Follow the data from an app to a public investigation.
Compare the FTC's account with the roles of the platform, application developers and consulting firm, keeping track of which source describes each part.
Some questions can be answered directly from a document. Others require comparing accounts or following a lead to another source. A useful research trail preserves both the findings and the questions that remain open.
How do several documents become a research trail rather than a pile of links?
Research is the discipline of keeping provenance while changing your mind. This journey uses Cambridge Analytica to build a trail that records what each document supports and what remains open.
See the subject in Counterpedia Cambridge Analytica · research Entry · Research · not admittedOr begin with step 1 ↓A reading exercise
Follow two accounts without blending them
Compare the research account with the FTC documents reached from its release. Which document supports the statement you want to make?
Cambridge Analytica research entry
The research entry brings together accounts of the platform, app/developer and consulting firm. It is a research starting point, not a final finding.
Cambridge Analytica · research Entry Research · not admittedFTC release and linked documents
This link opens the FTC release. Follow it to the opinion for a finding or the final order for an operative requirement, then keep the supporting passage and its procedural scope with your statement.
FTC · Press release on the Cambridge Analytica opinion and order (Dec 2019) Original publisher materialSee the distinction
A useful trail keeps each voice attached to its document and leaves unanswered questions visible.
Carry this forwardWrite one supported point and one open question, each with its source.
Follow the trail
- 01
Open the research dossier
State the question first: which document establishes each step of the described data collection and enforcement sequence, and what remains uncertain? The dossier frames that question and lists leads. Treat it as a starting point, not a summary of findings; a dossier or candidate relationship is not automatically an admitted or independently verified source.
Open Cambridge Analytica · research dossier Research · not admitted
Keep: One research question and the dossier's leads, labeled research-only.
- 02
Read the FTC action
Start at the FTC release, then follow its links to the document supporting your chosen statement: the opinion for a finding, or the final order for an operative requirement. Record the document, date, exact supporting passage and procedural scope. The FTC's 2019 proceedings examined Cambridge Analytica's collection and use of Facebook-user information through an app and its access to friends' data, including alleged deceptive practices. Attribute the statement to the FTC and the document you actually inspected; the release announces the action and is not a substitute for the opinion or order.
Open FTC · Press release on the Cambridge Analytica opinion and order (Dec 2019) Original publisher material
Keep: An FTC-attributed statement tied to an exact opinion or order passage if inspected.
- 03
Compare accounts without merging them
Set the FTC account beside other material. The developer, the platform, the consulting company, the regulator and affected users occupy different roles, and each document speaks from one of them. Record who said it, in what capacity, and where the accounts differ, rather than averaging them into one consolidated narrative.
Open FTC · Press release on the Cambridge Analytica opinion and order (Dec 2019) Original publisher material
Open Cambridge Analytica · research dossier Research · not admitted
Keep: A source-by-source record of positions and disagreements, not a merged narrative.
- 04
Keep a findings column and an open column
Write what the documents establish, each with its reference, and separately what they do not. Keep evidence, attributed positions, interpretation and derivation apart as you go; in a growing workspace they otherwise become indistinguishable. Record disagreements and derivations explicitly. The open column is the valuable one.
Open Cambridge Analytica · research dossier Research · not admitted
Keep: Two columns: supported findings with references, and unresolved questions.
- 05
Finish with a reusable trail
Retain the identities of the documents, the attributable positions, the derivations and the open questions so a later reader can review or reproduce the inquiry. Leave with identifiable references, one supported finding and the questions that remain genuinely unanswered. Dated anchors, such as the 2018 bankruptcy of Cambridge Analytica and the 2019 FTC opinion and final order, orient the trail; each still needs its own source and procedural scope.
Open Cambridge Analytica · research dossier Research · not admitted
Keep: A reusable trail with document identities, one supported finding, and open questions.
Open the material
- Cambridge Analytica · research EntryResearch · not admitted
- Cambridge Analytica · research dossierResearch · not admitted
- FTC · Press release on the Cambridge Analytica opinion and order (Dec 2019)Original publisher material
Research material is not admitted evidence; publisher links identify someone else’s document.
What to notice
- A dossier is research material, not admitted evidence.
- Agreement between accounts is not independence.
- An open question is a result, not a failure.
- Evidence, attributed positions, interpretation and open questions are different things and belong in different places.
A useful finish
A short research trail that separates evidence, attributed positions, interpretation and open questions: identifiable document references with the role of each issuer, one supported finding and explicit open questions, ready for a later reader to reuse.
Worked example
Observed on 4 November 2026 on the public site, signed out. This is what one actual run of these steps produced.
A labelled research starting point: Counterpedia's Cambridge Analytica research entry and dossier, with the dossier's source inventory and its recorded contradictions, such as “C-1 — 270,000 vs 50–65 million vs 87 million”, which it says “are not interchangeable counts.”
Positions kept apart by the issuer the dossier records: “FTC / DOJ settlement” (the Facebook $5 billion privacy settlement), Facebook, the UK Information Commissioner's Office, the House of Commons Digital, Culture, Media and Sport Committee and the U.S. Federal Election Commission. The FTC's Cambridge Analytica Opinion and Order are listed under “Issuer not labelled in the dossier”.
Open questions carried over from the dossier, including whether Cambridge Analytica's work measurably changed an election outcome.
Finish: not reached. No FTC document was read, so the trail had no finding tied to an inspected passage.
What it rests on
- Counterpedia's Cambridge Analytica research entry, marked “Research entry · Not admitted”, and its research dossier.
- The dossier's card for the FTC's final Cambridge Analytica Opinion and Order: FTC matter 182 3107, docket 9383. Its “Opinion date: December 6, 2019” is the FTC case-timeline date, not the date the documents were issued (see the later check).
What it did not establish
- ftc.gov refused this run's automated requests, including the press-release link on this page, so no FTC opinion, order or complaint was read.
- The FTC's position here is as the dossier describes it, not as read in an FTC document.
- No save or reopen operation was offered on this signed-out path, so the trail was kept outside Counterpedia.
Later check, 5 November 2026: A reviewer, using a research retrieval service rather than this run's client, opened the FTC's Commission Opinion and Final Order in the Cambridge Analytica matter. Both state that they were issued on 25 November 2019 (Opinion, printed page 16; Final Order, printed page 5). The FTC case timeline lists them, with the announcement, under 6 December 2019. That is the timeline and announcement date; it does not establish when the documents first became public.
Later check, 5 November 2026: In the same reading, the Opinion (section III.A, printed pages 11–12) finds the app's representation about collecting identifiable information deceptive, including because the app collected Facebook User IDs. The decision followed Cambridge Analytica's failure to answer or oppose summary decision, and that procedural context stays with the finding. The Final Order (section IV.B, printed page 4) requires deletion of covered information and work derived from it, with timing provisions and exceptions for government custody and preservation obligations.
Later check, 5 November 2026: What the reviewer's reading does not establish: the Order imposes a duty to delete but does not show that deletion took place, and neither document establishes any effect on an election outcome. These are the reviewer's readings, not part of the 4 November run, which still did not reach its finish.
If it stops
- This journey does not establish facts beyond the documents it cites.
- Candidate relationships stay candidates.
- Cambridge Analytica is an illustrative subject; the Guide does not offer one consolidated narrative of the actors involved.
- A question the documents leave unresolved is recorded as unresolved, not settled by inference.
Go deeper
Before you go deeper: The first chapter, Using the research library, stays with Cambridge Analytica but begins from a research pack and asks for the pack's author, document identities and status. If Counterpedia lists no relevant pack, the chapter's fallback is the FTC case record as an external starting point, not an opened pack.