Poliakoff, S., & Toepfl, F. (2026). Prigozhin’s propaganda team: The st Petersburg internet research agency (2013–2021). Europe-Asia Studies, 1–22. https://doi.org/10.1080/09668136.2025.2588334
Summary
This study offers the first systematic academic examination of the organizational structure of the St Petersburg Internet Research Agency (IRA) — Yevgeny Prigozhin’s infamous “troll factory” — rather than its content output. Drawing on an open-source intelligence (OSINT) approach, the authors mine 350 CVs that former IRA employees self-published on Russian job search platforms, reconstructing the agency’s workforce demographics, recruitment timeline, internal hierarchy, and functional divisions across 2013–2021. The central argument recasts the IRA not as an isolated covert operation but as a labor-market actor structurally resembling a mid-sized media or PR firm, tightly integrated into Russia’s legitimate media industry and best understood as Kremlin-aligned “outsourced trolls” privately operated by Prigozhin rather than a directly state-funded body.
Key Contributions
- First systematic academic study of the IRA’s organizational structure, workforce, and evolution instead of its content.
- Introduces a novel OSINT methodology using self-published CVs to study covert disinformation organizations — transferable to comparable global cases.
- Provides reliable demographic data resolving contradictions in prior journalistic accounts.
- Documents a detailed five-level hierarchy and six functional task categories.
- Demonstrates the IRA’s embedding in the legitimate Russian media labor market, extending framings of disinformation-for-hire as nested within national media industries.
- Argues for a “Kremlin-aligned outsourced” rather than “state-funded” characterization, with implications for theorizing state–private disinformation arrangements.
Methods
The authors manually collected 393 CVs (deduplicated to 350) from HeadHunter and SuperJob in late 2021–early 2022, searching by known IRA-affiliated entity names (GlavSet, MediaSintez, Internet-issledovaniya, etc.). Structured data were extracted via JavaScript and then manually coded for education, university location, and job category, with strong inter-coder reliability (Krippendorff’s α = 0.947). Free-text job descriptions were analyzed qualitatively in MaxQDA using grounded-theory open and axial coding. The design included IRB approval, pseudonymization, and explicit assessment of self-selection, temporal, and social-desirability biases.
Findings
- Workforce was 52% male / 48% female; median age 26; 98.7% Russian citizens; median tenure 2.3 years.
- 78.3% held university degrees, mostly from St Petersburg universities; dominant fields were economics/management (22.2%) and journalism/PR/advertising (19.2%).
- 27.4% had no prior work experience; only 14.3% reported advanced English, implying limited foreign-language operational capacity.
- Recruitment peaked in Q1 2014 (51 hires) aligning with the Crimea annexation, with a sharp decline before the 2016 US election — suggesting the IRA was not primarily built to target that election.
- Six functional task areas: mimicking citizens, mimicking journalism, advertising/marketing, media monitoring/analysis, visual propaganda, and administrative/technical support.
- Five-level hierarchy: specialists; content managers/journalists; team leaders/editors; deputy heads/editors-in-chief; CEOs.
- Operations spanned many platforms (Twitter, Facebook, Instagram, Telegram, YouTube, VK, OK, etc.), with VK groups and YouTube channels still active in 2025.
- Sophisticated advertising tools (Google AdWords, Yandex.Direct, CPA and teaser networks) plus in-house tools like a Russian synonymizer and Telegram/VK bots.
- 32 CVs documented seamless transitions to Prigozhin’s Patriot Media Group, often with identical job descriptions; camouflage companies kept employing staff after formal closure.
Connections
This paper’s focus on the labor and organizational infrastructure behind Russian influence operations complements platform-side and content-side studies of state-linked information disorder, such as work on Russian disinformation campaigns and protest suppression narratives (Kuznetsova2025-nu, Kulichkina2026-zk). Its OSINT and production-studies lens on coordinated actors offers an organizational counterpart to the coordinated-inauthentic-behavior detection literature that emphasizes network and content signatures (Luceri2025-tr, Minici2024-tf, Kansaon2025-id).
Podcast
A research-radio episode discusses this paper: 🎧 MP3 · Spotify · Apple Podcasts