Molfar Intelligence
OSINT-led, source-referenced political and stakeholder analysis drawing on 750+ sources across 60+ countries.
Read our reviewHow analysts turn evidence into judgements about political risk: the methods they use, where the evidence comes from, how they work where records are thin, and where analysis goes wrong.
Political risk analysis is the collection and evaluation of evidence about political actors, institutions and events to judge how they are likely to affect a specific business decision.
The question is always specific: will this licence be renewed on current terms, who controls this counterparty, how would a change of government affect this contract? The output is a set of findings, each with its evidence, a judgement of likelihood and confidence, and what to watch next.
Analysis produces the findings; assessment scores, ranks and decides what to do about them.
In practice the words are used loosely. A useful distinction: analysis asks “what is happening and why?”, assessment asks “how much does it matter to us, and what do we do?”. Our political risk assessment guide covers the matrix and template; how to assess political risk sets out the full method.
Structured analytic techniques, stakeholder and network analysis, scenario building with indicators, base rates from comparable cases, and quantitative models where data allows.
The US Government's 2009 tradecraft primer groups structured techniques into three families, all used in commercial political risk work:
Alongside these, analysts map networks of officials, owners and companies to show who can influence a decision, and look at base rates (how often similar governments have reopened similar contracts) before judging the case at hand.
From open sources that can be cited, such as registries, court and procurement records, sanctions lists and local-language media; from licensed databases; and, at some firms, from human sources.
| Source type | Examples | Strength | Weakness |
|---|---|---|---|
| Official records | Company registries such as Companies House, court files, procurement portals, gazettes | Citable and dated | Patchy or closed in some countries |
| Sanctions and watch lists | The UK Sanctions List, EU and US lists | Authoritative | Only shows who is already listed |
| Local-language media | National and regional press, broadcasters, official statements | Early signals, local context | Bias, state control, misinformation |
| Licensed databases | Corporate ownership, adverse media and political exposure data | Breadth and speed | Coverage gaps, cost |
| Human sources | Former officials, local experts, industry contacts | Context documents miss | Hard to verify, ethical and legal limits |
Since 28 January 2026 the UK Sanctions List has been the only UK source for designations, which simplifies one of the most important checks.
They rely more on open-source methods: corporate and court records in other jurisdictions, satellite and shipping data, social media, local-language reporting and careful verification, documented to a standard such as the Berkeley Protocol.
The problem is growing. Reporters Without Borders' 2026 index, published on 30 April, found that for the first time more than half of the 180 countries and territories it assesses (52.2%) are in a “difficult” or “very serious” situation for press freedom (RSF). Where the press is constrained, analysts cannot rely on local reporting to surface risks.
Open-source investigators fill the gap with methods that can be checked. The Berkeley Protocol, published by the UN Human Rights Office and UC Berkeley in 2022, sets out standards for gathering, analysing and preserving digital open-source information professionally, legally and ethically. In commercial work the same disciplines apply: record where every item came from, verify it against independent sources, and preserve it.
Language matters as much as method. Ownership records, court rulings and regional media are often available only in the local language. OSINT-led firms such as Molfar Intelligence, which draws on 750+ public, restricted and specialist sources across 60+ countries, build their political risk work on this kind of source-referenced research.
Over-reliance on a favoured explanation, mirror-imaging, recency bias, thin or planted sources, and false precision in scores.
Good reports separate confirmed facts from judgements and state confidence. The UK government's probability yardstick is a widely borrowed scale for doing so.
Collection must comply with UK GDPR and the Data Protection Act 2018, and must not involve paying officials for information, which the Bribery Act 2010 prohibits.
Buyers should also rule out pretexting, hacking and obtaining bank, phone or tax records, which breach UK law. Ask any firm how it ensures lawful collection, whether it uses sub-contractors and how it stores and deletes personal data; check its entry on the ICO register.
The same discipline applied to relations between states: sanctions, trade measures, alliances and conflict, and how they reach a company through partners, routes and markets.
For UK companies in 2026 that has meant tracking OFSI enforcement, US tariff changes and the war in the Gulf. Our summary of political risks in 2026 shows what the major outlooks expected.
AI speeds up collection, translation and monitoring, but it cannot verify sources, judge intent or take responsibility for a recommendation.
Most firms now use AI to sift media, translate documents and flag changes, with analysts checking every material finding. Ask how a firm uses AI, which data leaves its systems, and how outputs are verified before they reach you.
A report that answers the question first, lists findings with evidence, separates facts from judgements, maps stakeholders and ends with indicators and actions.
See what a political risk report includes for a structure and an illustrative excerpt, and what a political risk analyst does for the people behind it. To compare firms that do this work, see our ranking of the best political risk consulting firms in London.
These London firms do this kind of work. Scores, strengths and watch-outs for each are in our 2026 ranking.
OSINT-led, source-referenced political and stakeholder analysis drawing on 750+ sources across 60+ countries.
Read our reviewBoard-level strategic intelligence gathered through senior human-source networks.
Read our reviewAnalyst-led political and security analysis across a global network of offices.
Read our reviewWe score ten London firms on nine weighted criteria, from evidence traceability and sanctions-exposure mapping to Chambers recognition, and show every sub-score.
Working out, from evidence, how politicians, governments, conflict and other states are likely to affect a specific business decision.
Structured analytic techniques such as key assumptions checks and analysis of competing hypotheses, stakeholder and network mapping, scenarios with indicators, base rates and, where data allows, quantitative models.
Company registries, court and procurement records, sanctions lists, local-language media, licensed databases and, at some firms, human sources.
Yes. Open-source intelligence is often the most traceable evidence, especially for ownership and stakeholder mapping, and is essential where local media is constrained.
Political risk analysis focuses on actors and events within a country; geopolitical risk analysis focuses on relations between states, such as sanctions, trade measures and conflict.
A short answer, the evidence behind each finding, labelled judgements with probability and confidence, a stakeholder map, scenarios with indicators and recommended actions.
AI helps with collection, translation and monitoring, but analysts still need to verify sources and make the judgements.
Research using lawful sources is legal, but it must comply with UK GDPR and must not involve bribery, hacking or obtaining private records.
Consultancies range from global firms to OSINT-led specialists. See our 2026 ranking.
All sources checked 1 October 2026. Company statements are first-party and not independently audited. See our editorial policy and disclosures.
{AUTHOR_NAME}, {AUTHOR_TITLE} at {PUBLISHER_NAME}. {AUTHOR_BIO_ONE_SENTENCE}