Mind Bending · Digital Control · Psychographics · Data · 2016

Cambridge Analytica

A political data firm that harvested 87 million Facebook profiles, built psychographic models of individual voters, and deployed targeted psychological manipulation at unprecedented scale. Bernays's techniques, Goebbels's ambition, and Silicon Valley's data infrastructure — combined for the first time.

Active

2013–2018

Data harvested

87 million Facebook profiles

Campaigns

Brexit · Trump 2016 · 200+ elections globally

Exposed by

Christopher Wylie — whistleblower · 2018

The completion of the Bernays project. Bernays dreamed of targeting individuals with messages precisely calibrated to their specific psychological profile — but his tools were blunt. Goebbels had the ambition but not the data. Cambridge Analytica had both: 87 million psychological profiles, the computational power to model them, and the digital delivery infrastructure to reach each individual with the specific message most likely to move them. What Bernays theorised in 1928 and Goebbels approximated in 1933 became technically precise in 2016.

Section 01

What Was Cambridge Analytica

Cambridge Analytica was a British political consulting firm founded in 2013 as an offshoot of the Strategic Communication Laboratories group, backed primarily by American hedge fund billionaire Robert Mercer and his daughter Rebekah. Its stated offering was unprecedented: the ability to build detailed psychological profiles of individual voters using social media data, and to use those profiles to deliver targeted political messaging calibrated to each individual's specific psychological vulnerabilities and motivations.

The firm's CEO Alexander Nix described its approach with unusual candour in a 2016 presentation: "We have now worked in over 200 elections across the world... We have profiled the personality of every adult in the United States of America — 220 million people." This was not primarily a technology company. It was a psychological operations firm that used technology as its delivery mechanism — a direct descendant of the wartime propaganda operations that Bernays and Goebbels had developed, updated for the digital age.

The company collapsed in 2018 following the revelations of whistleblower Christopher Wylie and investigative reporting by The Guardian and Channel 4 News. But the techniques it had pioneered and the data infrastructure it had demonstrated was possible did not disappear with the company. They became the template for political data operations globally.

Section 02

The OCEAN Model

Cambridge Analytica's psychological profiling was built on the OCEAN model — the "Big Five" personality framework that is the most widely validated model in academic personality psychology. OCEAN measures five dimensions of personality: Openness (curiosity, creativity, intellectual interest), Conscientiousness (organisation, reliability, self-discipline), Extraversion (sociability, assertiveness, positive emotion), Agreeableness (cooperation, trust, empathy), and Neuroticism (emotional instability, anxiety, negativity).

The key insight that Cambridge Analytica operationalised — developed by Cambridge University researchers Michal Kosinski and David Stillwell — was that Facebook likes alone could predict OCEAN scores with remarkable accuracy. Kosinski's research showed that 68 Facebook likes were sufficient to predict a person's skin colour with 95% accuracy, their sexual orientation with 88% accuracy, and their political affiliation with 85% accuracy. 150 likes predicted personality better than a coworker. 300 likes predicted it better than a spouse.

This was the theoretical foundation of the entire operation: digital behaviour leaves a precise psychological fingerprint, that fingerprint can be read at scale, and reading it at scale enables the construction of psychological profiles of every individual in a target population — without their knowledge or consent.

OCEAN dimension

High Neuroticism — Fear Messaging

Individuals scoring high on Neuroticism are more susceptible to fear-based messaging — they process threat information more intensely and are more likely to be motivated by the desire to avoid negative outcomes than to achieve positive ones. Cambridge Analytica's targeting algorithms identified high-Neuroticism individuals and delivered messaging emphasising threat, danger, and the necessity of protective action. The message was not designed for the issue — it was designed for the psychological profile of its recipient.

OCEAN dimension

Low Openness — Authority Messaging

Individuals scoring low on Openness tend to prefer familiar, conventional, and authoritative information sources — they are more susceptible to messaging from established authority figures and more resistant to novel or unconventional ideas. Targeting low-Openness individuals with messaging from trusted authority figures and framing political choices as the protection of established tradition against threatening change was a specific psychographic strategy documented in Cambridge Analytica's internal materials.

The Precision

5 Profiles — Infinite Variations

Cambridge Analytica's approach was not to create five different messages for five personality types — it was to create a continuous spectrum of variations calibrated to the specific combination of OCEAN scores of each individual recipient. A single political advertisement might exist in hundreds of micro-variants, each tuned to a specific psychological profile. The voter saw a single ad. The system had chosen it from hundreds of alternatives based on their psychological fingerprint.

Section 03

The Facebook Harvest

The data that made Cambridge Analytica's psychographic profiling possible came primarily from Facebook — through an operation that exploited a loophole in Facebook's API that allowed third-party app developers to collect data not only from app users but from all of their Facebook friends. Dr Aleksandr Kogan, a Cambridge University psychologist, built a personality quiz app called "thisisyourdigitallife" that 270,000 Facebook users completed. Through the friend-network loophole, Kogan's app collected data from approximately 87 million Facebook profiles — most of whom had never heard of the app and had never consented to share their data.

This data was then sold to Cambridge Analytica — in violation of both Facebook's terms of service and, depending on jurisdiction, data protection law. Facebook knew about the data transfer in 2015 and asked for the data to be deleted. Cambridge Analytica certified that it had been. It had not. The data remained in use through the 2016 election campaign and beyond.

Facebook's fine for its role in the scandal — $5 billion from the US Federal Trade Commission in 2019 — was the largest privacy fine in history at the time. It represented approximately one month of Facebook's revenue. No individual Facebook executive faced criminal charges. Cambridge Analytica's CEO Alexander Nix was banned from serving as a company director in the UK. No criminal prosecution followed.

Section 04

Targeting in Practice

Cambridge Analytica's internal documents — leaked by whistleblower Christopher Wylie and published by The Guardian — reveal the specific practical application of psychographic targeting in political campaigns. The approach had three phases: identify persuadable voters by psychographic profile, determine which specific fears, desires, or values were most salient for each profile, and deliver targeted content designed to activate those specific psychological levers.

Suppression Targeting

Demobilise rather than persuade

One of Cambridge Analytica's documented strategies was not persuasion but suppression — identifying likely opposition voters by psychographic profile and targeting them with demoralising content designed to reduce their motivation to vote. Content emphasising corruption, futility, and the irrelevance of individual political action was delivered specifically to identified opposition-leaning profiles. The goal was not to change their vote but to ensure they did not cast one.

Dark Posts

Invisible targeted content

"Dark posts" — Facebook advertisements visible only to their specifically targeted recipients and not to the general public — were the primary delivery mechanism for psychographically targeted content. Unlike public political advertising, dark posts could not be monitored by journalists, regulators, or opposing campaigns. A candidate could simultaneously deliver contradictory messages to different psychographic segments — telling one group one thing and another group the opposite — with no public record of the contradiction.

Fear Architecture

Threat · Vulnerability · Solution

The basic structure of Cambridge Analytica's most effective content followed a consistent psychological architecture: identify the specific fear most salient for the target's psychographic profile, present content that activates and amplifies that fear, and position the campaign's candidate or message as the solution. High-Neuroticism profiles received immigration threat messaging. Low-Agreeableness profiles received messaging about crime and social breakdown. The fear was real. The solution was predetermined. The targeting ensured the right fear reached the right person.

Section 05

Brexit & 2016

Cambridge Analytica's two most consequential deployments — the Brexit referendum and the 2016 US Presidential election — both produced results that defied pre-election polling and that remain among the most politically significant events of the early twenty-first century. Whether Cambridge Analytica's specific intervention was decisive in either outcome is genuinely contested among researchers. What is not contested: it was deployed, it was unprecedented in its psychological precision, and both campaigns won by margins narrow enough that any systematic advantage could have been determinative.

Brexit · 2016

Leave.EU & AggregateIQ

Cambridge Analytica's parent company SCL and its Canadian data subsidiary AggregateIQ worked with multiple Brexit campaigns. The Leave.EU campaign — backed by Nigel Farage and funded by Arron Banks — worked directly with Cambridge Analytica on voter data and messaging strategy. The specific psychological targeting of English working-class voters with immigration fear messaging in the final weeks of the campaign has been documented, though the full extent of data use remains subject to ongoing investigation by the UK Information Commissioner's Office.

Trump 2016

The Make America Great Again Data Operation

Cambridge Analytica was hired by the Trump campaign in June 2016, funded primarily by Robert Mercer. Steve Bannon — later Trump's chief strategist — was a Cambridge Analytica board member. The firm's psychographic targeting was deployed specifically in key swing states — Pennsylvania, Michigan, Wisconsin — where the final margins were measured in tens of thousands of votes. Trump's data operation, described by his own team as "Project Alamo," used Cambridge Analytica's profiles to deliver targeted digital content to approximately 13.5 million identified persuadable voters.

The Contested Question

Did It Work?

Whether psychographic microtargeting actually changed enough votes to determine the outcomes of Brexit and the 2016 election is genuinely uncertain. Academic researchers are divided: some argue the effect sizes documented in experimental settings are too small to have been decisive at scale; others argue that in elections decided by margins of 1-2%, even small systematic effects are determinative. What is certain: the industry believed it worked, invested accordingly, and the techniques have been deployed in every major election since.

Section 06

The Legacy

Cambridge Analytica was dissolved in May 2018 — but the operation it represented did not end with the company. Its staff dispersed into other political data firms. Its techniques became the template for political data operations globally. Its demonstration that psychographic microtargeting at scale was technically possible — and that the regulatory framework to prevent it did not exist — was absorbed by governments, campaigns, and commercial interests worldwide.

The deeper legacy is epistemological: Cambridge Analytica demonstrated that in a sufficiently data-rich environment, the gap between individual psychology and delivered message can be closed to near-zero. Every person can be shown the specific content most likely to activate their specific psychological vulnerabilities — at the speed of digital delivery, at the scale of entire electoral populations, without their knowledge, without their consent, and without any public record of what was shown to whom.

Bernays needed a mass medium and a generalised understanding of crowd psychology. Goebbels needed total control of national media infrastructure. Cambridge Analytica needed a smartphone and a Facebook account. The barrier to entry for psychological mass manipulation has not been eliminated. It has been reduced to the price of a digital advertising campaign.

"We exploited Facebook to harvest millions of people's profiles. And built models to exploit what we knew about them and target their inner demons."

Christopher Wylie — Cambridge Analytica whistleblower, 2018