OSINT Investigation
Open-source intelligence on the Wikipedia activity around your brand.
When a Wikipedia page about you, your competitors, or your category needs scrutiny — we apply OSINT techniques to map who's editing, what sources they're pushing, and what patterns matter.
What you get
OSINT Investigation
We use open-source intelligence techniques to investigate editor activity, coordinated edit patterns, and disputed sources on Wikipedia pages tied to your brand or competitors. Output is an evidence-led report — useful for legal teams, communications leads, and decision support before sensitive edits.
- Edit-history & editor-cluster analysis
- Source quality & independence review
- Coordinated-editing pattern detection
- Confidential reporting only — no public statements
Starting price
From €700 · three fixed tiers
Typical timeline
1-3 weeks
Best for
- · Pages with suspicious edit history
- · Pre-publication intel for high-profile subjects
- · Legal & comms teams preparing disputes
When teams hire us
Four typical engagements
Suspicious edit history
Repeated negative edits, anonymous editors who only edit pages tied to your brand, suspiciously sourced claims that won't go away.
Pre-publication intel
Before publishing a high-profile page, know who watches related categories and which sources will get challenged on review.
Pre-dispute preparation
Legal or PR team is preparing a complaint to Wikipedia about defamatory content. An evidence-led report supports the case.
Competitor monitoring
Track edits to competitor pages, edits about your category, and emerging narratives — without inserting yourself into any edit war.
What you receive
Evidence-led, decision-ready
- Edit-history & editor-cluster analysis
- Source quality and independence review
- Coordinated-editing pattern detection
- Timeline of significant changes
- Sanitised, evidence-led report for legal / comms / leadership
Our reports are confidential and intended for internal use. We don't make public statements, file edit-warring complaints, or interact with Wikipedia editors on behalf of clients — that work, if needed, is a separate engagement with a different policy frame.
Scope of an investigation
Six angles that produce findings
An edit history is public but not readable — thousands of revisions with no narrative. The work is turning that log into a story with names, sequence and motive where the evidence supports one.
Editor history and behaviour patterns
Who edited the page, when, and what else they edit. Editing rhythm, topic range and account age separate a passing volunteer from a single-purpose account.
Signs of coordination
Accounts that appear together, edit in sequence, restore each other's changes, or share phrasing. Coordination leaves a pattern in public logs even when identities do not.
Administrator involvement
Which admins acted on the page, what they protected or reverted, and whether any of them carry a declared or documented conflict on the subject.
Conflicts of interest and undisclosed paid editing
Edits that track a competitor's messaging, or accounts whose history reads like agency work without the disclosure Wikipedia requires.
Cross-language comparison
The same subject often reads differently in EN, DE, FR or UA. Selective translation — an award present in one edition and missing in another — is a deliberate act, and it shows.
Timeline reconstruction
Every material change plotted against external events: funding rounds, litigation, leadership changes. Edits rarely arrive at random moments.
What turns up
The patterns this work exists to find
Categories of finding, not client cases — investigations run under confidentiality and the edit histories are public, so naming a case would name the client.
- A cluster of anonymous accounts editing one page in a way that consistently favours a single competitor.
- An administrator with a documented paid-editing history acting on a page where they hold a stake.
- Divergence between language editions — recognition present in one, quietly absent in another.
- A multilingual pattern of softening negative history across smaller editions, where fewer reviewers watch.
- A single-purpose account adding unsourced negative claims in the weeks before a funding round or a listing.
Method and limits
What we will not do to get an answer
This is intelligence work on a public platform, which makes the boundary worth stating before you ask.
- We work only from publicly available data: revision histories, user contributions, talk pages, noticeboard archives, public logs.
- We do not attempt to de-anonymise editors through hacking, IP tracing, or any access Wikipedia does not publish.
- We do not edit, revert or touch the page during an investigation. The report goes to you; what happens next is your decision.
- Findings are reported as evidence and probability, not as accusations — behavioural patterns rarely prove intent, and we say which is which.
- Reporting is confidential by default; we sign an NDA where the situation calls for it.
Pricing
Three tiers, scoped by how many editions get the full treatment
An investigation is bounded work with a fixed deliverable, so it is priced up front rather than quoted after a call. What scales is coverage — one edition or five — not the hours we bill.
Starter
€700one-time
One language edition. Editor profiles, conflict-of-interest check, coordinated-editing screen and a written report with the evidence behind each finding.
Advanced
Most common€1,000one-time
Up to three language editions, plus a reconstructed timeline of material changes, sockpuppet-pattern analysis and flagged critical risks with what each one exposes you to.
Enterprise
€2,200one-time
Full audit across up to five editions, administrator-action analysis, NDA as standard, white-label reporting for legal or board circulation, and post-report alerting on the pages examined.
What every tier includes
- A written report you can hand to legal, comms or the board without editing it first
- Every claim tied to a public revision, diff or log entry — checkable by anyone
- Findings separated into evidence, pattern and inference, so nothing reads as more certain than it is
- A recommendation on what to do next, including the option of doing nothing
What sits outside the fee
- Acting on the findings — editing, dispute resolution or escalation is separate work, quoted after you decide
- Ongoing monitoring of the pages examined — that is the WikiMonitoring service
- Off-wiki OSINT beyond Wikipedia itself, which is scoped as an add-on on any tier
- Legal advice. We supply the evidence pack; your counsel decides what to do with it
Prices exclude VAT. Turnaround is typically one to three weeks depending on tier and how much history the page carries. An NDA is standard on Enterprise and available on request at any tier.
Frequently asked questions
Answers to the questions that come up before signing
OSINT investigation — common questions
What exactly does a Wikipedia OSINT investigation cover?
Is the investigation confidential?
How long does an OSINT investigation take?
Can you identify who is editing a Wikipedia page?
What is the starting price for an OSINT investigation?
Where to next
Pages readers open after this one
AI Reputation Stack
Turn findings into a defense program.
Wikimonitoring
Catch the next coordinated edit within minutes.
Who edits my Wikipedia page?
Reading the fingerprints behind hostile edits.
Verify any agency's portfolio
The same public-evidence techniques, buyer edition.
A History of Paid Wikipedia Editing
20 years of scandals, rule changes, and the diligence checklist they dictate.
The Wikipedia Risk Report 2026
An evidence-based risk model and the Source Strength Index.
Have a Wikipedia situation that needs a careful read?
Tell us what's happening and what you'd like to know. We'll come back with scope, timeline and an NDA proposal.