Key Takeaways
Test a helping idea before scaling it; good intentions can burden recipients.
The PlayPump looked perfect. Children could spin a playground roundabout to pump water, an appealing fix for villages where water collection was arduous. But assessments found that the pump produced less water than a standard hand pump, children tired quickly, repairs were difficult, and women often had to operate it themselves. At about $14,000 per unit, it cost four times as much as the alternative.
Deworming looked less exciting. Economist Michael Kremer tested education programs in Kenyan schools and found that textbooks and smaller classes did little to improve results. Treating children for intestinal worms, however, reduced school absence by 25 percent. The contrast is a warning: a solution’s appeal is not evidence that it works.
What's striking is the distance between designing for a compelling story and designing for the people who must live with the result. Human-centered design asks whose experience defines the problem, while controlled trials ask whether the proposed fix changes outcomes. Both checks were missing from the PlayPump rollout. Trials are not a universal answer: they can miss local preferences, long-term maintenance, or harms that are difficult to measure. But pairing evidence with community feedback makes it harder for a funder's imagination to substitute for a beneficiary's needs.
Treat affluence as leverage: a modest gift can matter far more abroad.
Global wealth changes the comparison. In the book's 2014 figures, a person earning more than $52,000 annually was in the world's richest 1 percent, while 1.22 billion people lived on less than the purchasing power of $1.50 per day. That poverty measure already accounts for local prices and goods people grow for themselves.
Money has unequal value. The author calls the estimated difference in the benefit of a dollar to an affluent giver and an extremely poor recipient the 100x Multiplier. The reasoning is that doubling income appears to improve reported well-being by a similar amount at different income levels. Since affluent people earn many times more, a dollar can mean much more to someone in extreme poverty. This is a rough comparison, not a universal exchange rate for moral value.
The central insight is about marginal value: an extra dollar usually matters more when basic needs remain unmet. That idea connects economic inequality to the diminishing usefulness of additional consumption at higher incomes. The multiplier is best read as a prompt to examine privilege, not as a precise rule for every donation. Well-being depends on more than income, and large-scale transfers can affect prices, institutions, and other people beyond the direct recipient. The argument is strongest as a reason to investigate high-impact giving, rather than as proof that every overseas intervention is superior.
Measure the lives improved, not the activity counts that flatter charities.
Outputs are not outcomes. Counting books delivered, people reached, or treatments distributed says what a program did, not whether anyone became better off. A charity might report how much it costs to provide books, for instance, while offering little evidence that the books improve learning.
Health measures can help compare unlike benefits. A quality-adjusted life year, or QALY, represents one year of life in perfect health, combining lifespan and health quality. In the book's hypothetical comparison, AIDS treatment produced 6.5 QALYs, while surgery preventing blindness produced 30. Such estimates help clarify choices when resources are limited. The book also describes a broader well-being-adjusted measure, which would compare changes in life satisfaction across different kinds of support.
Measurement disciplines attention by making trade-offs visible. It also creates a risk: a clean metric can make contested moral choices look like settled arithmetic. QALYs rely on judgments about how different conditions affect life, and people living with a condition may assess it differently from outsiders. Quantification should therefore expose assumptions, not hide them. A useful measure is a map of consequences, not the territory itself. The alternative, however, is not neutral: choosing by vividness or familiarity also embeds values, only less transparently.
Search for exceptional programs, because effectiveness is wildly uneven.
The best options can dominate. A fat-tailed distribution is one in which a small share of cases accounts for a large share of the total value. The book argues that aid effectiveness often has this shape: many programs help a little, while a few help enormously. Its historical estimates put school uniforms at 7.1 additional years of attendance per $1,000, compared with 139 years for deworming and 0.2 years for incentives to keep girls in school.
Good is not automatically best. The book also compares health interventions, estimating that bed-net distribution could deliver hundreds of times the benefit of treating a less cost-effective condition. These figures depend on specific studies and contexts. The practical lesson is to compare plausible alternatives, while treating estimates as evidence to examine rather than promises.
Fat-tailed outcomes explain why averages can mislead donors. If a few highly effective programs account for much of the benefit, funding the typical intervention may leave substantial impact on the table. The same pattern appears in fields where a small number of discoveries or products reshape outcomes. But searching for winners can encourage overconfidence: estimates of unusually high impact may be precisely the ones most vulnerable to optimistic assumptions. Comparing alternatives matters, and so does asking how sturdy the evidence is behind the apparent standout.
Aim resources at neglected needs, where extra help can still change outcomes.
Attention is not allocation. After the 2011 Japanese earthquake and the 2010 Haitian earthquake, each received roughly $5 billion in international aid, although the book reports far more deaths in Haiti and much greater capacity for Japan to respond. The 2008 Sichuan earthquake killed about 87,000 people but attracted only about $500 million in international aid. Public attention did not track either need or local resources.
Ongoing emergencies compete too. Preventable disease and poverty claim lives every day, yet rarely prompt the same surge of giving as a dramatic disaster. The economic idea of diminishing returns explains why: if an area already attracts abundant resources, an additional donation may achieve less than support directed to a neglected problem. Neglect alone is not proof of promise, but it is a reason to investigate.
The contrast between headline disasters and chronic suffering reveals how attention can become a poor proxy for need. Media coverage and vivid stories help people notice distant crises, but they do not reliably show where money will do most good. This is a classic allocation problem: resources follow what is salient, not necessarily what has the greatest unmet need. Still, neglected causes may be neglected because they are hard to solve. A donor needs to ask both whether others are overlooking the problem and whether there is a credible way to improve it.
Count only the difference your action causes compared with the alternative.
Good outcomes need a baseline. A counterfactual is an estimate of what would have happened if an action had not occurred. The book contrasts D. A. Henderson, who led the successful smallpox eradication campaign, with Viktor Zhdanov, whose earlier advocacy helped persuade the World Health Organization to begin it. Henderson was crucial to execution, but the author argues that someone else might have filled his role. If Zhdanov accelerated eradication by a decade, the delay avoided could have prevented 10 to 20 million deaths.
A program can look successful and cause harm. Studies of Scared Straight, which took young offenders into prisons to deter crime, found that participants were more likely to offend than comparable youths who did not attend. Before-and-after improvement can mislead when behavior might have changed anyway.
Counterfactual thinking is the backbone of causal reasoning: the relevant question is not whether something good happened, but whether the intervention made it more likely. It guards against credit-taking, and against scaling programs that merely coincide with improvement. The hard part is that the alternative world cannot be observed directly; researchers approximate it with comparison groups, models, or historical evidence. Those methods can be imperfect, but ignoring the baseline is worse. Without it, dramatic stories can overwhelm evidence about what actually caused change.
Give long shots a fair hearing when small odds meet enormous stakes.
Expected value combines odds and impact. It is the average outcome implied by multiplying each possible result by its probability. A small chance of a very large benefit can outweigh a more certain, modest gain. The book uses voting as an example: a 2008 US election estimate put the average chance of one vote deciding the result at about one in 60 million. With a hypothetical benefit of $1,000 per American from the better result, the expected collective benefit was about $5,200. The calculation depends heavily on those assumptions.
Risk can run both ways. Fukushima's safety planning treated an extremely unlikely catastrophe as practically impossible, overlooking the scale of the harm if it occurred. Effective altruism applies the same reasoning to high-impact political change and catastrophic risks.
Expected value helps correct a common intuition error: people often treat low probability as if it meant no probability. It also offers a shared language for comparing reliable interventions with political or scientific work whose upside is much larger but uncertain. The arithmetic is only as good as its inputs. A voter’s estimate of which party is better, for example, can dominate the result, and distant risks are difficult to quantify. The framework is most useful when it makes uncertainty explicit, not when a precise-looking number disguises disagreement.
Judge charities by outcomes, evidence, execution, and unmet funding needs.
Low overhead does not mean high impact. Books For Africa had very low administrative costs, but research cited in the book found that simply providing textbooks often did little to improve learning. GiveDirectly transferred most donated money directly to poor households and had strong evidence from independent evaluation. Development Media International's radio health campaigns had a potentially greater estimated impact, but the evidence behind that estimate was less robust.
A sound review asks more than one question. Identify the actual program, estimate its cost-effectiveness, inspect the quality of evidence, assess whether the organization can deliver well, and check whether it can use additional funds. These comparisons came from a 2015 snapshot, not current recommendations. The underlying discipline remains useful: choose based on what donations change, not a financial ratio alone.
The charity comparison exposes a principal-agent problem: donors hand over resources but often cannot see whether the organization uses them well. Overhead ratios seem reassuring because they are easy to compare, yet they tell little about the value created. Evidence, implementation, and funding capacity together provide a more meaningful picture. There is still a trade-off between confidence and potential impact: a well-tested program may offer a smaller estimated benefit than a promising but less-tested one. Good evaluation makes that uncertainty legible instead of pretending it disappears.
Choose consumer actions carefully; targeted giving often buys more leverage.
A boycott can hurt the workers it aims to help. In poor countries, factory jobs may be safer and better paid than a worker's alternatives. When Bangladesh factories dismissed about 50,000 children amid pressure over child labor, some reportedly turned to more dangerous ways of earning a living. That does not excuse abusive conditions; it questions whether withdrawing jobs improves them.
Some purchases do little for the intended recipient. The book describes evidence that only a small fraction of the extra price of fair-trade coffee reaches producers. For climate, it argues that travel, home energy, and meat choices matter more than small gestures such as avoiding plastic bags. For animal welfare, it highlights reducing chicken, eggs, and pork. Its broader point is to compare these actions with targeted donations.
Consumer choices can matter through market demand, but their effects are often indirect and hard to track. A purchase may signal a preference, change a supplier's decisions, or simply shift costs to workers with few alternatives. That makes the counterfactual especially important. It also cautions against treating ethical labels as evidence of ethical outcomes. The chapter's strongest contribution is not that consumption is irrelevant, but that people should distinguish high-impact choices from gestures that mainly provide moral reassurance. Coordinated consumer action can still be powerful when it changes production practices rather than merely withdrawing support.
Build a career around fit and impact, not a romanticized passion.
Job satisfaction is not a treasure hunt. The book argues that independence, varied work, visible completion, useful feedback, and contribution to others are better guides to satisfying work than searching for a pre-existing passion. Interests change, and people are poor at predicting what will make them happy. Trying jobs and learning from experience can reveal personal fit more reliably.
Career impact has several routes. A person can contribute labor, earn money to donate, or influence others. Greg Lewis estimated that an additional doctor in the UK might save only one or two more lives over a career once replacement effects were considered, while earning to give could fund more direct aid. Early choices can also build career capital: useful skills, connections, and credentials that expand future options.
Career choice combines identity with resource allocation, so slogans about passion can feel more certain than the evidence warrants. The book's emphasis on fit and experimentation resembles learning-by-doing: short tests can reveal whether a field suits someone's abilities and temperament. Its counterfactual approach also challenges the assumption that visible service always creates the most additional value. Still, high earnings are not automatically a route to greater impact. Work culture, motivation, opportunity costs, and the effectiveness of eventual donations all matter. A career strategy should be revisable, not treated as a permanent moral label.
Choose causes by scale, neglect, tractability, and your particular strengths.
A large problem may be hard to solve. The book compares causes using scale, or how much harm exists; neglectedness, or how few resources address it; and tractability, or how readily effort can produce progress. For time and career choices, personal fit matters too. Global poverty scores well in the book's assessment because the need is great and proven health interventions can help at low cost. International labor mobility promises enormous gains but faces political resistance. Criminal justice reform appears more tractable, while factory farming is highly neglected.
Values affect the ranking. The scale of animal suffering or catastrophic risks depends partly on how much moral weight is assigned to animals or future generations. The framework organizes inquiry, but it cannot settle every disagreement about what matters most.
The framework is valuable because it separates three questions that are often collapsed into one: Is the problem serious, is it overlooked, and can action help? This helps explain why the loudest or largest cause is not automatically the best target. It also makes personal competence part of the analysis rather than an afterthought. The difficult judgment lies in weighing dimensions that do not share a common scale. A transparent disagreement about values is more productive than pretending that a scorecard can make the moral choices disappear.
Turn concern into a recurring commitment, not a one-off emotional response.
Make the next step concrete. The book closes by recommending a regular donation to a well-evaluated charity, even if the initial amount is small. It also suggests writing down a specific plan for giving, career exploration, or changes in purchasing, with dates and actions rather than general intentions. These steps make altruism part of ordinary life instead of leaving it dependent on the next crisis or appeal.
Build social support. Joining a community, sharing useful ideas, and inviting others to give can reinforce the commitment. The author presents effective altruism not as a demand to sacrifice everything, but as a continuing effort to improve the results of whatever time and money a person chooses to devote to others.
A recurring commitment converts moral intention into a practice that can survive shifting attention. This connects the book's analytical emphasis with behavioral design: regularity reduces the need to reconsider the same decision from scratch. Social accountability can reinforce follow-through, provided it does not turn giving into status competition. The caution is that sustainable action matters more than a dramatic pledge that leads to burnout or resentment. A modest, durable commitment can be both ethically serious and psychologically realistic.
Analysis
Doing Good Better treats altruism as a problem of both moral attention and practical inference. Its central move is to ask not merely whether an act is kind, but what changes because of it, compared with the available alternatives. The PlayPump and deworming stories make that distinction vivid: an intervention can be emotionally compelling, publicly celebrated, and still be inferior to an unglamorous alternative tested against real outcomes.
The book's strongest contribution is methodological. It brings marginal analysis, causal evaluation, and expected value into ordinary decisions about donations, careers, consumption, and political action. This widens the idea of helping beyond charity work and challenges the assumption that proximity to beneficiaries, low overhead, or personal passion reliably signals impact. It also preserves a role for uncertainty: a speculative political intervention may deserve attention when its possible benefits are vast, even if success is unlikely.
The framework's limits are part of its value. QALYs and cost-effectiveness estimates make trade-offs visible, but they depend on contestable judgments about health, fairness, and which interests count. A measurable intervention can look superior partly because its effects are easier to count. Once a proxy becomes a target, organizations may optimize the proxy rather than the underlying good, a familiar problem in measurement-driven systems. The book's own emphasis on evidence quality and counterfactuals helps guard against this, but cannot eliminate it.
Effective altruism is therefore most persuasive not as a machine that produces one unquestionable ranking, but as a discipline for making hidden assumptions explicit. Its questions encourage readers to compare alternatives, seek neglected opportunities, and revise beliefs when evidence disappoints. That combination of ambition and self-correction is the book's defining ethic: care should motivate action, while reason should help determine whether the action actually serves its intended purpose.
Review Summary
Reviews of Doing Good Better are generally positive, praising its thought-provoking ideas on effective altruism and evidence-based charity. Many readers found it eye-opening and life-changing, appreciating MacAskill's logical approach to maximizing charitable impact. Some critics argue the book oversimplifies complex issues or relies too heavily on quantitative measures. Despite occasional controversy, most readers recommend it as an important read for anyone interested in making a meaningful difference in the world through donations or career choices.
People Also Read
Glossary
Effective altruism
evidence-guided helpingA way of thinking about helping others that combines concern for their well-being with evidence and careful reasoning. It asks how to use available time, money, and influence to produce the greatest positive difference, while remaining willing to revise beliefs when evidence shows that an appealing intervention is ineffective.
100x Multiplier
unequal value of moneyThe author's label for the estimated difference between the benefit an additional dollar provides to an affluent person and the benefit it provides to someone in extreme poverty. The comparison draws on income and reported well-being research. It is a rough illustration of unequal marginal value, not a precise rule for ranking every transfer or intervention.
QALY
healthy life-year measureA quality-adjusted life year is a measure combining length of life with health-related quality of life. One QALY represents one year lived in perfect health, or an equivalent combination of time and health improvement. The metric helps compare health interventions, although its quality weights depend on judgments about how different conditions affect people's lives.
Well-Being-Adjusted Life Years
well-being over timeA proposed extension of health-adjusted life measures that compares changes in overall well-being rather than health alone. It treats well-being as varying over a person's life and combines the size of an improvement with its duration. The book uses it to illustrate how benefits such as income gains might, in principle, be compared with health improvements.
fat-tailed distributions
value concentrated in outliersDistributions in which a relatively small number of cases account for a large share of the total. The book applies this idea to charitable programs, arguing that a few interventions may produce far more benefit than typical ones. It explains why choosing the strongest opportunities can matter more than funding an average program.
counterfactual
what would otherwise happenThe alternative outcome that would occur if a particular action were not taken. Assessing impact requires comparing what happened with what would likely have happened without the intervention. This distinguishes additional benefit from good outcomes that were already likely, and can reveal when a program produces harm despite appearing successful.
expected value
probability-weighted outcomeA decision method that multiplies each possible outcome by its probability and combines the results. It allows uncertain actions to be compared with more reliable ones by accounting for both the chance of success and the size of the benefit or harm. It is especially relevant when small probabilities accompany very large consequences.
micromort
one-in-a-million death riskA unit of risk equal to a one-in-a-million chance of death. The book uses micromorts to compare risks from activities such as travel, smoking, or recreation, including risks that may take years to affect life expectancy. Converting risks into a shared unit can make their relative size easier to understand.
earning to give
earn more to donateA career strategy in which a person prioritizes earning income so a substantial portion can support effective causes. Its impact depends on the person's earning potential, the effectiveness of the organizations receiving donations, and what would have happened if the person had chosen another career. It is one possible route to impact, not a universal recommendation.
career capital
skills, networks, credentialsThe skills, professional connections, and credentials accumulated through work or study that can expand future career options and influence. A job with limited immediate social impact may still be valuable if it builds abilities that enable greater contribution later. The concept supports treating career plans as revisable investments rather than fixed callings.
three-factor model
cause-prioritization frameworkA framework for comparing causes by their scale, neglectedness, and tractability. Scale concerns the magnitude of the problem, neglectedness asks how much attention and funding it already receives, and tractability concerns whether additional effort can make progress. For career or volunteer choices, the book adds personal fit as another relevant consideration.
moral licensing
good deed, weaker restraintA psychological pattern in which people who have acted, or even imagined acting, morally may feel more permission to behave selfishly or unethically afterward. The book raises it as a possible downside of promoting low-impact ethical actions if those actions make people feel they have already done enough to help.
lower-bound reasoning
conservative impact estimateA method for assessing uncertain opportunities by constructing an estimate believed to be lower than the true possible impact. The book applies it to political careers: if a conservative estimate still compares favorably with a more reliable alternative, the uncertain path may merit serious consideration. The method depends on making cautious assumptions explicit.
FAQ
What's Doing Good Better about?
- Focus on Effective Altruism: The book by William MacAskill explores effective altruism, a philosophy that combines data and reason with altruistic acts to maximize positive impact.
- Five Key Questions: It introduces five essential questions to guide effective altruism, focusing on impact, effectiveness, neglected areas, alternative outcomes, and success chances.
- Real-World Examples: MacAskill uses case studies like PlayPump and cash transfer programs to illustrate effective altruism principles and potential pitfalls of ineffective charitable actions.
Why should I read Doing Good Better?
- Transformative Perspective: The book challenges common assumptions about altruism, encouraging a more analytical approach to making a difference.
- Practical Framework: It provides a framework for evaluating charities and making decisions about resource allocation for maximum impact.
- Inspiring Call to Action: MacAskill motivates readers to consider their potential to effect change, emphasizing that even small contributions can lead to significant benefits.
What are the key takeaways of Doing Good Better?
- Effective Altruism Defined: The book defines effective altruism as using evidence and reason to determine the most effective ways to help others.
- Importance of Evidence: MacAskill stresses the need for robust evidence when evaluating charities, arguing that many initiatives may not be as effective as they seem.
- Focus on Cost-Effectiveness: The author highlights the significance of cost-effectiveness in charitable giving, urging readers to consider the impact relative to costs.
What is effective altruism as defined in Doing Good Better?
- Philosophical Approach: Effective altruism encourages using resources—time, money, and skills—to do the most good possible, combining altruistic intentions with rigorous analysis.
- Evidence and Reason: It emphasizes using evidence and reason to determine the most effective ways to help others, rather than relying on emotional appeals.
- Long-Term Impact: The focus is on long-term solutions to global problems, aiming for sustainable change rather than temporary relief.
How does Doing Good Better suggest evaluating charities?
- Five Key Questions: MacAskill introduces questions to assess charities, focusing on impact, effectiveness, and neglected areas.
- Cost-Effectiveness Analysis: The book emphasizes evaluating how much good a charity can achieve per dollar spent.
- Transparency and Evidence: It advocates for transparency in charity operations and the use of robust evidence to support claims about effectiveness.
What are the five key questions of effective altruism in Doing Good Better?
- Impact Assessment: How many people benefit, and by how much? This encourages a concrete assessment of the impact of charitable actions.
- Effectiveness Evaluation: Is this the most effective thing you can do? It pushes individuals to evaluate the best use of their resources.
- Neglected Areas: Is this area neglected? It highlights the importance of focusing on causes that receive less attention and funding.
How does Doing Good Better illustrate the pitfalls of charitable giving?
- Case Study of PlayPump: MacAskill discusses the PlayPump initiative, which failed due to practical shortcomings, illustrating the need for evidence-backed solutions.
- Community Needs: The book emphasizes understanding community needs to avoid wasted resources and unmet needs.
- Evaluating Outcomes: It stresses the need for rigorous evaluation of charitable programs to ensure they deliver intended benefits.
What is the 100x Multiplier mentioned in Doing Good Better?
- Concept Explanation: The 100x Multiplier suggests that individuals in affluent countries can do at least one hundred times more good for others than for themselves.
- Impact of Donations: It illustrates how a dollar donated to someone in extreme poverty can have a much greater impact than if spent on oneself.
- Encouragement to Act: Understanding this multiplier motivates readers to leverage their resources to help those in need.
How does Doing Good Better address ethical consumerism?
- Critique of Effectiveness: The book argues that ethical consumerism often has minimal impact compared to targeted charitable donations.
- Moral Licensing Effect: MacAskill discusses how ethical consumerism can lead to moral licensing, potentially undermining overall impact.
- Focus on Donations: It advocates for prioritizing donations to effective charities over small changes in consumer behavior.
What are some examples of effective charities mentioned in Doing Good Better?
- Against Malaria Foundation: Highlighted for its effectiveness in distributing bed nets to prevent malaria at a low cost per QALY.
- GiveDirectly: Provides cash transfers to the poor, allowing recipients to use funds as they see fit, with significant positive impacts.
- Deworm the World Initiative: Focuses on deworming programs that improve health and educational outcomes for children, demonstrating high cost-effectiveness.
What are the best quotes from Doing Good Better and what do they mean?
- “Good intentions can all too easily lead to bad outcomes.”: Emphasizes that wanting to help is not enough; effectiveness must be considered.
- “Effective altruism consists of the honest and impartial attempt to work out what’s best for the world.”: Highlights the core principle of using rational thinking and evidence in altruistic efforts.
- “The best ways of doing good are very good indeed.”: Reminds readers that highly effective methods of helping others can lead to substantial positive change.
How can I apply the principles of Doing Good Better in my own life?
- Ask the Five Key Questions: Apply these questions to your charitable actions to make informed decisions about donations.
- Consider Earning to Give: Maximize impact by pursuing high-earning careers and donating a portion of income to effective charities.
- Stay Informed and Flexible: Remain open to new evidence and adjust charitable strategies based on effectiveness and impact.
Download PDF
Download EPUB
.epub digital book format is ideal for reading ebooks on phones, tablets, and e-readers.