Every decision is a choice between alternatives. Another element is the right time for the decision to be made. Very few decisions force the manager to decide immediately. In itself this is an undesired situation where a threat has emerged in complete surprise. Most decisions leave enough time to the decision maker.
Facing substantial uncertainty suggests that every decision should be delayed until the last moment that allows executing the decision in full. The underlining assumption is that time adds information that reduces the uncertainty.
There are serious negative branches to the above logical claim. All of them look at what might go wrong with the suggestion. Here are several of them:
- We don’t truly know the exact timing of “the last moment”, so we may miss it.
- We might forget the decision at the right moment.
- Our attention might be occupied by more important issues at the critical time.
- Making the decision at the very last moment makes it urgent and generates stress. The stress might affect us to make the wrong decision!
Inspired by the idea of time buffers we should treat every non-trivial decision as a mission that should be completed at a given time and to assign a time buffer for that mission. According to the DBR interpretation of time buffers the mission should not start prior to the buffer. The time buffer itself should provide enough time to deal with all the other requirements for attention, without creating stress, except the need to make the RIGHT decision.
Managing the execution of missions by individuals or teams through assigning time-buffers, and using buffer management as a control mechanism, is a much more effective process than check-lists. It reduces multi-tasking through buffer management priorities and limits handling missions, especially decisions, too early. Only non-trivial tasks should be included in the missions. It is a step forward in understanding the behavior of the capacity (attention) of individual managers. It would also clarify the issue of distinguishing between missions and projects.
An example
Suppose a decision to stop working with a certain supplier and switch to another one is considered. The decision process requires updated information on the trouble with the current supplier and mainly finding alternative suppliers, inquiring how they are evaluated by their clients, whether they have the specific capabilities and, of course, their pricing.
When is the deadline to make the above decision?
Suppose the contract with the current supplier ends on December 31st 2016. If the contract is not going to be extended, it is fair to announce it by December 1st, which also leaves enough time for finalizing the contract with the new supplier. The mission includes getting the relevant information, bringing it to the decision maker(s) and letting 1-2 hours for the decision itself. Assigning three weeks for the whole mission is reasonable. This means no one should work on that mission prior to November 10th!
The impact of the criticality of the decision
Goldratt said: “Don’t ever let something important to become urgent”.
The practical lesson is: important decisions should be given a reasonable time buffer. Very important decisions, those that we call ‘critical’, should be given longer time buffer, ensuring the decision is not going to be taken under stress. Of course, a critical decision might be taken under stress because of possible negative ramifications, for which no viable solution has been successfully developed. This post only focuses on the time element.
The suggested process expands what TOC has developed for the production-floor to the work of managers taking responsibility for important missions.
Comment: a mission can be viewed as a very small and simple project, but it does not make sense to work on the mission continuously, unlike the expected behavior in projects, especially along the critical chain, where we strive for continuous progress.
Batching of Decisions
Batching of decisions, usually by periodical planning sessions, is widely done. The tendency to plan according to time periods can be explained by the need to gather together the relevant management team to come up with a periodical financial budgeting process based on forecasts. The targets for the various functions are derived from that periodic plan.
I’ve expressed in previous posts my negative view on one-number forecasts and how they reduce the overall performance of the organization. My focus here is to highlight that the planning sessions provide an “opportunity” to include other decisions that are not directly related to the purpose of the periodical planning.
Any plan is a combination of decisions that are interconnected in order to achieve a certain objective. The plan should include only the decisions that any deviation from them would impact the objective. This message is explained in a previous post called “What is a good plan – the relationships between planning and execution”, including the need to plan buffers within the planning.
Does it make sense to include the decision to switch suppliers within the annual planning session aimed at determining the financial boundaries for next year? Is the identity of the specific supplier critical to the quality of that high-level planning? Suppose there is a small impact of the switch on the total cost of goods – does it justify forcing a decision too early?
The key point is that including decisions, with very limited impact on the objective, within the planning, disrupts the quality of the plan that needs to be focused just on the critical factors for achieving the objective. It forces timing that does not support the quality of the particular decision.
Planning, execution and the right timing of decisions are all part of handling common-and-expected-uncertainty. We need to vastly improve processes that dictate what is in the planning, what are left for execution and the handling of all the variety of non-trivial decisions including making sure they are made at the right time.
Eli Schragenheim 
Eli,
I recall the TOC-ICO Conference in Barcelona that may be relevant. I believe it was an afternoon session with the Upgrade Workshop with Dr. G. I do not recall the specifics of what led to the question, but a gentleman from India, that I did not recognize, in reference to projects or something, asked Eli to speak about “good delay.”
I recall that it really stopped Eli in his tracks so much that Eli asked “who said that?,” but not in a negative way, but in a respectful way. Eli spoke for less than a minute on the concept of when delays occur in a project that may or may not be moving along smoothly, and good things happen. This occurs commonly in software design. Maybe something strange interrupts the push for progress on a new release and the extra time allowed results in the software engineers discovering a new way to do something that is a sea change. Something that would not have occurred without the delay. So one would ask … when is delay good?
This can easily be applied to making a decision. I believe all of us can reflect on a decision or two that if we had made them at the time we waned to make them, knowing everything we knew at the time, believing our information was sufficient, but something caused us to delay, then new information was available and we thought “goodness gracious, I just dodged a bullet!”
Its not so easy to even reflect on when this occurs because we are part of he system. Observing ourselves is no easy task. It’s even more difficult to be aware of this while we are in the decision making process, and even more difficult when we are feeling time pressures (… which is pretty much most of the time …).
Seems like we can tell (can we really??) when we shouldn’t have made a decision, but we dont know it at the time.
Yup, its a great question: when is the right time to make a decision.
LikeLike
Thanks, as always, Eli.
LikeLike
And then there is the notion of Blitzkrieg. It works wonderfully when the other guy is trying to decide when to make his decision.
LikeLike
Henry, Blitzkrieg raises the question of the capability of the other side to identify emerging threats. One of my older posts on this blog was dedicated to: Identifying the Emergence of Threats. Once you are under attack you have to react immediately. Not a good situation to be. Remember Goldratt warning: Never let something important to become urgent.
LikeLike