and complexity profile, and the search intent that drives buying questions. There is no shortcut to a thin profile.
Inside a single persona
Open any persona and you get the full decision model.
Each persona carries pain points, decision criteria, a risk and complexity profile, and the search intent that drives their buying questions.
Take the CIO persona. The description is specific: an executive leading technology strategy at a multinational financial institution, navigating risk, regulatory mandates, and cost control while supporting mission-critical low-latency workloads. The pain points are concrete: fragmented infrastructure that requires costly integrations, escalating cloud costs, strict data residency and audit requirements, long procurement cycles. The search intent is sharp: proven strategies to replace legacy tech with a secure, cloud-native, on-prem alternative that meets compliance and total cost mandates.
None of that is decoration. Each pain point is a content opportunity. Each criterion is something you can win or lose on. The search intent tells you the exact framing the buyer uses when they go to an AI, which is the framing your content needs to match.
See the whole buying committee at once
A single persona is a model of one buyer. A B2B deal is rarely decided by one buyer. The Persona Decision Map positions every persona in your set by problem urgency and buying power, so you can see your full buying committee on one view.
The Persona Decision Map positions every persona by problem urgency and buying power, so you can see at a glance which buyers are worth winning first.
The CIO of a Fortune 100 financial firm sits in the top-right: high urgency, high buying power. That is the buyer to win first. A persona low on both axes might not be worth building content for at all. The map turns a pile of profiles into a prioritization.
The same product, different buyers, different verdicts
Here is where vague personas do the most damage. A single generic persona flattens your entire market into one imagined buyer. But the buyers are not the same, and they do not decide the same way.
The CIO of a Fortune 100 financial firm walks through the decision weighing compliance, data residency, and total cost at scale. The VP of Engineering at a growth-stage SaaS company walks through the same decision weighing unpredictable cloud bills, vendor lock-in, and migration complexity. Same product. Different criteria. Different verdict.
Run a buying journey with the generic "B2B IT buyer" and you get a generic answer that maps to neither of them. Run it with the CIO persona and the recommendation turns on compliance and cost-at-scale. Run it with the VP of Engineering and the recommendation turns on cloud-cost relief and lock-in risk. The product never changed, but the criteria that decide the deal did.
The same buying question, two personas. The vague persona returns a generic result; the sharp persona returns a specific, actionable verdict with a different win rate. The persona is the variable that decides what you learn.
If your persona is vague, you never see these differences, and you optimize for a buyer who does not exist.
From persona to monitored intent
A persona in Ansehn is not a static profile. Each one generates the high-intent search queries that buyer would actually bring to an AI, and you can put those queries under monitoring or schedule them as recurring buying journeys.
Each persona surfaces the buying questions that buyer would ask an AI, ready to monitor or run as scheduled buying journeys.
The CIO persona surfaces questions like "How can I reduce cloud costs without losing developer agility?" and "What is the fastest path to migrate finance workloads off VMware?" Those are not keywords. They are the actual decision-stage questions a real buyer asks. The CIO persona's scheduled buying journey on those questions currently shows a 29% win rate against named competitors. That number is the link between persona quality and pipeline. Improve the persona, improve what you learn, improve the win rate. The persona is the bridge between knowing your buyer and measuring whether you win them.
FAQ
How are personas built in Ansehn? Ansehn generates personas from your brand details, market, and competitive context, then enriches each with pain points, decision criteria, a risk and complexity profile, and search intent. You can also add fully custom personas, and regenerate the set as your market understanding evolves.
How many personas should I have? Enough to cover the distinct ways your buyers decide, not one per job title. The Oxide demo uses seven because the product is evaluated very differently by a CIO, an engineer, a security director, and a procurement lead. The Persona Decision Map helps you see whether your set covers the buyers worth winning.
What makes a persona good enough to be useful? It has to model a decision, not describe a person. If a persona cannot tell you what the buyer weighs, what they fear, and what they would ask an AI, it is too vague to drive content or buying journeys. Specificity in the pain points and search intent is what makes a persona functional.
How do personas connect to buying simulations? A persona is the primary input to a buying journey. The journey runs the buyer's decision using that persona's criteria and search intent, so the verdict is only as realistic as the persona behind it. This is why persona quality determines simulation quality. See Buying Simulations in Ansehn for the full walkthrough.
Build a persona that models a real decision
Start with the buyer you most want to win. Give them real pain points, real decision criteria, and the questions they would actually ask an AI. Then run a buying journey and see whether you win them.