How to Choose the Most Impactful Cause to Work On

Most people care about many issues: poverty, climate change, animal welfare, mental health, global health, and more. Yet our time, skills, and money are limited, so we face a strategic question: which causes should we focus on if we want to help as much as possible?

This article offers a practical framework for comparing causes and selecting the ones where you can have the greatest impact. It is inspired by ideas from effective altruism and evidence-based policy, but written as an independent, original guide.

Why Strategic Cause Selection Matters

Social impact is not distributed evenly across issues or interventions. Research in global health and development has shown that some programs can be orders of magnitude more effective than others at improving or saving lives. Small differences in cause choice and intervention choice can therefore translate into huge differences in overall impact.

• Resources are scarce: Philanthropic spending and public budgets are small compared with the scale of global problems.
• Opportunities vary: Some problems are closer to being solved, some are better understood, and some have neglected but promising solutions.
• Comparisons are possible: With reasonable assumptions and available data, we can often make approximate comparisons that are good enough to guide decisions.

Being deliberate about which causes we work on is therefore a powerful lever for doing more good.

The Core Idea: Three Foundational Criteria

A widely used starting point for cause comparison is to assess each candidate problem on three core dimensions:

• Scale: How large is the problem in terms of total harm or unmet need?
• Neglectedness: How many resources are already devoted to tackling it?
• Tractability: How solvable is the problem, at least at the margin?

These dimensions are not rigid rules, but they help you build a more systematic view of where additional effort is most promising.

Comparing Causes with the SNT Framework

This table is a simplification, but it provides a reference point for the rest of the framework.

Step 1: Clarify Your Objective and Ethical Assumptions

Before comparing causes, be explicit about what you are trying to maximize and what counts as a benefit. Different ethical views and risk attitudes can lead to different prioritization.

Questions to Ask Yourself

• Do you focus mainly on improving human welfare, reducing animal suffering, safeguarding future generations, or some combination?
• Do you prioritize saving lives, improving quality of life, increasing autonomy, or reducing inequality?
• How do you weigh near-term versus long-term benefits and risks?
• Are you especially concerned with catastrophic or existential risks (e.g., pandemics, nuclear war, AGI accidents)?

You do not need a fully worked-out moral philosophy. However, articulating your assumptions makes trade-offs more transparent and helps you understand why you might rank causes differently from others.

Step 2: Generate a Longlist of Candidate Causes

Next, create a longlist of causes that seem at least plausibly high-impact. You can refine later; the aim here is breadth.

Sources of Candidate Causes

• Global burden of disease data (e.g., leading causes of death and disability).
• Major international agendas (e.g., Sustainable Development Goals).
• Academic and policy work on catastrophic risks and global priorities.
• Personal exposure, lived experience, or professional expertise.

At this stage, list causes at a relatively broad level, such as “global health,” “climate mitigation,” “pandemic preparedness,” or “criminal justice reform,” rather than individual projects.

Step 3: Qualitative Comparison Using Key Questions

For each cause on your longlist, perform an initial qualitative assessment. This does not require detailed numerics; you can use rough ratings (e.g., low/medium/high) for each dimension.

Assessing Scale

• How many people (or animals) are affected each year?
• How severe are the harms (e.g., death, disability, extreme poverty, chronic suffering)?
• What is the cumulative impact over decades if the problem continues?

For example, the Global Burden of Disease project provides systematic estimates of deaths and disability-adjusted life years (DALYs) by cause, which can be used to approximate the scale of health-related problems.

Assessing Neglectedness

• How much money is spent annually on this issue by governments, philanthropies, and markets?
• Is the issue well-represented in policy debates and research agendas?
• Are the existing actors coordinated and effective, or fragmented and under-resourced?

Issues like basic scientific research into pandemic risks or certain non-commercial diseases may be relatively neglected compared with their potential importance.

Assessing Tractability

• Are there evidence-based interventions or policies that already work at small scale?
• Can you see plausible pathways from additional resources to measurable progress?
• Are there “low-hanging fruit” or policy windows that could make change easier?

Randomized evaluations and quasi-experimental studies can help identify interventions that reliably improve outcomes, particularly in global health and development.

Step 4: Move Toward Quantitative, Back-of-the-Envelope Estimates

Once you have a qualitative sense of the most promising causes, you can deepen the analysis by building back-of-the-envelope (rough) quantitative estimates. These will never be perfect, but they can reveal orders-of-magnitude differences between options.

Elements of a Rough Quantitative Model

• Baseline: How much good is currently being done in this area per unit of resource?
• Marginal impact: How much extra good might be done by adding one more person or an additional dollar?
• Room for more funding or talent: Are there credible opportunities that could absorb more resources?
• Uncertainty: What are the main assumptions that drive your estimate?

To make the process manageable, treat these estimates as tools for learning rather than precise forecasts. Adjust them when you get new data or feedback.

Step 5: Integrate External Evidence and Expert Opinion

Your own reasoning is valuable, but it should be complemented by external evidence and expert knowledge.

Useful Sources of Evidence

• Systematic reviews and meta-analyses in medicine, economics, and social science, which synthesize evidence across multiple studies.
• Global policy assessments that weigh costs and benefits of interventions (e.g., some international economic analyses of development priorities).
• Technical reports on risk assessment, such as those on pandemic preparedness, nuclear security, and climate scenarios.
• Direct conversations with practitioners, researchers, or policymakers working in each cause area.

When incorporating expert opinion, pay attention to track records, diverse perspectives, and incentives. Where possible, favor evidence that is transparent and methodologically rigorous.

Step 6: Consider Personal Fit and Strategic Positioning

Cause selection is not only about which problems matter most in the abstract. It also depends on what you personally can do.

Dimensions of Personal Fit

• Skills and expertise: Do you have relevant knowledge, or could you realistically acquire it?
• Motivation: Can you see yourself working on this problem for years without burning out?
• Network and access: Do you have connections or credibility that could unlock opportunities?
• Risk tolerance: Are you willing to pursue uncertain, long-shot projects, or do you prefer safer, incremental work?

Two people using the same cause-prioritization framework might sensibly choose different causes because their comparative advantages differ.

Step 7: Build a Portfolio of Causes

You do not necessarily have to commit all your resources to a single cause. In many situations it makes sense to build a portfolio that balances different types of impact.

Possible Portfolio Approaches

• Core focus with side bets: Devote most of your career or giving to one main area, while reserving a portion for exploratory or high-risk causes.
• Diversification across time horizons: Combine near-term interventions (e.g., disease control) with longer-term risk reduction (e.g., pandemic preparedness or climate resilience).
• Balancing evidence and innovation: Support both well-tested programs and more experimental work that could open up new opportunities.

Portfolio choices reflect your beliefs about uncertainty, learning, and the value of information. In fields like causal inference and risk analysis, combining multiple models and approaches can yield more robust decisions than relying on a single narrative.

Step 8: Treat Cause Selection as an Ongoing Learning Process

Cause prioritization is not a one-time decision. Evidence changes, new technologies emerge, and your personal situation evolves.

Principles for Continuous Improvement

• Regularly revisit your views: Schedule periodic reviews (e.g., annually) to update your cause rankings.
• Track outcomes: Where possible, monitor measurable indicators of progress in your chosen causes.
• Reflect on mistakes: When predictions fail or projects underperform, analyze why and adjust your models.
• Stay informed: Follow high-quality research outlets, policy reports, and expert communities working on global priorities.

This mindset mirrors how scientific fields use causal inference frameworks to gradually refine their understanding of interventions and policies over time.

Practical Worksheet: Applying the Framework

To make the ideas above concrete, you can use a simple worksheet when comparing causes. For each candidate cause, rate it from 1 (low) to 5 (high) on each dimension, then add notes.

The scores are rough and subjective, but the exercise encourages explicit reasoning and comparison rather than relying solely on intuition.

Frequently Asked Questions (FAQs)

Q: Isn’t it arrogant to compare causes and say some are more important?

Comparing causes does not imply that some people’s suffering matters less. It simply acknowledges that different actions produce different amounts of good. Evidence from global health and development clearly shows that some interventions save many more lives per dollar than others, and being aware of this allows us to help more people overall.

Q: What if my values differ from standard economic or health metrics?

You can adapt the framework to your own values by changing what you measure (e.g., putting more weight on mental health, animal welfare, or long-term risks). The structure—thinking about scale, neglectedness, and tractability—still helps, but you fill in the details with your own ethical assumptions.

Q: How do I handle very uncertain, long-term risks?

For issues like existential risks or advanced technology, empirical data is limited. In these cases, you may rely more on expert judgement, scenario analysis, and robustness: favor actions that look beneficial across a range of plausible models. Some researchers argue that reducing extinction risks can have immense expected value because it affects all future generations.

Q: Do I need advanced statistics to use this framework?

No. While sophisticated methods from causal inference and economics are valuable for research, individuals and small organizations can make significant progress with simple comparisons, transparent assumptions, and a commitment to revising their views when new evidence appears.

Q: How often should I reconsider my choice of cause?

A reasonable approach is to review your cause priorities on a regular schedule (for example, every one to three years) or when major new evidence or opportunities arise. This balances stability—important for building expertise—with flexibility to update when the world changes.

Written by

Medha Deb

Medha Deb is an editor with a master's degree in Applied Linguistics from the University of Hyderabad. She believes that her qualification has helped her develop a deep understanding of language and its application in various contexts. Medha specializes in the areas of beauty, health, sports, and wellness and is committed to ensuring that the content on the website is…