Previous
Beyond the survey: what policy citations can tell us about think tanks
Every year, On Think Tanks’ State of the Sector survey offers a valuable snapshot of how think tanks understand their own work, the issues they focus on, how they define impact, and whether they believe research influences policy.
At Overton, we look at the same world from a different angle. Rather than asking think tanks how they see themselves, we observe where their work appears in policy documents.
Neither perspective tells the whole story. Surveys capture experience and perception. Citation data captures one visible route through which research enters policymaking.
Read together, however, they allow us to ask a more interesting question:
How closely does what think tanks say about themselves match what appears in the policy record?
To explore this, we compared responses from Anglosphere (UK, Canada, USA and Australia) think tanks in the 2026 State of the Sector survey with policy citation data from Overton. Because the survey is anonymous, these aren’t matched think tanks, but two independent views of the same sector.
A sector confident in evidence
Compared with respondents globally, Anglosphere think tanks are notably more optimistic about the role of research in policymaking. Around three-quarters believe policymakers value evidence; they’re more likely to identify policy change as their primary measure of impact, and they see the media as more receptive to research than respondents elsewhere.
That paints a picture of a sector that feels research has a genuine role to play in the policymaking process. 
Anglosphere vs Global – self-reported confidence in evidence uptake
The citation data can’t confirm those perceptions directly, but it can offer another perspective by showing where research appears in published policy documents.
Where priorities align
Looking across 438 Anglosphere think tanks with at least 50 policy citations, we compared the policy areas organisations report working on in the survey with the UN Sustainable Development Goal (SDG) categories assigned to the policy documents that cite their research in Overton. The two classification systems aren’t directly equivalent, but they offer a useful basis for broad comparison.
At that level, the picture is largely consistent. Economic policy and governance are the two most commonly reported areas of work in the survey, while the policy documents citing these organisations are most frequently classified under SDG 8 (Decent Work and Economic Growth) and SDG 16 (Peace, Justice and Strong Institutions).
Artificial intelligence is one of the survey’s most frequently reported priorities, yet it has no clear counterpart within the SDG framework used for citation analysis. That reflects a limitation of the taxonomy rather than evidence that AI research isn’t influencing policy.
Health tells a different story. Policy documents classified under SDG 3 (Good Health and Well-being) account for a larger share of citations than health does as a self-reported area of work in the survey. These data alone can’t determine whether that reflects genuinely greater policy uptake, broader application of the SDG 3 classification, or differences between the two classification systems, but it highlights an area worth exploring further.
Citation share by SDG category
Policy visibility is highly concentrated
Among the 438 think tanks in our sample, policy citations are highly concentrated. Just five think tanks receive almost 29% of all policy citations, while the top twenty receive just over half, despite representing fewer than 5% of think tanks in the dataset.
Citation concentration across Anglosphere think tanks
This doesn’t mean the rest of the sector lacks policy influence. Rather, it highlights an important distinction between the two datasets.
The survey measures confidence that evidence reaches policymakers. Citation data measures one specific observable outcome: research being formally cited within published policy documents. Those are related, but they’re not identical. Research can shape consultations, advisory groups, private briefings and informal decision-making without ever producing a formal citation. Formal citations are only one way think tanks appear in the policy record; policy documents may also discuss or mention think tanks without referencing a specific research output.
Rather than contradicting one another, the two datasets appear to describe different dimensions of policy engagement.
Reading the two together
Taken together, these analyses suggest that the sector’s own understanding of itself broadly matches what appears in the policy record. Economic policy and governance emerge as leading priorities in both datasets, while differences on topics such as AI and health raise useful questions about how research is classified and how policy engagement becomes visible.
The citation data also adds an important nuance to the survey’s picture. Confidence that evidence reaches policymakers is widespread across the sector, but visible policy citations are concentrated among a relatively small number of established think tanks. That doesn’t imply other think tanks lack influence. Citation patterns likely reflect factors such as think tanks’ size, publication volume, international reach, and the types of policy engagement they undertake. Formal citations capture one important pathway through which research informs policy, but not the only one.
Neither dataset tells the whole story on its own. Surveys reveal how think tanks experience policy engagement; citation data shows one way that engagement appears in the policy record. Used together, they provide a richer perspective than either could alone.
What is Overton?
Overton connects research with public policy, helping users to find, evidence and grow the real-world impact of their work.
Our flagship product, Overton Index, is the world’s largest searchable database of policy documents and grey literature. Discover millions of documents from governments, IGOs, think tanks, and NGOs, and see what research, people, and policy they reference. Understand how evidence is applied across the policy landscape, offering deep insight into how ideas move from research into policy documents and between policies themselves.
A few caveats
The State of the Sector survey is anonymous, so individual think tanks cannot be linked to citation records. The Anglosphere survey sample is relatively small (n=29) and should be interpreted as indicative rather than definitive. Around 62% of citing documents could be mapped to SDG categories. Also, recent citation counts are likely to be affected by indexing lag and citation conventions used by policymakers.