Too Many Interruptions at Work?

Robison (2006) ยท Gallup Business Journal

[NOT A STUDY. A trade-press interview. COMPLETE INTERVIEW READ IN FULL AT SOURCE at news.gallup.com/businessjournal/23146, and RE-READ IN FULL at source on the verification pass of 2026-10-22. Page metadata gives article:published_time 2006-06-08 and article:modified_time 2024-10-16. The item carries no DOI, so this entry is keyed on first author and year] Robison, J. (2006). 'Too Many Interruptions at Work?' A Q and A with Gloria Mark. Gallup Business Journal, 8 June 2006. A JOURNALIST'S INTERVIEW, NOT A PEER-REVIEWED STUDY, and it is where the most quoted number in the attention trade actually comes from. Interviewed by Jennifer Robison; Mark's title as printed is associate professor at the Donald Bren School of Information and Computer Sciences, University of California, Irvine. VERBATIM AND COMPLETE: 'GMJ: How long does it take to get back to work after an interruption? Mark: There's good news and bad news. To have a uniform comparison, we looked at all work that was interrupted and resumed on the same day. The good news is that most interrupted work was resumed on the same day, 81.9 percent, and it was resumed, on average, in 23 minutes and 15 seconds, which I guess is not so long.' FOUR THINGS THE CIRCULATED VERSION DESTROYS: the figure is CONDITIONED on the 81.9 per cent resumed the same day, so it UNDERSTATES rather than overstates the cost across all interrupted work, while the trade uses it as an overstatement; her own verdict on her own number is that it is small, 'which I guess is not so long'; the bad news in her framing is not the clock but the detour, 'when you're interrupted, you don't immediately go back to the task ... There are about two intervening tasks before you go back to your original task, so it takes more effort to reorient back to the original task'; and the interview's whole argument is that interruptions can be good, with Gallup's own standfirst reading 'Interruptions can actually be quite beneficial' and Mark saying 'when they were in a working sphere, and someone interrupted with the same content as that working sphere, people called it interaction. And then it's beneficial.' ALSO HERE, AND CONFIRMED WORD FOR WORD ON THE VERIFICATION PASS, her answer when the interviewer puts it to her that there must be a cost to productivity: 'There must be, but we don't know what [it is].' THE DATASET QUESTION, TO BE STATED CAREFULLY: the interview describes a different and larger dataset from the published CHI 2005 paper, 'we observed for a half day, then we shadowed 36 managers, financial analysts, software developers, engineers, and project leaders for three days', against twenty-four informants in the paper, and gives 12.2 working spheres a day against 11.7, 10 minutes 29 seconds before switching against 11 minutes 4 seconds, and 23 minutes 15 seconds resumption against 25 minutes 26 seconds. A PAPER PUBLISHING THE THIRTY-SIX-PERSON DATASET AND THE 23:15 FIGURE WAS SEARCHED FOR TWICE AND NOT FOUND. WHAT MAY NOT BE SAID: the figure may NOT be described as invented, fabricated, wrong or misreported, and Mark may not be said to have misreported anything. She conditioned the figure when she gave it, published the larger one for the same measure, and called the number short in the same sentence. THE FAULT IS ENTIRELY IN THE CIRCULATION and must be stated in the same paragraph as the point. ALSO IN THIS INTERVIEW: 'the average amount of time that people spent on any single event before being interrupted or before switching was about three minutes. Actually, three minutes and five seconds, on average. That does not include formal meetings', and with sub-two-minute interruptions removed 'people still worked 12 minutes and 18 seconds in a working sphere before switching', and self-interruption 'about 44% of the time' against the published 52.0 per cent, so the circulated 'every 3 minutes' is not an interruption rate at all. NOT ESTABLISHED: a 2008 Fast Company appearance of the figure, claimed by secondary blogs; no such article was reached, and an earlier appearance than June 2006 cannot be ruled out because pre-2006 conference coverage was not searched. ALSO OFFERED UNPROMPTED IN THE SAME ANSWER AND NOT LOCATED INDEPENDENTLY: a study of extreme collaboration with NASA engineers, which may be cited only as something she says in this interview and never as a finding.

[NOT A STUDY. A trade-press interview. COMPLETE INTERVIEW READ IN FULL AT SOURCE at news.gallup.com/businessjournal/23146, and RE-READ IN FULL at source on the verification pass of 2026-10-22. Page metadata gives article:published_time 2006-06-08 and article:modified_time 2024-10-16. The item carries no DOI, so this entry is keyed on first author and year] Robison, J. (2006). 'Too Many Interruptions at Work?' A Q and A with Gloria Mark. Gallup Business Journal, 8 June 2006. A JOURNALIST'S INTERVIEW, NOT A PEER-REVIEWED STUDY, and it is where the most quoted number in the attention trade actually comes from. Interviewed by Jennifer Robison; Mark's title as printed is associate professor at the Donald Bren School of Information and Computer Sciences, University of California, Irvine. VERBATIM AND COMPLETE: 'GMJ: How long does it take to get back to work after an interruption? Mark: There's good news and bad news. To have a uniform comparison, we looked at all work that was interrupted and resumed on the same day. The good news is that most interrupted work was resumed on the same day, 81.9 percent, and it was resumed, on average, in 23 minutes and 15 seconds, which I guess is not so long.' FOUR THINGS THE CIRCULATED VERSION DESTROYS: the figure is CONDITIONED on the 81.9 per cent resumed the same day, so it UNDERSTATES rather than overstates the cost across all interrupted work, while the trade uses it as an overstatement; her own verdict on her own number is that it is small, 'which I guess is not so long'; the bad news in her framing is not the clock but the detour, 'when you're interrupted, you don't immediately go back to the task ... There are about two intervening tasks before you go back to your original task, so it takes more effort to reorient back to the original task'; and the interview's whole argument is that interruptions can be good, with Gallup's own standfirst reading 'Interruptions can actually be quite beneficial' and Mark saying 'when they were in a working sphere, and someone interrupted with the same content as that working sphere, people called it interaction. And then it's beneficial.' ALSO HERE, AND CONFIRMED WORD FOR WORD ON THE VERIFICATION PASS, her answer when the interviewer puts it to her that there must be a cost to productivity: 'There must be, but we don't know what [it is].' THE DATASET QUESTION, TO BE STATED CAREFULLY: the interview describes a different and larger dataset from the published CHI 2005 paper, 'we observed for a half day, then we shadowed 36 managers, financial analysts, software developers, engineers, and project leaders for three days', against twenty-four informants in the paper, and gives 12.2 working spheres a day against 11.7, 10 minutes 29 seconds before switching against 11 minutes 4 seconds, and 23 minutes 15 seconds resumption against 25 minutes 26 seconds. A PAPER PUBLISHING THE THIRTY-SIX-PERSON DATASET AND THE 23:15 FIGURE WAS SEARCHED FOR TWICE AND NOT FOUND. WHAT MAY NOT BE SAID: the figure may NOT be described as invented, fabricated, wrong or misreported, and Mark may not be said to have misreported anything. She conditioned the figure when she gave it, published the larger one for the same measure, and called the number short in the same sentence. THE FAULT IS ENTIRELY IN THE CIRCULATION and must be stated in the same paragraph as the point. ALSO IN THIS INTERVIEW: 'the average amount of time that people spent on any single event before being interrupted or before switching was about three minutes. Actually, three minutes and five seconds, on average. That does not include formal meetings', and with sub-two-minute interruptions removed 'people still worked 12 minutes and 18 seconds in a working sphere before switching', and self-interruption 'about 44% of the time' against the published 52.0 per cent, so the circulated 'every 3 minutes' is not an interruption rate at all. NOT ESTABLISHED: a 2008 Fast Company appearance of the figure, claimed by secondary blogs; no such article was reached, and an earlier appearance than June 2006 cannot be ruled out because pre-2006 conference coverage was not searched. ALSO OFFERED UNPROMPTED IN THE SAME ANSWER AND NOT LOCATED INDEPENDENTLY: a study of extreme collaboration with NASA engineers, which may be cited only as something she says in this interview and never as a finding.

Written from the nOT A STUDY. A trade-press interview. COMPLETE INTERVIEW READ IN FULL AT SOURCE at news.gallup.com/businessjournal/23146, and RE-READ IN FULL at source on the verification pass of 2026-10-22. Page metadata gives article:published_time 2006-06-08 and article:modified_time 2024-10-16. The item carries no DOI, so this entry is keyed on first author and year.

What our sources record

Recorded when this source was checked, figures included, with page references and quotation marks removed.

A Q and A with Gloria Mark. Gallup Business Journal, 8 June 2006