# Psychographic Profiling The practice of constructing individual-level inferred-personality profiles — typically along the *Big Five* / *OCEAN* personality dimensions (Openness, Conscientiousness, Extraversion, Agreeableness, Neuroticism) — from behavioral data, for the purpose of delivering personality-tailored persuasive content. The behavioral data inputs commonly include social-media likes, browsing patterns, purchasing history, survey responses, location traces, app-usage patterns, and (in the most ambitious versions) cross-platform integrated data. The practice's public visibility crystallized through the Cambridge Analytica scandal of 2014–2018; the underlying behavioral-research lineage long-predates that specific firm and long-survives it. ## The Big Five / OCEAN psychological foundation The contemporary version of the practice rests on the *Big Five* / *Five-Factor Model* of personality — an empirical-statistical framework developed across decades of psychometric research (Allport-Odbert 1936 lexical hypothesis → Cattell 16PF → Tupes-Christal 1961 → Goldberg 1981 → Costa & McCrae NEO-PI-R 1985+). The framework identifies five broad personality dimensions that explain substantial variance in human behavior: 1. **Openness** to experience — imaginative, curious, intellectually-engaged vs. conventional, traditional, narrowly-focused 2. **Conscientiousness** — organized, dependable, disciplined vs. spontaneous, careless, impulsive 3. **Extraversion** — outgoing, energetic, sociable vs. reserved, solitary, reflective 4. **Agreeableness** — cooperative, trusting, helpful vs. competitive, skeptical, challenging 5. **Neuroticism** — emotionally-reactive, anxiety-prone, mood-variable vs. emotionally-stable, calm, even The Big Five is one of the most-replicated frameworks in psychological research; cross-cultural validity is robust though not universal; the framework has substantial predictive validity for various life outcomes (occupational performance, relationship satisfaction, health behaviors, political attitudes). ## The Kosinski-Stillwell-Graepel inflection The contemporary digital-era psychographic-profiling project crystallized around the work of Michal Kosinski, David Stillwell, and Thore Graepel at the Cambridge Psychometrics Centre. Their 2013 PNAS paper *Private traits and attributes are predictable from digital records of human behavior* demonstrated that Big Five personality scores — plus a range of other traits including political views, sexual orientation, ethnicity, religious affiliation, and substance use — could be inferred from Facebook *likes* data with measurable predictive accuracy. The methodological insight: aggregate-level patterns in *like* data (which pages a user has liked) provide signal-rich behavioral fingerprints that correlate predictably with personality dimensions. A subject who likes *philosophy*, *jazz*, *political-podcasts*, *art-museums* may score predictably high on Openness; a subject who likes *Harley-Davidson*, *country music*, *NRA*, *military* may score predictably high on Conscientiousness and low on Openness. The pattern-aggregation approach allows personality inference from observable-behavior data alone, without requiring subjects to take personality surveys. The 2013 PNAS paper is the academic-research inflection point. The subsequent commercialization of the method is the inflection point for the public-controversy phase. ## The Cambridge Analytica deployment (2014–2018) [[Cambridge Analytica]] — the SCL Group political-affiliate, funded by [[Robert Mercer]], run by Alexander Nix — built its political-targeting business on a psychographic-profiling framework explicitly modeled on the Kosinski-Stillwell-Graepel research. The firm's technical contributions included: 1. **Acquiring the data** — through the *thisisyourdigitallife* personality-quiz Facebook app developed by [[Aleksandr Kogan]] / Global Science Research, which harvested both the data of the ~270,000 quiz-takers and the data of their Facebook friends (then ~87 million people total) under the permissive Facebook Graph API v1 in place at the time 2. **Building personality models** — fitting OCEAN scores to the harvested data using the Kosinski-style methodology 3. **Segmentation by personality-and-political-attitude clusters** — partitioning the target electorate into segments defined by both demographic-political variables and inferred personality dimensions 4. **Personality-tailored content delivery** — different ads, different framings, different emotional appeals delivered to different personality segments, with the claim that the personality-matched content produced measurably higher persuasion-rates than untailored content The firm contracted with the 2016 [[Ted Cruz]] primary campaign, the 2016 Trump general-election campaign (after Cruz dropped out), the 2016 Brexit Vote Leave campaign, and multiple international clients including operations in Kenya, Trinidad, Nigeria, the Czech Republic, and Mexico. The 2018 [[Christopher Wylie]] whistleblower disclosures — published in March 2018 in the *Observer*, *Guardian*, and *New York Times* — exposed the operation in detail. Channel 4's hidden-camera footage of Alexander Nix discussing entrapment-and-honeypot political tactics provided the visual-public-face of the scandal. Cambridge Analytica filed for bankruptcy and dissolved in May 2018; Facebook faced multi-year regulatory investigation and the $5 billion FTC penalty in 2019. ## The empirical-effectiveness controversy A complication in the analytical literature: Cambridge Analytica's *marketing claims* about psychographic-profiling effectiveness substantially exceeded what subsequent independent researchers found. Specifically: - **The firm's claim**: personality-matched content produced large measurable persuasion gains over conventional targeting - **Subsequent independent research**: the marginal effect of personality-matched content over demographic-and-attitudinal targeting is *measurable but modest*; the firm's marketing claims were inflated The Kosinski-Stillwell-Graepel research and subsequent academic literature support the *predictive-validity* claim (Big Five scores can be inferred from behavioral data with measurable accuracy) but find smaller *persuasion-effectiveness* effects from personality-tailored content than the Cambridge Analytica pitch suggested. The honest analytical position: psychographic-profiling-based targeting is a real capability that produces real effects, but not at the dramatic-impact scale the firm's marketing claimed. The *structural* concern is independent of any particular firm's effectiveness: a *market* for psychographic-targeting services using personal-data inputs exists at scale; the regulatory regime governing the market is still uneven; the long-run trajectory of the capability is upward as both the data-inputs and the AI-inference layer continue to improve. ## The contemporary state The post-2018 institutional environment: - **GDPR / EU regulation** — substantial constraints on profile-building from personal data; political-targeting subject to additional restrictions under the Digital Services Act and 2024 EU Political Advertising Regulation - **Facebook / Meta policy changes** — substantial narrowing of available targeting categories for political advertising; partial throttling of the data-export infrastructure that enabled Cambridge Analytica - **Continued commercial operations** — many firms in the space continue to operate under different names with different specific methods; Cambridge Analytica's successor firms (Emerdata, Auspex International) operate in adjacent markets; commercial psychographic-targeting infrastructure exists across the political and commercial advertising sectors - **AI-era escalation** — large-language-model technology has dramatically lowered the cost of *personality-aware content generation*; the personality-inference and content-tailoring loops are now near-real-time - **Reduced research-side visibility** — the post-2022 weakening of academic disinformation-research infrastructure has narrowed the independent monitoring capacity for psychographic-targeting operations The 2026 midterm cycle will substantially test the post-CA equilibrium; whether psychographic targeting in the LLM era is being meaningfully constrained or is operating at substantially greater scale than the 2016 cycle is an open empirical question. ## See Also - [[Microtargeting]] — the broader political-advertising-decomposition category that psychographic profiling enables - [[Cambridge Analytica Scandal]] — the canonical public-controversy case - [[Michal Kosinski]] — the principal academic researcher - [[Christopher Wylie]] — the whistleblower - [[Surveillance Capitalism]] — the structural-economic frame - [[Disinformation]] — the broader categorical context - [[Disinformation Hub]] — Subject hub - [[Disinformation Lexicon]] — alphabetical MOC