2016 is here. Worldwide, managers are setting SMART – Specific, Measureable, Achievable, Relevant, and Timely – performance goals. Even though the underlying theory – Management by Objectives (MBO)– has lost credence, this catchy mnemonic, which George Doran coined in 1981, has not. The time is upon us to retire SMART for all managers and executives from whom we need discretionary effort.
MBO lost credence because “The boss knows best” paternalism no longer works well. Global companies are transforming structurally to free emerging businesses from top-down strictures. Open source communities, coordinated by respect for expertise, not central authority, are creating technologies and products. Innovative workplaces are giving employees time off the clock and free resources, and benefitting from their unmeasured, untracked tinkering. Such environments thrive on distributed leadership and decentralized, uncounted action, and SMART goals can’t add to, and inevitably subtract from, them.
The problems with SMART run deeper and can damage even organizations that don’t need to unbundle business units or use open source approaches. The business environment has fundamentally changed. Companies no longer compete individually, but as members of networks: Apple couldn’t create the iPhone, or Airbus the A350 aircraft, without collaborating with others. Network members may be located half a world away, and inevitably have their own strategies, processes and cultures. So, complexity, uncertainty, and ambiguity abound, which allow problems and opportunities flash across these networks with blinding speed, meaningfully affecting performance. SMART goals implicitly assume staid environments that are far removed from these realities and can keep executives from responding appropriately.
Problems with SMART arise from virtually all elements of the acronym. ‘Specific’ goals, clear-cut and definite, are easy to articulate and act on. They enable quick assessments of individuals’ successes. However, when used extensively, they reduce discretionary activity and limit broader action. I once facilitated a meeting between two groups of senior executives, each from a well-known global company, whose businesses had merged. One group described how its corporate values drove performance evaluation and gave it freedom to act. The other retorted that its values were the five tasks set for each by his/her boss; each manager could, and did, decline to work on any initiative unless specifically tasked to do so. Guess which company had acquired the other? Guess which one’s stock price has usually outperformed the other’s?
‘Measurable’ goals have become an unshakeable article of faith, commonly justified by physicist Lord Kelvin’s dictum, “If you can not measure it, you can not improve it.” Such goals make it is easy to decide not just whether someone has performed, but how well. In so doing, they implicitly emphasize efficiency (doing something optimally, even if it is the wrong thing) over effectiveness (doing the right thing, which may be hard to discern). To make this point, I often ask senior executives to identify a single factor whose absence would destroy their businesses. They inevitably – and quickly – converge on ‘Trust.’ They are right – how much would you get done if you had to personally check every single word you were told? I then ask, “How do you measure trust?” They don’t – and can’t: this critical driver of business success is immeasurable. Instead of spouting Lord Kelvin out of context, executives should internalize the words attributed, perhaps aphoristically, to Albert Einstein: “Everything that can be counted does not necessarily count; everything that counts cannot necessarily be counted.”
Since ‘Achievable’ may produce inadequate outcomes, many companies set ‘Ambitious but achievable’ goals. Regardless, this criterion disregards our knowledge about human motivation. As Daniel Pink has brilliantly summarized in a YouTube video popular in business schools, carrot-and-stick approaches improve performance only when work is physical. Intellectual work benefits from allowing people to develop mastery in a field and giving them the autonomy to act therein. So, the quintessential carrot-and-stick nature of Achievable goals limits their relevance to managers. To drive supernormal performance, we should instead give people responsibility for accomplishment, and allow them to set their own targets consistent with organizational goals.
Goals are supposed to be ‘Relevant’ – not just to the organization, given its environment – but also to specific individuals and groups for whom they are set. The first criterion is undoubtedly reasonable, and on a prima facie basis, so is the second: why set a goal that isn’t applicable to, or deliverable by, the people in question? In reality, ‘relevance’ inevitably results in an enduring, widespread problem: organizational silos that hinder collaboration. For example, a sales team accountable for customer satisfaction is likely to have conflicts with a supply network team accountable for minimizing inventory. However, these silos would collaborate in their own interest if each was assessed (in part) on the other’s accountability – in effect measuring them for something that wasn’t relevant to their daily work.
How could ‘Timely’ not be legitimate? Very simply because it has become code for “as soon as possible.” We have made a virtue of speed to the exclusion of every other meaningful, and important organizational goal. Business textbooks assign critical importance to “first mover advantage” even though irrefutable examples of its falseness are readily available. When was the last time Apple launched a truly first-in-the world product? Was Google the first search company? Was Facebook the first social media offering in its niche? How are Chinese and Indian multinationals, Johnny-come-latelies to international markets, giving established Western firms a run for the money? When ‘timely’ equates to solely to speed, creativity, effectiveness and yes, even efficiency, suffer, sometimes irreparably.
What should executives do? They should reserve SMART goals solely for people who have limited discretion. For everyone else, they should begin goal setting with non-specific, qualitative, “can’t be done” diffuse and “time is one of several criterions” goals that give people autonomy and mastery. Indeed, they should urge them to propose goals for themselves. They should add SMART goals, only where they are truly unavoidable, and there too, with enough fudge-factors to ensure they don’t become limiting or constraining.
In effect, throughout the goal setting process, they should ask themselves: Am I paying attention to issues that truly matter? Am I truly leading an organization of people, or am I merely checking boxes to show that my job matters? Being SMART is easy, but that doesn’t make it right.
A shorter version was published by Forbes on January 12: http://www.forbes.com/sites/forbesleadershipforum/2016/01/12/it-may-be-time-to-get-rid-of-smart-management/#2715e4857a0b6e160d5e3bbc