Margen
You need a stranger to lend you a phone for two minutes, and before you open your mouth you run the sums in your head: not this one, not the one beside him either, maybe the fourth. That sum is wrong, and it is wrong by almost double: to get five favours out of five strangers you think you will have to approach twenty people, and ten will be enough.
Both figures come from a field experiment with a clipboard and a biro, and it is worth knowing what each one measures before trusting them.
Each participant was given a pad of questionnaires and a single sentence to say: "Will you fill out a questionnaire?" Three rules. Strangers only. No explanations if they asked for any. And write down on a sheet the yes or the no of every person approached. The task did not end until five people had filled out the questionnaire.
Half of them, before going out, were asked one more question: how many people do you think you will have to approach to get those five? The average of their answers was 20.5 people.
The real average was 10.5.
The other half of the group went out without making any prediction, in case the act of predicting changed the way they asked. They needed 11.2 people, practically the same. Predicting spoiled nothing; it was simply twice as pessimistic as reality.
It is signed by Francis Flynn, then a professor at Columbia Business School and today at Stanford's, and Vanessa Lake, of the same university's psychology department, which is where the fieldwork was done, on the New York campus. It was published in 2008 in the Journal of Personality and Social Psychology, the leading journal in social psychology.
The explanation for that double is in the fourth and fifth experiments of the series, and it is the part that actually helps you. The authors put some people in the role of asking and others in the role of being asked, and questioned them about the costs.
The person asking works out the cost of saying yes, and works it out well: the time you take from somebody, the bother, a stranger's phone in another stranger's hands. What they leave out of the sum is the cost felt by the person being approached, which is the cost of saying no. Refusing somebody who is half a metre away, looking at you, who has asked a short question with no catch, is genuinely uncomfortable, and that discomfort pushes towards yes every one of the times you have not counted.
With that asymmetry in hand, the study's other favours read themselves. To get three people to lend them a phone, participants reckoned on 10.1 approaches and needed 6.2. To get a single person to walk them to the gym, they reckoned on 7.2 and two point three were enough. Of the twenty-nine who completed that second experiment, twenty-three erred on the same side: they reckoned too high.
These are small samples — forty-two participants in the first, twenty-three of them making a prediction, and twenty-nine in the second — on a university campus in New York and with minuscule favours. Nobody has shown from this that you will get a raise if you ask for one, or that a stranger will hand you their car keys. What has been measured is narrower than that, and it is free: asking in person for something small and specific works out sooner than you calculate.
There is one more finding from the same line of work. Vanessa Lake, who has signed for years now as Vanessa Bohns, later published with Mahdi Roghanizad an article whose title is already half a piece of advice: "Ask in person: You're less persuasive than you think over email." The channel changes the result. Email spares you the bad moment of asking and, in exchange, takes away your advantage, because the person reading has nobody in front of them to say no to.
So the next time you need something small, do it the way the experiment did: one sentence, to their face, and the tally on your phone, writing down the noes before the first yes. That list comes out at half of what you are carrying in your head right now.