The effective decision - The Elements of Effective Decision

The effective decision
– synopsis of an article by Peter F. Drucker.

This is definitely worth re-publishing …. a classic  read.

NOTE: best practice process and tools available here

Best Practice Decision Making

While browsing through the B School library the other day, I happened across an article on decision making by Peter Drucker, originally written in 1967. Naturally as this is our business I was curious to see what one of the true masters of management had to say on the topic. Here is the synopsis together with our commentary.

Drucker commences by stating that an effective decision making process must go through some basic steps. These steps will not “make” the decision – it will always be a judgement call – but if the steps are ignored, the decision is not likely to be effective nor right. The 6 steps he recommends are:

1. The classification of the problem
2. The definition of the problem
3. The specifications which the solution to the problem must satisfy (the “boundary conditions”).
4. The decision as to what is “right”, rather than what is acceptable, in order to meet the boundary conditions
5. The building into the decision of the action to carry it out.
6. The feedback which tests the validity of the decision against the actual course of events.

We will show how these steps fit into the overall Genesis Decision Making framework a little later, but first a brief description of the steps and why they are important.

Step 1: Problem classification

Drucker postulates that a decision falls into 2 broad categories: generic (where the situation has happened before and a set of rules or principals may be applied) or unique (must be treated individually and pragmatically). There is a further categorisation of (a) generic although unique to the organisation and (b) generic, although only the first event of a new trend or genus. Both the latter appear to be unique but are not truly so. He states that if the problem is incorrectly classified at this stage, then the decision will inevitably go wrong.

Step 2: Problem definition.

Here the decision maker must work out what the situation is all about and what are the key issues. The danger, he claims, is that of an incomplete definition but one that is plausible. The only safeguard being to check the definition again and again against all the observable facts and throw out the definition the moment it fails to encompass them. That is, doing what we now call, seeking dis-confirmatory evidence as well as confirmatory evidence. Although much has recently been written about this under the title of behavioral economics, he reminds us  that these are simply the rules of scientific observation first formulated by Aristotle and then reaffirmed by Galileo.

Step 3: The specifications (“boundary conditions”)

It must be clearly defined what the decision must accomplish, that is what are the minimum goals it has to attain. In science, these are known as boundary conditions. Drucker says that a common problem in decision making is not necessarily the wrong decision, but a circumstance when the boundary conditions change while the decision is being implemented – such as may have happened to organisations who started a decision process pre-recession and are now trying to implement it in the midst of the economic crisis.

He also states that another reason to have boundary conditions clearly defined is in when one is making the most dangerous of all decisions – which is when the conditions are essentially incompatible. That is when the decision might, if nothing goes wrong, work. This is what he calls little more than “gambling”.

Step 4: The decision: what is right

It is critical to decide what is right. That is not to say that a compromise may not eventually have to be made when implementing (inevitably it will), but rather start with the best decision that meets all the boundary conditions and  then, if necessary, compromise from that position. Drucker brilliantly demonstrates this by explaining there are two types of compromise. One is expressed in the proverb: “Half a loaf is better than no bread”. The other in the story of the Judgement of Solomon where it is realised that “half a baby is worse than no baby at all”! In a nutshell, he is saying that we should not be thinking about “what will be acceptable” to others (at least initially), rather “what is the right answer?”.

Step 5: Converting the decision into action.

Drucker says that a decision is not a decision until it has been acted upon. He goes further to state that the action should be built into the decision from the outset. He suggests 4 distinct questions:

• Who has to know of the decision?
• What action has to be taken?
• Who is to take it?
• What has to be done so that these people can take the action?

He notes that the first and last questions are most frequently overlooked and then reminds the reader that the action must be appropriate to the capacities of the people who have to carry it out.

Step 6: Feedback

Drucker reinforces that men are fallible and decisions can go wrong and may not achieve their desired results. Therefore a feedback mechanism must be put in place to monitor and report back on the success or otherwise of the outcome. He says that effective decision makers realise that often they should not rely on reports but, like military commanders, must go into the field themselves to see how the decision is being carried out. Peter Drucker, with accurate foresight (remember this was written in 1967) warns that with the advent of computers this is even more important as computer-generated reports only can report back on abstractions. A final comment in this section is “Failure to go out and look is the typical reason for persisting in a course of action long after it has ceased to be appropriate or even rational.”

In more recent years, Genesis Management Consulting has constructed their own proprietary strategic decision making framework that incorporates all of Drucker’s steps within a slightly larger and more detailed framework. Although Drucker does not mention all of the steps in the Genesis framework specifically, we have little doubt that he would recognise them all as being valid and a valuable complement to his own original work. The Genesis framework is shown below:

(Note: this decision framework together with detail on each of the steps is available to purchase at

The Decision Shop for a nominal cost)

The way in which Drucker’s steps overlap with the Genesis Decision Framework can best be shown in the diagram to the right where the initial three steps fall into the decision structuring; the fourth step under evaluation and the fifth and sixth steps under decision and implementation.

Drucker’s original article is available for purchase from the Harvard Business Review library and is worth reading as it contains a little more detail and some examples of the  steps and the pitfalls. Warning, it is written in a somewhat old-fashioned way and does not make for easy-reading – but it is probably worth the effort.

催生有效的決策

The Effective Decision

出 處 / 2013年8月號(大師的四堂決策課)

作 者 / 彼得．杜拉克 (Peter F. Drucker)

主 題 / 領導

杜拉克認為，主管想要發揮高效能，在決策上，就必須集中心力，鎖定少數重要的決策，而不是事無巨細地什麼決定都握在自己的手裡。因此，在作出決策時，他們會循序進行六大步驟：分類、定義、設定範圍、決定、行動、測試。

有效能的主管並不會做出許多決策，而只專注在重要事務上。他們致力制定的少數重大決策，是放在概念理解的最高層次上。他們希望找出的，是狀況中的常數，思索其間的策略性與共通性，而不是「解決問題」。因此，他們並不特別看重決策的速度；反而認為熟練操控許多變數，是思考馬虎的一種症狀。他們希望知道決策的所有相關事宜，以及決策背後需要滿足的現實條件。他們希望的是影響力，而不是技巧。他們希望決策是健全的，而不是聰明的。

有效能的主管懂得視個案狀況，決定決策何時須根據原則，何時應務實。他們知道，最棘手的決定，是判斷何者是對的妥協，何者是錯的妥協，也學會如何區分兩者。他們知道，流程中最花時間的步驟，並不是做出決定，而是將決策付諸施行。除非決策已落實到工作上，否則就不是決策，充其量只能算良好的立意而已。這就是說，雖然有效決策本身是根據概念理解的最高層次，但行動承諾則應盡量接近負責執行者的能力。最重要的是，有效能的決策者了解，決策本身有系統化流程與定義清楚的元素。

︱依序決策六步驟

決策元素本身並不會「做」決策。其實，每一個決策都是冒風險的判斷。但除非以這些元素為決策流程的基礎，決策主管將無法得出正確的、當然也不是有效的決策。因此，本文描述決策流程涉及的步驟順序如下。

1. 問題分類。是普遍的問題，還是例外而獨特的問題？或者，是一種首度現身的新類型，相關處理準則尚待整理歸納？

2. 定義問題。我們要處理的是什麼？

3. 針對問題的答案，設定目標範圍（Specifying）。也就是，「邊界條件」（boundary conditions）是什麼？

4. 決定什麼是「正確的」，而不是「可接受的」，以求符合邊界條件。先注意什麼才能充分滿足所有的範圍設定？然後，再談需要哪些妥協、調整與讓步，好讓決策可被接受。

5. 將執行決策所需的行動納入決策中。行動承諾應該如何？有誰該知道？

6. 以實際事件測試決策的正確性與有效性。決策如何付諸實施？根據的假設是否適當，或是已經過時？

以下就一一檢視這些元素。

步驟1：分類

有效能的決策者會問：這個症狀是根本的脫序，或偶然偏離常軌？普遍性的問題，必須根據某種準則或原則來解決，但真正的例外事件，就只能視狀況隨機應變了。

嚴格來說，決策者可區分出四種不同類型的事件，而不是兩種。

首先是真正普遍性的事件，個別發生的問題只是一種症狀。主管工作上碰到的大多數「問題」，都屬於這種性質。例如，企業的存貨決策，就不是「決策」，而只是調整處理準則。這個問題具有普遍性，尤其在製造部門的組織上更是明顯。例如：

某產品管制與工程團隊，通常每月要處理幾百個問題。然而，在分析問題時，卻發現大部分只是潛在基本情況的病兆與展現。每位流程控制工程師或生產工程師，工作範圍只局限在工廠的一部分，通常看不到這點。他們或許每個月會碰到幾件管路內連結器有水蒸氣或熱水的問題，但也僅止於此。

只有分析整個團隊好幾個月以來的全部工作，普遍性的問題才會浮現。然後就可看出，現有設備溫度或壓力過高，需要重新設計連繫不同管線的連結器，足以承受更大的負荷。除非做過這項分析，否則流程控制單位會耗費許多時間修理漏水，卻無法真正控制狀況。

第二類情況是，雖然問題對個別機構屬於獨特事件，但實際上具普遍性。例如：

某家公司接到另一家較大公司合併的提議，如果接受了，以後就不可能再次接到這類提議。對那家公司及他們的董事會與管理階層而言，這是不會重複發生的狀況。但擴大至整個商業界來看，這當然是隨時都在發生的普遍狀況。考量是否接受或拒絕這項提議，需要一些通用的準則。這個時候，決策者就必須參考他人的經驗。

第三類，是主管必須清楚分辨的真正例外事件。例如：

1965 年 11 月的大停電，讓美國東北部從聖勞倫斯到華盛頓都陷入黑暗，根據第一時間的說明，算是真正例外狀況。另外，像是 1960 年代初，鎮定劑沙利竇邁（thalidomide）導致許多畸形新生兒的悲劇。根據報導，發生其中任一事件的機率，是千萬或一億分之一，而這些事件再度連續發生的機率更微乎其微，就像我現在坐的椅子會分解為原子大小一樣，是幾乎不可能的事。

真正獨特的事件很罕見，不過一旦出現，決策者就得問：這是真正的例外，還是一個新類型首次現身。「一個新的問題類型乍現」，是決策流程要處理的第四類，也是最後一類事件。因此：

我們現在知道，不論美國東北部大停電或沙利竇邁的悲劇，都只是在現代電力科技或現代藥理學環境下，最先發生的事件，除非找到普遍的解決方案，否則可能會變成常見的事件。

http://www.genesismc.co.uk/blog/drucker-effective-decision/