Vision and values: Craft your message carefully
One result of my getting sick at the beginning of this year is that piles of the academic journals I normally read got put into a big unread pile on a table. Recovering now from my hospitalization, I have just begun to attack the stack. And in doing so, I just came upon a shining golden nugget, in the form of a paper with the intriguing title "A (Blurry) Vision of the Future: How Leader Rhetoric about Ultimate Goals Influences Performance." The authors are three young business school assistant professors -- Andrew Carton, Chad Murphy and Jonathan Clark -- and the paper appears in the December 2014 issue of the Academy of Management Journal.
Their study is both an extremely careful piece of academic research (which professors like, but which practitioners also should care about because it affects how believable the findings are), and also one with enormous practical implications for government managers.
The paper studies the messages managers can give employees presenting a vision for what the organization seeks to accomplish and the values for which the organization stands, and how such messages are used by managers as a way to improve an organization's performance. The authors argue there are effective and ineffective ways to present both vision and values. For vision, the more concrete the better. For values, the fewer the better.
A less effective vision presents a general concept ("Our vision is that our toys will be enjoyed by all our customers"). A more effective vision presents an image that employees can truly visualize ("Our vision is that toys will make wide-eyed kids laugh and proud parents smile"). The authors believe an actual image is both more vivid and less subject to varying interpretations of what it means.
In terms of values, a short list of declarations promotes common purpose. As the number of professed values increases, so to does the potential for disagreement among employees about which are really important.
Communicating vivid images and a limited number of values, they argue, will improve performance by creating a shared view that makes it easier for employees, especially those working on different tasks, to coordinate their activities. That, in turn, contributes to more efficient and effective production.
The authors tested their theory with two studies. For one, involving a dataset of all the acute care hospitals in California, they tracked hospital readmission rates (patients who come back to the hospital after release because they fall sick again), which are related, among other things, to successful coordination among caregivers and staff. With careful measurement of the kinds of messages hospital executives deliver to staff (using a mixture of publicly available hospital documents and interviews at a sample of hospitals), they provided evidence that use of strong images reduces readmission, but only when combined with a smaller number of values emphasized.
In a second study, a lab experiment involving collaborative design of a toy, they measured successful coordination in terms of the number of errors made by team members, and found similar results.
These results are very important for government managers. Crafting effective messages for employees is not resource-intensive or even particularly time-intensive (especially compared with crafting ineffective messages, which presumably takes the same amount of time). Doing so requires no statutory changes or union negotiations; any manager can do it.
That's not to say that it's easy, however; the study found that effective messages are rare. The researchers' analysis found that in vision statements, there were 16 conceptual words for every image word. And the average number of values referred to was greater than four per statement. "In sum," they write, "most leaders exhibit a 'blurry vision bias in which they (a) provide conceptual (rather than concrete) visions and then (b) communicate a number of values that further obscures the vision."
So here is a scrupulously careful piece of scholarship with conclusions that are of immediate practical relevance to government managers who want to improve their organization's performance.
In my last blog on "nudge," I noted that the message was that the message matters. In a different way, we see the same result here: the messages managers give -- both how they give them and how many -- can have measurable ramifications.
Carton, Murphy, and Clark, thanks for writing! Now, managers – will you listen?
Posted by Steve Kelman on Sep 21, 2015 at 12:48 PM