The previous post on why companies will always miss their targets, immediately raises the question: how should companies set targets, then? The immediate, and provocative, response would be: why should they set targets? Is there any evidence that companies who set targets get any benefit from the exercise? McKinsey and Co., for instance, does not set targets, there is no evidence that they have become laggards in their field. Ok, maybe that is too radical for some people to swallow. If we do insist on setting targets for the company as a whole, here are some thoughts on how to do it: 1. By strict analogy to Theory of Constraints thinking: top management should set an overall target for the company and announce it. Every account manager runs as hard as he/she can, and top management keeps monitoring the buffer (the shortfall, in this case). (In TOC applied to project management, the buffer is maintained only at the project level, not at the level of individual tasks, and it is the only metric monitored continuously). As the year progresses, top management shifts resources around, or adds resources to close the gap. Whoever puts up his hand and volunteers to do something extra or innovative to cut the shortfall is a hero, and so on. The point is, no account-level deficits or buffers, only company-level. 2. How is the company-level target to be set? By reference to external factors only, certainly not by an aggregation of account-level targets. Systems thinking requires us not to make the mistake of thinking global optimization is to be achieved by optimizing at subsystem level. 3. The difficult trick will be – how to evaluate and incentivize account managers. I would again think about this at two levels: at a philosophical level, the important thing is company performance; evaluating people is a secondary question and it doesn’t matter much if we come up with inexact ways of doing it. At an operational level, I would suggest a subjective evaluation by the CEO/COO of each account manager, based on assessment of his/her reality, and how much effort/innovation he/she put in to grow the business. We can even have a metric made up of absolute growth, customer satisfaction, innovation, and people satisfaction. Just because we want to be able to measure the performance of our managers, let us not make that itself a constraint to our growth! In the first place, there is no evidence that incentives improve performance (and a lot of evidence that they reduce performance), in the second place, let us not hurt the main goal – which is performance, just because we want to achieve a secondary goal – measuring performance. It is like weighing down Usain Bolt with a 10 kg speedometer just because we want measure how fast he runs. Let us find another way to evaluate and reward managers, that doesn’t directly constrain performance itself.

Why a company will always miss its target

Dr. Goldratt, in his wonderful work on project management, demonstrates that a project is guaranteed to be late if it is managed in the conventional way break down the project into work units, assign deadlines to each and build in buffers in each time estimate. His reasoning is roughly as follows: nobody is going to finish early because there is no incentive to do so, anyway work always expands to fill the time available, and so on – when we bring in what he calls ‘Murphy’, by which he means unpredictable uncertainty, ‘some s---‘ happening, basically, which will cause some overrun somewhere, the project as a whole is guaranteed to overrun, since nobody is going to make up for it. If we apply this same kind of thinking to revenue targets, we get a similar law – a company as a whole will never meet its target if its target is simply the aggregation of individual targets, for instance, on accounts. I have observed the same kind of behavior by account managers that Goldratt describes in the context of a project: as the year end, or quarter end approaches, an account manager who knows he is going to meet his target, eases off and coasts. Meanwhile, something is bound to go wrong in some other account, which will end up falling short of its target, thereby guaranteeing that the aggregate target is not met! This is a behavioral issue and cannot be resolved by adding or subtracting buffers to targets. It can only be met by recognizing, rewarding absolute performance, not performance against targets.. ‘run as hard as you can’ in Goldratt’s words. I believe the very process of target setting is causing us to underperform as an organization.