This article was originally published on the Future Matters website by School for Thinktankers alum Vegard Beyer. Read it here.
In 1991, researchers asked participants how much they were willing to pay to prevent migrating birds from drowning in oil ponds. One group was told two thousand birds were at risk; another, 20,000; a third: 200,000. The answers came back: $80, $78, and $88. A hundred-fold difference in scope had almost no effect on how people responded.
Policy advocates make equivalent mistakes every day – just with much higher stakes. When deciding which policy to pursue, experienced advocates tend to anchor on what is trending in the news, what their peers are working on, or what their organisation has historically done. While you can hardly work 100x as hard or as effectively, your choice of policy objectives can very well make a 100-fold difference in how much impact you achieve.
In 2023, Future Matters applied a systematic method to identify EU climate policy priorities for the incoming 2024–2029 mandate. One result was an early focus on EU-India cooperation on industrial decarbonisation, specifically steel, as an urgent opportunity to avert a risk of decades-long industrial emissions lock-in – at a time when the few climate advocates with a Europe-India focus were looking at India’s energy transition and coal exit, and had not yet turned their attention to heavy industry. Two years later, the EU and India agreed at their January 2026 summit in New Delhi to cooperate on green steel production. Future Matters recommendations were mirrored in policy documents along the way. What made this opportunity visible to us early was a replicable method for identifying high-impact, underrated policies before they become obvious.
This article describes that method, developed by Future Matters through three research projects and dozens of consulting cases since 2020. The method draws on the importance-tractability-neglectedness framework first developed by Holden Karnofsky at GiveWell Labs in 2013 and later formalised by 80,000 Hours – and adapts it to the specific challenges of policy advocacy, where the measure of impact is not allocated donor funds but policy outcomes achieved. The method is domain-agnostic – Future Matters uses the same logic for climate policy, AI governance, and biosecurity – and can be run as a workshop of one to three hours with a small group of experts.
Why instinct fails
We observe that three cognitive biases regularly distort advocacy priorities.
- Scope insensitivity. As the bird study illustrates, our intuitions about scale are poor. We struggle to feel the difference between a policy that could prevent 50 million tonnes of CO₂ emissions and one that could prevent 500 million tonnes. Both feel like “big wins” – the numbers are too large to carry emotional weight. Only deliberate analysis can surface the difference.
- Availability and recency bias. We overweight what is in the news and what colleagues are currently discussing. A piece of legislation attracting media coverage will draw advocates even when less-covered opportunities have far higher potential impact.
- Groupthink on neglectedness. The more advocates are already working on a problem, the more it can seem like the right problem to work on. But a policy that twenty well-resourced organisations are already pushing hard on likely has less room for additional contribution.
These are well-known properties of how human minds process information under uncertainty. A systematic method is more reliable than intuition for decisions of this complexity.
Core criteria for policy prioritisation
To evaluate which policy deserves your attention, assess it against five criteria – separately.
- Impact potential: How much could this policy advance your goals globally, over the long term? Look for policies that operate at national or international scale and link clearly to the problem you are trying to address. A clean energy transition partnership between major economies may prevent hundreds of megatons of emissions; a domestic efficiency programme in a mid-sized country, perhaps a few. The relevant question is not “is this good?” but “how much good could this realistically do?”
- Effort: What would it take to get this policy adopted? Account for the full scope of work – research, policy development, coalition-building, advocacy – from initial idea to successful adoption. Some policies are already well-researched, well-regarded, and just need a champion to carry them into a few specific fora. Others require years of groundwork, even research or underlying norm shifts, before advocacy can begin.
- Feasibility: Given current political, economic, and social conditions, how likely is a serious push for this measure to succeed? Consider political opposition, social acceptance, windows of opportunity, institutional capacity, and whether conditions are ripe for change. A proposal with high impact potential is worthless if it truly has no path to adoption in a relevant timeframe. (In practice, Future Matters models this as separate feasibility scores at different stages of the policy process – policy proposal, committee, adoption, implementation – since the task may be easy at one stage and hard at another.)
- Neglectedness: How many other advocates are already working on this? The concept of counterfactual impact is essential here: What would not happen if you did not contribute? If a policy is already being pushed effectively by well-resourced organisations, your additional effort substitutes for work that would happen anyway – and produces no net impact at all. An opportunity where your team could genuinely make the difference between success and failure is worth far more.
- Fit with your skills: Do you have the specific expertise, political relationships, or institutional credibility this policy requires? Your comparative advantage is where you can contribute something others cannot easily replicate.
It is important to evaluate policies on each of these criteria separately, in order to capture orders-of-magnitude differences and interactions between the criteria. A policy requiring considerable effort may still be worthwhile if its impact is proportionally larger and few others are attempting it. A seemingly high-impact policy becomes much less attractive if it is already crowded with well-resourced, well-positioned advocates.
One important consideration alongside the positive criteria: complex policy systems can generate harms that are hard to foresee. Systematically consider how your proposed policy might be misused or create harmful side effects. This is not a reason to avoid ambitious advocacy, but a reason to design your process to consider risks. For advocates working on dual-use technologies or biosecurity, this step is particularly non-trivial.
A systematic four-step process
The following process is designed to overcome the limitations of intuitive assessment when comparing multiple policy options across several dimensions simultaneously. It can be run as a one-to-three-hour workshop with a small group of policy experts and a facilitator.
Step 1: Generate, research, and eliminate
Brainstorm the policy options that could plausibly have high long-term impact on your goals. Conduct initial research on each: What is the scale of potential impact? Who is already working on it? What are the main barriers? What would sustained advocacy actually require?
Use this initial pass to eliminate obvious non-starters: policies with negligible impact, insurmountable political barriers, or those already being pursued by virtually every major organisation in your field.
One discipline we’ve found worth enforcing is to research all options at similar depth. A thoroughly researched option might appear stronger simply because you know more about it, not because it is actually better.
Step 2: Rank options under each criterion separately
Rather than weighing everything at once, rank all remaining options under each criterion separately – one criterion at a time. Human minds cannot reliably compare options along multiple dimensions simultaneously. Isolating each criterion produces more reliable judgements. Have each participant rank independently before comparing notes and reaching consensus; independent preparation reduces anchoring on whoever speaks first.
For each criterion, first identify which policy performs best (rank 1) and which performs worst. Then arrange the others in between. (For Effort: Which solution takes the most effort? That is your number 1.)
Step 3: Assign quantitative scores to capture relative differences
Rankings tell you which option is better, but not by how much – and the difference between “twice as good” and “ten times better” drives very different decisions. Our intuitions handle relative comparisons far better than absolute quantitative assessments. Give the top-ranked option under each criterion a score of 100, then rate all others relative to that anchor.
For example:
- If the second-ranked policy has about half the impact of your first, give it a score of 50.
- If the last-ranked policy has roughly one-tenth the impact of the first, give it 10.
Repeat for all criteria. Again, prepare individual ratings first, then discuss and reach consensus. The exercise reliably surfaces disagreements that were invisible in a qualitative conversation.
Step 4: Calculate counterfactual impact per effort and visualise results
Calculate a priority score for each option:
Priority Score = (Impact × Feasibility × Neglectedness × Fit) ÷ Effort
Impact and Effort use your 0–100 scores. Feasibility, Neglectedness, and Fit are expressed as percentages (0–100%). One thing to watch: Effort is ranked so that the most resource-intensive policy scores 100 – the higher the score, the more effort required. Dividing by it in the formula then correctly penalises costly options.
The score represents expected counterfactual impact per unit of effort, in relative terms. If Policy A scores an overall 0.6 and Policy B scores 0.3, the same investment of effort in Policy A is likely to produce twice the impact.
To visualise results, plot each policy with Effort on the horizontal axis and the numerator (Impact × Feasibility × Neglectedness × Fit) on the vertical axis. Policies in the upper-left quadrant – high impact, low effort – are your most efficient opportunities. Policies in the lower-right should likely be avoided. The upper-right contains high-impact policies requiring substantial effort – potentially worthwhile if resources allow.
The slope of the line from the origin to each plotted point represents impact efficiency: a steeper slope means more impact per unit of effort invested.
Building a strategic portfolio
Most policy options will not achieve their full potential. Political conditions shift, coalitions collapse, key officials leave. A resilient strategy does not bet everything on a single priority.
With limited resources, a sensible portfolio might combine one or two medium-effort policies with solid impact potential alongside a single high-effort, high-impact bet. This spreads risk while preserving the possibility of transformative wins. Think of it as the policy advocacy equivalent of a diversified investment strategy – not because all options are equally good, but because any individual option carries meaningful execution risk.
Applying the methodology: the EU-India steel case
In 2023, Future Matters asked how to maximise impact on global emission mitigation through European Union policy over the 2024–2029 period.
The team conducted a prioritisation study drawing on five datasets covering global emissions through the end of the century. After reviewing 585 publications and interviewing 104 policy experts, the process produced eight EU policy priorities – those with the highest potential impact on global emissions, sufficient political feasibility, and comparatively few existing advocates.
One of those eight was an EU-India partnership on low-carbon industrial production, focused on the steel sector. India’s steel industry has the potential to lock in up to 125 Gt CO₂e of cumulative emissions by 2100 if it follows a conventional development pathway – or to avoid most of those emissions on a low-carbon one. Most European advocates were focused on India’s energy transition and coal exit; industrial decarbonisation was comparatively neglected. And 2025 offered a clear political window: the EU was planning to revise its strategic partnership with India.
Future Matters dedicated a significant share of its climate team’s resources to this opportunity. When specific EU-India cooperation proposals were being developed, the team applied the same prioritisation logic at a more granular level: which bottlenecks in India’s clean steel transition could EU support address most effectively, and which of those was the EU well-positioned to address?
In January 2026, the EU and India agreed at their summit in New Delhi to cooperate on industrial decarbonisation in heavy industries, with green steel as a focus. Future Matters was one of several actors contributing to this outcome. Systematic prioritisation enabled the team to identify an opportunity that most actors had underweighted, and to commit resources to it before the policy window opened – which is the whole point of the method.
When to apply this methodology
The approach works best when you have genuine flexibility in choosing which policies to pursue, several options with potentially significant differences in impact, and enough capacity to do some research across options. It is less useful when your mandate is narrow and predetermined, or when only one viable option exists.
Natural moments to apply it might be:
- Choosing your priorities at the start of a new legislative period
- When the political landscape shifts significantly
- When a new policy package is proposed and you need to decide which provisions to prioritise
- When setting or reviewing an annual advocacy strategy
The method can be applied at multiple levels – annual strategy, quarterly review, specific proposal selection within an already-identified priority – and becomes more powerful when the outputs of one level feed into the next.
Getting started
A first run through the method rarely produces definitive answers. It more typically reveals what you do not yet know, and where additional research would most change your conclusions – which is itself a useful result. A two-hour workshop with a small group of advocates and experts can surface significant differences in expected impact per effort that were previously invisible.
A few practical starting points:
- Adapt the criteria to your specific policy domain – the exact dimensions may need adjustment even if the underlying logic is similar
- Treat the first round as diagnostic: it will tell you what you need to research next
- Involve people with different vantage points – the most valuable disagreements will emerge during the consensus-building steps
You can read about how Future Matters applied this methodology across EU climate policy in the research report 8 EU Policy Priorities for Global Decarbonisation. The same approach aids our impact on AI governance and biosecurity. If you want to learn more about our approach or discuss how systematic prioritisation could inform your advocacy strategy, reach out at [email protected]
Carl Frederick Luthin is a Project Manager in Climate Protection at Future Matters, where he serves think tanks, NGOs, and policymakers.
Vegard Beyer is the co-founder and Director of Partnerships & Communications at Future Matters, where he leads EU climate work and strategy consulting.