Email and Organisational Effectiveness
By Dr. Thomas Jackson
Employees at the Danwood Group in the UK were monitored to see how they used email. The results from this research have shown that email messages do have some disruptive effect by interrupting the user. It was found that most employees had their email software check for incoming messages every 5 minutes and responded to the arrival of a message within 6 seconds. A recovery time between finishing reading the email and returning to normal work also existed, though it was shorter than published recovery times for a telephone interrupt.
Cutter Business Technology Council Fellows, Tom DeMarco and Tim Lister in their book, “Peopleware: Productive Projects and Teams,” have described the high impact of phone calls in engineering environments: developers routinely receive 15 telephone calls a day, which can make the whole day non-productive [1]. The findings of the email analysis show that, on average, about 60 seconds is spent per email excluding recovery time, and a total of two minutes is spent if the recovery time is assumed for each email message. This compares well with the interruption reported for telephone calls, [1]. Clearly email has less of an interrupt effect than the telephone although it is still significant. It is important, therefore, that any suggested methods for reducing the effect of email interrupt should not encourage employees to seek alternatives as this is likely to increase the interrupt effect.
The following methods for reducing the interrupt effect of email are considered.
1. Employees education on interrupts
In Solingen’s research, educating employees about the negative effects of interrupts has decreased interruptions by 30 percent [2]. The Danwood research showed that a large number of email messages received were not relevant to the person receiving them as they were sent to all staff. Many employees then used the reply-to-all facility generating further non-relevant emails. Education on the importance of reducing this email by targeting email to only those who need to receive it is bound to have an overall beneficial effect. The company can assist this targeting by providing suitable email groups to enable the targeting to be performed simply and effectively.
Simple education could also assist employees to minimise the inconvenience of non-relevant emails. The employees who used the facility to display the first three lines of each message were able to make more effective use of their email monitoring activity. Not only did it enable non-relevant emails to be deleted immediately it also allowed employees to prioritise their handling of each message. If all employees used this facility it would reduce the effect of the widespread interruptions caused by email-to-all messages.
2. Educating employees about the nature of interrupts
The research at the Danwood Group showed that the average interrupt time taken for an email was about two and a half minutes. This was assuming that every email interrupt would generate an interrupt recovery time. In practice on some occasions two or more emails could be received at one time so this would further reduce the average interrupt recovery time for each interrupt. Educating employees of these facts could persuade employees to use email rather than telephone calls. Reducing telephone calls by 4 per employee per day would in turn save 50 minutes’ recovery time as well as the time taken for the actual telephone calls. De Marco and Lister’s research into telephone interrupt recovery times was carried out on software developers, and the 15 minutes it takes a software developer to recovery from a telephone interruption is unlikely to apply to most of the employees studied at the company. However, the 15-minute interrupt recovery time gives an indication of the time that could be saved if an employee decided to use email instead of the phone for some of their communication.
3. Changing the settings of the email software.
Microsoft Outlook 2000 has a facility, as many other email clients do, that allows the user to set the frequency of when the email application should check for new mail. The results at the company showed that the majority of employees have their email application set to check for email every 5 minutes. It would be more productive to the user if the email was checked at an interval when interrupts would be less frequent and possibly even beneficial. Many people find that long periods of concentration on one single task reduces their effectiveness, and for that reason it can be beneficial to take a break and think about something else for a few minutes. The length of time a person can effectively concentrate on one task will vary from person to person, but typically a break after 45 minutes or so is likely to be beneficial. This gives an indication of how long an interval would be desirable for email interrupts. If employees set their email applications to check for email no more frequently than every 45 minutes (or whichever interval they find most suitable for themselves) then the interruption that does take place is clearly less disruptive and could even have a beneficial effect.
With the email software checking for new email every 5 minutes over an 8-hour day this means that in the worst case these employees could have 96 interruptions in the day. The total interrupt recovery time would then be 102 minutes. If the checking for incoming email is reduced to every 45 minutes this will mean there is a maximum of 11 interruptions in an 8 hour day, with a total interrupt time of only 12 minutes per day. Setting the email software-checking interval to an even longer period would make still further time savings. The monitoring of the Danwood employees showed that the peak of email activity occurred at four points in the day; at the start of the day, just before lunch, just after lunch and at the end of the day. If employees could be persuaded to restrict their use of email to these four periods by checking for incoming email manually, removing the automatic checking, then arguably this would reduce the interrupt recovery time altogether.
4. Changing the method of incoming email notification
Most email applications allow different means of notifying the user of incoming email. These different means can give different levels of intrusion, as some are easier to ignore than others. For example, a prominent sound accompanying a pop-up dialog box has a far more intrusive effect than a small icon that appears in one corner of the screen, yet the user is just as aware of the incoming email whichever is used. A less intrusive notification method may encourage users to respond in their own time, at a moment which is more convenient for them, and which will give less of an interruption to their concentration on the job in hand.
5. Encourage the use of one line emails
It is noticeable that many employees used a verbose style when writing emails. An informal observation in this research suggested that 50% of emails communicated could be handled in a one-line message. A message expressed in one line is obviously quick to read and minimizes the overall time spent reading emails. Furthermore, if the whole message is then put in the subject line, the reader need not even open the message and can decide what to do with the email very quickly.
To exploit the one line potential for email messages a sample of employees at the company were provided with an in-house developed extra email facility. The short message activity was monitored and the results collated to see whether the Short Message Service had increased the effectiveness of communicating within the workplace, by reducing the time spent composing and reading emails.
The incentive for the message senders to use the program was that their messages automatically had high priority. For the receiver the message was displayed in the Inbox list of messages, eliminating the need to “open” each individual message (see Figure 1).
Figure 1 – Outlook 2000 Inbox with SMS and Traditional Email Messages
In a survey of the users who tried the SMS emails, 80% said they found the facility saved them time. A side effect was that users of the SMS facility learned to be more succinct with their messages in general so that even their traditional emails were shorter and better structured.
It is clear from the observations made that many employees could reduce the time lost to interrupts through simple changes of practice. It has been possible to make recommendations to the company that would allow more effective and efficient use of email. The recommendations given below are based only on the experience of the Danwood Group, however, informal enquiries of other companies suggest that the experience at the Danwood Group is not likely to be unique, and so it is believed that other companies would also benefit from putting these recommendations into practice.
The recommended guidelines for email use in the workplace are as follows:
This research has shown the value of measuring communication processes. The implication for managers in other companies is that if their own employees have similar practices in using email then the companies would also benefit from following these recommendations.
[1] DeMarco T., and Lister T. 1999. Peopleware. Productive Projects and Teams 2nd Ed. New York: Dorset House Publishing Company.
[2] Solingen R., Berghout E., and Latum F. September/October 1998. “Interrupts: Just a Minute Never Is.” IEEE Software 15:97 - 103.