I’m sure you’ve all received plenty emails that have been sent to multiple recipients, where the gist of the message is a request:
“Will one of you please do [something useful]?”
These messages are usually an exercise in futility — nobody bothers to handle the request, because we each say to ourselves “someone else will handle it”.
In an attempt to understand the reasons behind this, I did some digging. I had never realized this topic could be so fascinating.
No, I have not mistakenly pasted in the wrong text for the title. It’s from a story in the New York Times (March 27, 1964), which describes a shocking phenomenon. I’ll show you how it connects with our topic.
On March 13, 1964 at about 3:20 am, 28 year old Catherine “Kitty” Genovese parked her car outside her apartment building at 8720 Austin St., Queens, New York City. Her car was only 100 feet from the entrance to the building, but she was followed and brutally stabbed multiple times by an attacker before she reached it. Kitty’s screams awoke many of the neighbors, some of whom shouted at the attacker, who ran off. As soon as the neighbors has turned their lights back off, Kitty’s assailant returned until her screams triggered more opened windows and lights. After driving away, the attacker returned yet again and finished her off.
The police were called by a neighbor at 3:50 am, half an hour after Kitty’s screams had first woken the neighbors. They arrived within two minutes. The caller admitted that he’d phoned a friend for advice before finally calling the police.
The police detectives discovered that no less than thirty-eight people had witnessed various parts of the crime, yet not one of them called the police before it was too late. These are some of the excuses they made:
“We thought it was a lover’s quarrel!”
“Frankly, we were afraid.”
“I didn’t want my husband to get involved.”
“We went to the window to see what was happening, but the light from our bedroom made it difficult to see the street.”
“I was tired.”
“I tried … I really tried, but I was gasping for air and was unable to talk into the telephone.”
I am sure that if any of these witnesses had thought that they were the only person who could help, they would have each called the police in time to save Kitty Genovese.
Sociologists, economists and psychologists have all researched this type of behavior.
- Social cueing theory predicts that people are less likely to act when they believe that others are capable of acting.
- Economists call this the volunteer dilemma. The larger a group of people is, the less likely a specific member is to volunteer to perform a service that benefits the group.
- Psychologists call this phenomenon diffusion of responsibility.
It makes sense. People generally act to further their own interests. In a large group, assuming that a volunteer will actually come forward, the convenience of not volunteering outweighs any advantage of volunteering. In small groups, the assumption that someone else will come forward is a bit shaky, especially when peer pressure is brought to bear. Self-interest is often well-aligned with the interests of a small group, so each individual is much more likely to volunteer.
How does this apply to email? Greg Barron and Eldad Yechiam of the Technion technology institute in Haifa, Israel performed the following experiment:
- They sent emails to 240 students, researchers and administrative staff, from a fictitious student called Sarah Feldman.
- The emails asked whether the Technion had a biology department.
- Messages were either addressed to a single recipient or addressed to five recipients.
Apart from the respondents who tried to chat Sarah up, the responses were divided into three categories:
- Unhelpful: “Find the web page and look it up yourself!”
- Helpful: “Yes, there is a biology department.”
- Very Helpful: “Yes, there is a biology department, and here are the phone numbers …”
I was surprised to read that of the messages sent to five people, 50% of the recipients actually responded. (I like to attribute this high figure to the “damsel in distress” effect on the Technion’s predominantly male student body. In fact, other studies have shown that women do indeed elicit higher response rates than men.) However, only 16% of these responses were considered to be Very Helpful.
The response rate of the people who received individually addressed messages was 64%, and 33% of these were classed as Very Helpful.
The study clearly shows that individually addressed messages elicit both a signifcantly higher response rate and responses of a significantly better quality.
Every time I see a poorly written email that ignores this phenomenon, I feel sorry for the author who is wasting his time. He’s not wasting my time, though — having set up my Outlook rules to color-code my messages, messages where I’m the only recipient on the To: line get a special color. These are the ones I read first, because I know that if I don’t handle them, nobody else will.
It’s very easy to use this behavioral phenomenon to get people to respond. If you require action, you must explicitly ask for it, and in order to avoid diffusion of responsibility, you must assign it to a specific person:
- The person whose responsibility it is to act, should be the only person on the To: line.
- If it is not clear whose natural responsibility it is, arbitrarily nominate one person to be responsible, and address it to this person, as mentioned in the previous point.
- If different actions are required of multiple people, consider using separate messages. If it still makes more sense to use a single message (e.g., to show everyone how the individual assignments fit into a wider context), explicitly list the assignees’ names alongside their corresponsing tasks, and do this towards the top of the message.
If you are too lazy to personalize the messages you send, how lazy do you think the recipients will be?