Would you rather help one person or 200 people, if it took the same effort? If you do what most people do, you’ll be lucky if you help even one.

Let’s say you recognize that giving to charities can make a profound impact in others’ lives and perhaps you even believe it’s morally the right thing to do. Perhaps you once met someone who was blind and now you are drawn to helping the blind. You’ve made the choice of a cause, but there are hundreds of organizations that help the blind and thus seem deserving of your money.

Some think it inappropriate to compare charities and especially to criticize a decision about which charity to give money to. But it is not just appropriate, it is supremely important. Your decision where to give will determine whether 200 or just one person gets helped.

You find a US charity that trains guide dogs for the blind. It’s a commendably austere charity; over 99% of its received donations go directly to its goal: training the dogs. So you make your donation of US$1,000 and are confident your money will be used well. Months later you receive a photo of the dog with its new owner and you feel great, knowing you have helped provide a dog that will help its owner ‘see’ for the next 10 years.

No one will say you’ve done a bad thing, and you haven’t. But you could have done something better. You probably would have liked to help several people as much as you’ve helped the lucky recipient of your guide dog – you just didn’t have much money to give. It turns out, you could have done far better; you could have cured about 20 cases of blindness instead! Let’s see how.

Another organization, which you overlooked because it uses only 70% of its donations on its goal, works in the developing world. Because of the low salary of third-world doctors and other factors, the cataract surgery this organization focuses on costs about $35.[1]

Let’s compare the two charities directly: via the first it costs $10,000 to train a guide dog and thus costs about $1,000/year to provide a proxy for eyesight for a blind person. Via the second it costs about $35/person and, even if they don’t live long, it costs no more than $5/year to restore sight to a blind person. It seems reasonable to say you’ve done at least 200 times worse than you could have – if your goal was to help as many people as you could. In this example, you’ve given 1 year of proxy for eyesight, rather than 200 years of actual eyesight.

You may be skeptical of the numbers I’ve used, but I made them favorable to the first cause: one eye-seeing dog charity claims it costs ‘$42,000 or more’ to train the dog and provide instructions to the user.[2] a cataract charity claims to be able to perform the procedure for $25. even if you’re skeptical that the costs may be higher in practice, they are unlikely to exceed $100/person.

I think it an important lesson to learn: what matters most is the cost-effectiveness of the intervention, not the overhead costs of the organization. Choosing where to give may be more important than whether to give at all. If your goal is to help as many people as possible, you’ll do more good by comparing charities by the effects of every dollar you donate to them.

Choosing among the causes is important as well. For more information on that, read The Best Causes.

[1] To avoid singling out a specific charity for attack or acclaim, these numbers are approximate to the average of existing charities that focus on the interventions.

[2] The figures used in this post are meant to make the dog-training charity look as favorable as possible. One US charity claims it costs ‘approximately $42,000 or more’ per dog (guidedogsofamerica.org) and their non-program expenses are not 1% as in the example but about 30% (charitynavigator.org).