AI can simulate buyers, but it misses personality

Synthetic audiences promise marketers faster customer research by simulating buyer responses using AI. But many models share the same blind spot: they rely on demographic, firmographic, and behavioral data that describe buyers, but not how they make decisions.

I kept running into that gap while working on my book, “The Hidden Buyer Journey.” I studied the personalities of 10,000 buyers across 15 industries over seven years and found consistent personality patterns by role and industry.

Buyer personalities cluster by role and industry

One finding has important implications for synthetic audience accuracy: buyer personalities aren’t randomly distributed. They cluster around role and industry.

We call this the Two-Thirds Rule: within a given role and industry, two personality types account for at least two-thirds of the people who hold that job. Data scientists in life sciences skew heavily toward one personality profile, while operations professionals in the airline industry skew toward another.

Most synthetic personas are built from data that is relatively easy to collect, such as firmographics, job history, public behavior, and survey responses. That data tells you what category someone falls into. It says far less about how they weigh risk, process decisions, or respond to messaging.

Without personality as an input, a synthetic persona captures only part of what shapes buyer behavior. Incorporating the Two-Thirds Rule grounds the model in the personality profile most likely for that role and industry.

Why professional signals miss personality

The Two-Thirds Rule helps explain why personality matters. There’s another reason many synthetic personas miss how buyers make decisions: they’re trained primarily on professional signals.

When someone takes a job, they put on what we call a work persona — a professional identity that may or may not reflect who they actually are. What doesn’t change is the behavior tied to their underlying personality.

A synthetic persona trained on signals such as LinkedIn activity, published content, or meeting behavior learns the professional persona rather than the underlying decision-making patterns that influence how someone buys.

Personality data complements demographic, firmographic, and behavioral data by helping models account for those underlying patterns. Without it, synthetic audiences simulate only part of what drives buying decisions.

Recommended Articles

Why LinkedIn doesn’t really care about AI slop

Aug 17, 2026

The latest AI-powered martech news and releases

Aug 13, 2026

The marketing-to-sales handoff is where revenue leaks

Aug 10, 2026

How to build more accurate synthetic audiences

If personality is missing from today’s synthetic audiences, improving them isn’t about collecting more of the same data. It’s about adding a different kind of input. Our research points to three practical changes.

• Layer personality into persona generation: Don’t rely solely on demographic and firmographic data. If two-thirds of people in a role and industry share a personality type, that distribution should be another input alongside job title and company size.
• Model personas at the role-and-industry level: Personality concentrations emerge from the combination of role and industry, not either variable alone.
• Validate against personality, not just demographics: A synthetic audience can match firmographics yet still misjudge how buyers respond to messaging because risk tolerance, decision-making, and communication preferences are influenced by personality.

The missing ingredient in synthetic audiences

Synthetic audiences are here to stay, but they’re only as good as the data they’re trained on. Today, most models excel at describing buyers’ roles, behaviors, and organizations. They have far less visibility into the personality patterns that shape buying decisions.

Marketing organizations that incorporate personality alongside demographic, firmographic, and behavioral data will build synthetic audiences that better reflect how real buyers make decisions — not just audiences that look plausible.