Doing Good Better

Doing Good Better Plot Summary

Introduction

Most of us want to make a difference in the world, but our good intentions often lead to underwhelming results. We donate to charities without knowing if they're effective, we pursue careers based on vague notions of social impact, and we rarely consider the actual consequences of our altruistic actions. Even when we try to help others, we frequently fail to think critically about the most important question: how can we do the most good possible with our limited resources? Effective altruism offers a solution to this problem by applying data, reason, and rigorous analysis to our altruistic endeavors. Rather than relying solely on emotional appeals or conventional wisdom, effective altruism examines evidence to determine which interventions genuinely improve lives. Through this framework, we can discover that some approaches to helping others are drastically more effective than others—not just marginally better, but often hundreds or thousands of times more impactful. By combining both heart and head, we can transform our good intentions into astonishingly good outcomes and maximize the difference we make in the world.

Chapter 1: Measuring Impact: How to Evaluate What Truly Makes a Difference

When attempting to determine which interventions make the biggest difference, we must first establish how to measure impact meaningfully. Traditional metrics often focus on inputs (how much money was spent) or outputs (how many services were provided), but these fail to capture what truly matters: outcomes for the people we're trying to help. The quality-adjusted life year (QALY) represents one approach to measuring impact in healthcare. One QALY equals one year of life in perfect health, or equivalent combinations like two years at 50% health. This metric allows us to compare interventions treating different conditions—a program addressing blindness can be compared with one treating malaria. While imperfect, QALYs provide a concrete way to assess the actual benefit to recipients rather than just counting services delivered. Beyond health, impact evaluation requires considering the counterfactual—what would have happened otherwise. Many charities claim success when their beneficiaries improve, but without a control group, we cannot know if they would have improved anyway. Randomized controlled trials, where some randomly selected people receive an intervention while others don't, help establish causation. Michael Kremer's pioneering work in development economics demonstrated this approach, showing that many intuitive interventions (like providing textbooks) had minimal impact, while less obvious ones (like deworming) dramatically improved educational outcomes. Measuring impact also requires acknowledging opportunity costs. Every dollar spent on one program cannot be spent on another, which means we must make hard tradeoffs. Like a doctor performing triage after a disaster, we cannot help everyone, and failing to prioritize can result in helping fewer people overall. This reality may be uncomfortable, but ignoring it doesn't make it disappear. Effective impact measurement further demands attention to scale. Small pilot studies might show promising results, but these don't always translate when programs expand. Conversely, some interventions become more cost-effective at scale. Understanding these dynamics helps identify truly transformative approaches rather than merely promising experiments. Finally, long-term impacts matter tremendously but are often neglected in evaluation. Programs that appear successful in the short term might have negative consequences years later, while others might generate benefits that compound over decades. Comprehensive impact measurement requires grappling with this uncertainty through careful research and follow-up studies.

Chapter 2: Maximizing Effectiveness: Why Some Interventions Outperform Others by Orders of Magnitude

The most striking insight from rigorous impact evaluation is the vast difference in effectiveness between interventions. Within global health, the best programs aren't just marginally better than average ones—they're often hundreds of times more effective. This pattern, where a small subset of activities produces most of the value, follows what statisticians call a "fat-tailed distribution" rather than a normal bell curve. Consider deworming programs. For approximately $0.50 per child, medication can eliminate parasitic worm infections, leading to significant improvements in school attendance, cognitive development, and eventual adult earnings. Studies show this intervention provides approximately 100 times more benefit per dollar than distributing textbooks in the same contexts. This tremendous difference exists even though both programs target education in developing countries and both seem intuitively beneficial. These enormous effectiveness gaps appear across multiple domains. In poverty alleviation, direct cash transfers to the extremely poor have proven remarkably effective, with recipients typically investing in assets like metal roofs or livestock that yield substantial returns. Meanwhile, more complex microfinance programs often show minimal impact despite their popularity. The evidence suggests giving money directly to the poor is frequently more effective than implementing elaborate development schemes. The existence of these effectiveness gaps means our choices about which interventions to support have enormous consequences. If we support interventions that are merely "pretty good" instead of those that are outstanding, we might achieve only 1% of the impact we could have had. This makes choosing the most effective interventions not just marginally better but ethically imperative if we genuinely want to help others as much as possible. Several factors drive these effectiveness differences. First, interventions addressing preventable problems typically outperform those addressing harder problems. Preventing malaria is far more cost-effective than treating advanced cancer. Second, interventions in neglected areas usually have higher returns than those in crowded fields, as the "low-hanging fruit" remains unpicked. Third, simple, evidence-backed interventions typically outperform complex ones with multiple moving parts. Most critically, effectiveness follows a power law because the most successful approaches can be replicated and scaled. A breakthrough intervention like oral rehydration therapy for diarrheal disease can spread globally, saving millions of lives at minimal cost. This scalability creates a multiplicative effect that makes the best interventions dramatically more impactful than merely good ones.

Chapter 3: Neglected Problems: Finding Overlooked Opportunities for Greatest Impact

One of the most powerful insights from effective altruism is that we can often achieve dramatically more good by focusing on neglected problems rather than popular causes. The principle of diminishing marginal returns suggests that additional resources directed toward already well-funded areas typically produce less benefit than resources directed toward neglected areas of comparable importance. Consider global disease burden. Malaria kills hundreds of thousands annually and disables millions more, yet receives proportionally far less funding than diseases like breast cancer or ALS that affect fewer people. This funding disparity doesn't reflect differences in disease severity but rather factors like geographic proximity, media attention, and emotional appeal. The result is that interventions addressing neglected tropical diseases often achieve far more impact per dollar than those addressing well-publicized conditions. This pattern extends beyond health. After natural disasters, media coverage generates enormous funding surges that frequently exceed what organizations can effectively deploy. Meanwhile, ongoing crises like civil wars or chronic malnutrition receive minimal attention despite affecting far more people. A donation to an overlooked humanitarian crisis often helps more people than a donation to disaster relief after a high-profile earthquake or hurricane. The neglectedness principle also reveals opportunities within popular cause areas. Climate change receives substantial attention, but specific aspects remain surprisingly neglected. While renewable energy attracts significant investment, research into climate adaptation for vulnerable populations remains chronically underfunded. Similarly, while endangered mammals receive conservation resources, protecting keystone species in marine ecosystems often delivers greater ecological benefits despite minimal funding. Neglectedness frequently correlates with other factors that magnify impact. Neglected problems often affect the world's poorest people, who can benefit tremendously from even modest interventions. They typically lack powerful advocates, meaning that your support may be the difference between action and continued neglect. And they often involve preventable harms where early intervention proves far more effective than later remediation. Identifying neglected problems requires looking beyond emotional appeals and media coverage to examine actual resource allocation. Government spending, foundation grants, corporate investments, and public attention all provide indicators of which problems receive adequate resources and which remain overlooked. By deliberately seeking out important but neglected issues, we can identify opportunities where additional resources can make an extraordinary difference.

Chapter 4: Counterfactual Thinking: Assessing What Would Have Happened Otherwise

Counterfactual thinking—evaluating what would have happened without our intervention—stands as a critical but frequently overlooked element of effective altruism. Without this perspective, we may mistakenly believe we're making a significant difference when we're actually having little or even negative impact. Consider career choice. Many idealistic graduates pursue careers in international development organizations, believing they'll make a substantial difference. However, these positions often attract numerous qualified candidates, meaning if you hadn't taken the job, someone similarly capable would have. Your counterfactual impact isn't the full value of what you accomplish in the role, but rather the difference between your performance and that of the next-best candidate. This insight has led some effective altruists to pursue "earning to give"—taking high-paying jobs and donating substantially to effective charities—rather than working directly for nonprofits. The counterfactual perspective also reveals why supporting certain popular programs can prove counterproductive. The "Scared Straight" program, which brought juvenile offenders to prisons to deter future criminal behavior, appeared successful when evaluating participants before and after the program. However, controlled studies revealed that participants actually had higher reoffending rates than similar youth who didn't participate. Without comparing outcomes to the counterfactual, supporters mistakenly believed they were helping when they were actually causing harm. Counterfactual thinking similarly explains why funding established charities in wealthy countries often achieves less impact than one might expect. Many such organizations would receive adequate funding regardless of your specific contribution, meaning your donation doesn't enable additional activities. Conversely, newer organizations addressing neglected problems may entirely depend on your support to operate, making your counterfactual impact much larger. This perspective extends to policy advocacy as well. Supporting popular policies that would likely pass anyway produces minimal counterfactual impact. However, advocating for important but overlooked policies might completely change outcomes. The same principle applies to research: working on already crowded questions typically contributes marginal insights, while exploring neglected questions can produce transformative discoveries. Importantly, counterfactual thinking doesn't diminish the value of traditional helping professions like medicine or teaching. These roles undoubtedly create substantial value. However, it encourages us to consider the difference we personally make by filling these roles, not just the inherent value of the positions themselves. This perspective helps identify truly high-impact career paths rather than simply socially respected ones.

Chapter 5: Expected Value: Balancing Probability and Magnitude in High-Impact Decisions

Expected value thinking—multiplying the probability of an outcome by its magnitude—provides a powerful framework for evaluating uncertain interventions. While we naturally focus on activities with predictable outcomes, this approach can lead us to overlook high-risk, high-reward opportunities that might ultimately do far more good. Consider political advocacy. The probability that your individual efforts will change a major policy might seem vanishingly small, perhaps one in a thousand. However, if that policy change would benefit millions of people, the expected value remains substantial. One person working on nuclear security policy for a year might have just a 0.01% chance of contributing to risk reduction, but given that nuclear war could cause billions of deaths, even this tiny probability translates to saving hundreds of lives in expectation. This principle explains why certain unpopular actions can be highly rational. Voting in elections appears futile from an individual perspective—no major election has been decided by a single vote. However, expected value calculations reveal that voting in close elections in swing states can have an expected impact equivalent to thousands of dollars in charitable donations. The extremely low probability of being decisive is balanced by the enormous consequences if you are. Research provides another example of effective expected value thinking. Most research projects fail to produce breakthrough findings, making any individual project seem like a poor investment. However, the few projects that succeed can transform entire fields and benefit millions. This explains why research funding generally delivers tremendous returns despite high failure rates—the successes more than compensate for the failures. Expected value thinking also illuminates the importance of working on catastrophic risks like pandemics, nuclear war, or artificial intelligence alignment. While these events may have low probabilities in any given year, their potential harm is so extreme that even small reductions in risk justify substantial resources. Importantly, expected value calculations must include long-term impacts, not just immediate consequences. This approach doesn't suggest recklessly pursuing any intervention with a non-zero chance of enormous benefit. Expected value calculations should incorporate all relevant information, including the track record of similar interventions and expert consensus. The goal is rational assessment of uncertain outcomes, not wishful thinking about unlikely scenarios. Embracing expected value thinking often requires overcoming psychological biases. Humans naturally struggle to reason about very small probabilities and very large impacts. We tend to either entirely dismiss low-probability events or obsess over them disproportionately. Systematic expected value calculations help counteract these biases, ensuring we neither ignore nor overemphasize uncertain but potentially transformative opportunities.

Chapter 6: Strategic Career Choice: How to Maximize Your Positive Impact Through Work

Your career represents one of your greatest opportunities to make a difference, comprising approximately 80,000 hours over your working life. However, traditional advice about impactful careers often focuses narrowly on directly helping professions like medicine, teaching, or nonprofit work without considering broader factors that determine your true impact. Strategic career planning begins by critically examining the "follow your passion" paradigm. Research shows that passion typically develops after achieving competence in meaningful work rather than preceding it. More reliable predictors of job satisfaction include autonomy, engagement with interesting problems, supportive colleagues, and visible impact. Building your career around these factors rather than pre-existing passions typically leads to both greater satisfaction and impact. Assessing impact potential requires considering three key dimensions. First, direct impact examines the good you accomplish through your daily work. Second, earnings potential considers your capacity to donate to effective organizations. Third, career capital evaluates how your role develops skills, credentials, and connections that increase your future impact. The highest-impact careers often excel in at least one of these dimensions, though rarely all three simultaneously. This framework reveals several promising career paths. Research careers can develop solutions to pressing problems, with particularly high potential in neglected fields addressing global challenges. Policy careers offer opportunities to improve government priorities and resource allocation, especially in areas like biosecurity, artificial intelligence governance, or economic development. Entrepreneurship enables creating organizations that directly address important problems, with social enterprises sometimes achieving both financial sustainability and significant impact. Earning to give—pursuing a high-earning career specifically to donate substantially—provides another powerful option. While medicine or nonprofit work might directly help dozens or hundreds of people, high earners who donate effectively can potentially save thousands of lives through their contributions. This approach proves especially valuable for those with comparative advantages in fields like finance, technology, or consulting but without specialized skills aligned with pressing global problems. Strategic career planning also requires appropriate time horizons. Early career stages should typically emphasize building versatile skills and exploring different paths rather than maximizing immediate impact. Mid-career professionals often face different considerations, leveraging their established expertise while remaining open to pivotal role changes when greater impact opportunities arise. Importantly, career decisions should account for personal fit. Your impact in any role depends significantly on your performance, which correlates with your abilities, interests, and working style. Someone who excels in quantitative research but struggles with interpersonal dynamics might achieve far more impact as a researcher than as a manager, regardless of which role seems more directly impactful in theory.

Chapter 7: Cause Prioritization: A Framework for Selecting the Most Important Problems

Not all problems are equally important, tractable, or neglected, making cause prioritization—the systematic comparison of different issue areas—essential for maximizing impact. Rather than simply following personal passion or responding to emotional appeals, effective altruists use structured frameworks to identify where additional resources can accomplish the most good. The most widely used prioritization framework evaluates causes along three dimensions. Scale measures the magnitude of the problem—how many individuals are affected and how severely. Tractability assesses how easily progress can be made and measured. Neglectedness considers how many resources are already dedicated to addressing the problem. The most promising causes typically score highly across all three dimensions, representing large problems where additional resources can make substantial progress because existing efforts remain insufficient. Global health and development consistently emerges as a high-priority cause area through this analysis. The scale is enormous, with billions living in extreme poverty. Tractability is high, with well-established interventions like antimalarial bed nets, deworming, and direct cash transfers demonstrating robust evidence of effectiveness. Despite substantial existing resources, the area remains relatively neglected compared to domestic social spending in wealthy countries, with highly effective interventions still lacking adequate funding. Animal welfare presents another compelling priority. The scale is vast, with tens of billions of animals suffering in industrial agriculture annually. The area remains highly neglected, receiving minimal philanthropic attention relative to its scale. While tractability varies across interventions, corporate campaigns to improve welfare standards and research into plant-based alternatives have demonstrated significant success with limited resources. Long-term future causes increasingly attract attention within effective altruism. While conventional prioritization often focuses on immediate suffering, most potential human flourishing lies in the future if humanity continues to exist and thrive for thousands or millions of years. This perspective highlights the importance of reducing existential risks—threats that could permanently curtail humanity's potential. Climate change, nuclear war, engineered pandemics, and artificial intelligence alignment all represent potential existential risks that remain relatively neglected compared to their potential impact. Cause prioritization necessarily involves value judgments alongside empirical assessments. Different moral frameworks may place different weights on helping humans versus animals, present versus future generations, or preventing suffering versus promoting flourishing. Rather than prescribing a single correct answer, effective cause prioritization makes these value judgments explicit and identifies which causes most effectively advance different moral priorities. The prioritization process should remain dynamic rather than static. As additional resources flow to previously neglected areas, their neglectedness decreases. As research advances, our understanding of tractability evolves. And as our moral circle expands, our assessment of scale may change to encompass more beings. Regular reassessment ensures resources continue flowing to where they can accomplish the most good.

Summary

Effective altruism fundamentally transforms how we approach doing good in the world by replacing intuition and emotional appeal with evidence and rigorous analysis. By systematically evaluating interventions based on their actual outcomes rather than their intentions, we discover that some approaches to helping others achieve hundreds or thousands of times more impact than others. This insight carries profound implications: even modest resources, when directed toward the most effective solutions, can save numerous lives or prevent immense suffering. The framework presented throughout this analysis offers a powerful methodology applicable across multiple domains—from charitable giving and career choice to policy advocacy and research prioritization. By consistently asking which problems are most important, which solutions are most effective, and where additional resources can make the greatest difference, we can dramatically increase our positive impact. Rather than settling for making some difference, effective altruism challenges us to make the most difference possible with the resources available to us, transforming good intentions into extraordinary outcomes through the thoughtful application of both compassion and reason.

About Author

Loading...

William MacAskill

I'm Will MacAskill, an Associate Professor in Philosophy at Lincoln College, Oxford, and author of Doing Good Better (Gotham Books, 2015). I've also cofounded two non-profits: 80,000 Hours, which provides research and advice on how you can best make a difference through your career, and Giving What We Can, which encourages people to commit to give at least 10% of their income to the most effective charities. These organisations helped to spark the effective altruism movement.

Read more

Related Books by William MacAskill

Loading...

What We Owe the Future

William MacAskill

A Guide to Ethical Living for the Fate of Our Future

★★★½☆3.8