如何做出更好的決定

How to Make Better Decisions
瀏覽人數:1078


影片載入中...
哈佛商學院的約翰.畢謝爾(John Beshears)和法蘭西絲卡.吉諾(Francesca Gino),提供一個五步驟的流程,可減輕認知偏誤和動機低落對決策的影響。

艾莉森.比爾德:嗨,我是《哈佛商業評論》編輯艾莉森.比爾德(Alison Beard)。今天的兩位來賓約翰.畢謝爾和法蘭西絲卡.吉諾,都是哈佛商學院教授,他們合寫了一篇文章〈成為決策建築師〉。感謝兩位能撥冗前來。

法蘭西絲卡.吉諾:謝謝邀請我們。

約翰.畢謝爾:很高興能來這裡。

艾莉森.比爾德:你們曾花很多時間研究決策,以及大多數人包括員工、經理和顧客,怎麼會做出錯誤的決定。但你們已開發出一種建構工作的方法,應該可以協助我們解決這個問題。是否可以說明一下?

法蘭西絲卡.吉諾:好的。第一步是要了解,為何所有人常做出差勁的決定。我們過於依賴康納曼(Daniel Kahneman)所說的系統一思考。系統一的思考是情緒性、直覺的、自動的。過於依賴它,會導致你產生各種認知偏誤,像是過度自信、團體迷思、損失規避、偏好現狀。但人們也可能更頻繁進行康納曼所說的系統二思考。系統二更慎重、更有邏輯、速度更慢。它告訴我們,我們的直覺何時是錯的,我們的情緒何時會妨礙我們的判斷。

第二步是界定問題。我們建議,有興趣使用我們方法的領導人,應考慮三個因素:這個問題是否源自與人類行為有關的事物,是否可以狹義地定義這個問題,人們的行為是否違背本身的最佳利益。在界定問題之後,第三步是了解問題背後的原因。

你應該問兩個問題:這個問題的主要原因,是否在於缺乏動機或動機不足,所以人們沒有採取行動?或者出問題的原因,是否在於人們有採取行動,但採取錯誤的行動。例如,約翰曾與一家零售商合作,這家公司試圖削減部分醫療成本,做法是讓這些員工申請郵寄處方藥,而非到藥房領取。

約翰.畢謝爾:是的。這是狹獈的問題,人們的行為不符合自己的最佳利益。雇主明白,當員工從親自到藥房領取慢性病藥物,改為宅配到家,而這些都是用於治療高膽固醇、高血壓的藥物,那麼贊助處方藥計畫的公司和員工,其實都可以省錢。

但許多員工沒有改為申請宅配。原因是慣性,也就是缺乏動機。因此這種情況就是,人們應該做一些符合自己最佳利益的事情,同時也對公司有幫助,但他們就是沒有做,原因正是康納曼向我們解釋的那些問題。

艾莉森.比爾德:那家公司做了什麼來改變員工的行為?

約翰.畢謝爾:這是我們流程的第四步,設計解決方案。在這個案例中,公司決定要讓人們運用系統二。它的做法是鼓勵員工做出積極的選擇。員工可以決定要改為宅配慢性病處方藥,也可以繼續自行去藥房領取。但無論選擇哪一種,若要享受處方藥計畫的好處,員工必須主動表明更喜歡其中哪一種做法。

他們可以繼續親自去藥房領藥,但若要這樣做,必須透過郵件、打電話或上網表明自己的選擇。這套流程的設計完全不繁瑣,非常簡單且不費力。但它的效果是,讓員工暫停一下,思考自己是否想改為宅配到家。其實經過深思之後,很多人都決定轉換為宅配。這個非常簡單的改變,導致郵寄領取慢性病處方藥的員工人數增加六倍。

艾莉森.比爾德:哇。

約翰.畢謝爾:是的。我要強調,不同的情況需要不同的解決方法。在我剛才描述的例子中,雇主決定鼓勵員工運用系統二。但我們提出的另一項建議,是鼓勵人們啟動系統一。其實,法蘭西絲卡有個例子是與印度某家公司的合作,他們解決方案的基礎,是藉由激發情緒來啟動系統一。

法蘭西絲卡.吉諾:我們在這個案例中與該公司合作,設法降低客服中心面臨的高離職率問題。這家公司是威普羅,而我們進行實驗的地點,是該公司的商業流程外包客服中心。我們做的事情非常簡單,是在新進人員到職流程中,針對兩個情況進行實驗。

其中一種情況是,新進員工在第一天的新人訓練當中,被要求花時間思考自己的強項,以及如何應用這些強項在工作上。另一種情況是我們的控制組,員工沒有被要求這樣做。他們沒有得到自我反思的時間。我們發現,這種反思確實讓員工與組織的聯繫(情感聯繫)更強。結果,離職率大幅下降。

我們也能提高員工的績效和工作投入程度。我們能在許多其他組織複製這個發現,做法是再次啟動系統一,透過與公司建立更好的聯繫(更好的情感聯繫)來啟動。

艾莉森.比爾德:你提到了對照組。我知道測試是你那套方法的第五個步驟。能否談談為什麼這很重要?

約翰.畢謝爾:好的。我們確實提出許多不同的可能解決方案,來解決不同的組織問題。法蘭西絲卡舉了一個很好的例子,有關觸發系統一思考,在這個例子是透過激發情緒來啟動。但你也可以考慮利用偏誤,或者簡化流程。然後還有運用系統二的可能性。這麼做的一個方式,是我之前舉的例子,就是經由鼓勵反思來做到。但你也可以促使人們制定計畫,也可以嘗試激勵組織內部人員更廣泛地思考。

然後甚至還有第三種可能,那就是完全繞過這兩個系統。要做到這點,你可以設定明智的預設值,或是在流程中引入自動調整。但我們建議選擇其中一種解決方案,然後與對照組進行嚴格測試,以了解哪種方案更好。這是步驟五,也就是測試解決方案,找出最有效的方法。

艾莉森.比爾德:你為各個組織所做的工作中,是否曾發現某些策略的效果真的最好?

約翰.畢謝爾:事實證明,繞過這兩個系統通常非常有效。原因是,當你使用這種策略時,就不需要人們採取特別的行動。因此,這是影響結果非常有力的方法。但在很多情況下,不可能繞過這兩個系統,於是組織必須選擇要啟動系統一,還是系統二。

有個因素很重要,在做這個選擇時應納入考慮,那就是你希望涉及其中的人,要運用多少認知能力。在某些情況下,也許你根本不希望他們運用任何認知能力,因為這會分散他們關注其他重要面向的能力。法蘭西絲卡,你與募款人的合作就是一例,你其實不希望他們運用任何認知能力,對吧?

法蘭西絲卡.吉諾:是的。這個專案是與華頓商學院的格蘭特合作,我們想了解,該如何激勵某家美國公立大學的募款人員。做這個工作經常會遭人拒絕,要保持強烈的動機很難。因此,我們決定啟動系統一,設法讓募款人員的情緒更加強烈。我們的做法是請一位主管四處走動,造訪不同的中心,感謝員工所做的工作。

所以我們能夠經由啟動系統一來提高情緒。在這種情形下,你可以想像選擇不同的策略。例如,你可以選擇啟動系統二。你會要求員工花更多時間思考打電話這件事,然後才開始打。或者你可以想像,提高為打這些電話的結果負起的責任。但我們決定不要啟動系統二,以免人們把大量精力花在認知資源上,因為這可能會降低他們的努力和動機。

相反的,我們啟動系統一來提高情緒,因而能夠增加他們的努力和動機,而他們在接下來幾週的表現變得更好,主動打了更多電話。

艾莉森.比爾德:簡單的一句謝謝,就能產生強大效果。

法蘭西絲卡.吉諾:是的,沒錯。

艾莉森.比爾德:這真的是很有趣的研究。謝謝你們的分享。

法蘭西絲卡.吉諾:謝謝邀請我們來。

(劉純佑譯)


Alison Beard: Hi. I'm Alison Beard an editor at HBR, and I'm here today with John Beshears and Francesca Gino, both professors at Harvard Business School, and authors of the article, Leaders as Decision Architects. Thank you both for coming in.

Francesca Gino: Thank you for having us.

John Beshears: Pleasure to be here.

Alison Beard: So you both spend a lot of time studying decision making, and how most of us – employees, managers, customers -- make the wrong ones. But you've developed an approach to structuring work that's supposed to help us solve that problem. So can you walk me through it?

Francesca Gino: Sure. The first step is to understand why all of us often make poor decisions. We rely too much on what that Daniel Kahneman calls System 1 thinking. System 1 thinking is emotional, intuitive, automatic. And relying too much on it leads you to all sorts of cognitive biases, like overconfidence, groupthink, loss aversion, and a preference for the status quo. But it's also possible for people to engage what Daniel Kahneman calls System 2 thinking more often. And System 2 is more deliberate, more logical, and slower. And it tells us when our intuitions are wrong, and when our emotions are clouding our judgments.

The second step is to define the problem. We recommand that leaders who are interested in using our approach think about three factors -- whether the problem is caused by something that has to do with human behavior, whether the problem can be narrowly defined, and whether people are acting against their best interest. And then, after defining the problem, the third step is to understand the underlying causes of it.

You should ask two questions -- whether the problem is mainly due to a lack of motivation, or insufficient motivation. So are people failing to act? Or whether the problem is due to the fact that people are taking actions, but they're taking the wrong one. For example, John worked with a retailer who was trying to cut some health care costs by getting these employees to sign up for mail prescriptions rather than picking up their prescription at the pharmacy.

John Beshears: Yes. That was a narrow problem where people were not acting in their best interests. What the employer realized was that when employees switch from in person pharmacy pick of maintenance prescriptions to home delivery of maintenance prescriptions -- and these are things like drugs for high cholesterol, high blood pressure -- actually both the company sponsoring the prescription drug plan and the employees would save money.

But many employees were not getting around to signing up for home delivery. And the reason was inertia -- a lack of motivation. So that's a case where people should have been doing something that was in their best interest, and helping the company along the way, but they simply weren't because of the problems that Daniel Kahneman has laid out for us.

Alison Beard: So what did the company do to change their behavior?

John Beshears: Right. That's step four of our process -- design the solution. In this case, what the company decided to do was get people to engage System 2. And the way it did that was by encouraging employees to make an active choice. So employees could decide to switch to home delivery of their maintenance prescriptions, or they could stick with in person pharmacy pick up. But either way, in order to take advantage of the benefits of the prescription drug plan, what employees had to do was actively indicate which of those two options they preferred.

They could stay with in person pharmacy pick up, but in order to do so, they needed to indicate that that was their choice by mail, or by making a phone call, or by going online. And the process wasn't designed to be onerous at all. It was a very simple, trivial process. But what it did was it caused employees to pause for a moment to reflect on whether they might want to switch to home delivery. And in fact, upon reflection, a lot of them decided that they did want to switch to home delivery. This very simple change led to a sixfold increase in the number of employees receiving maintenance prescriptions by mail.

Alison Beard: Wow.

John Beshears: Yeah. So I should emphasize that different situations call for different solutions. And in the example I was just describing, the employer decided to encourage employees to engage System 2. But another recommendation that we put forward is encouraging people to have System 1 triggered. And actually, Francesca has an example of that in some of her work with a company in India, where the solution was based on triggering System 1 by arousing emotions.

Francesca Gino: What we were trying to do in that case was working with the company, trying to address the problem of reducing high levels of turnover that we are experiencing in their call centers. So this is Wipro, and the experiment that we conducted was in their business process outsourcing call centers. And what we did was very simple. We conducted an experiment with two conditions during the onboarding process.

In one condition, new employees came in on their first day of orientation, and they basically had the time to think about their strengths, and how they could apply them to their jobs. And in another condition -- which was our control, if you will -- they didn't do that. They were not given the time for self reflection. And what we found is that the type of reflection really made the bond with the organization –the emotional bond--much stronger. And as a result of it, turnover ended up being much lower.

And we also were able to increase the performance of the employees, as well as job engagement. And we were able to replicate the findings in many other organizations by again working on this idea of engaging System 1 by creating a better bond -- better emotional bond -- with the company.

Alison Beard: So you mentioned control groups. And I know that testing is step five in your approach. So can you talk a little bit about why that's important?

John Beshears: Sure. We do, indeed, lay out a number of different possible solutions to various organizational problems. Francesca gave us a great example of triggering System 1, in this case by arousing emotions. But you could also consider harnessing biases or simplifying processes. And then there's the possibility of engaging System 2. One way you do that is the example I put forward, which was by encouraging reflection. But you could also prompt people to form plans. You could also try to inspire broader thinking on the part of people within your organization.

And then there's even a third possibility, which is bypassing both systems altogether. And you can do that either by setting wise defaults, or by introducing automatic adjustments into a process. But what we recommend is choosing one of those solutions, and then rigorously testing it against a control group to see which works better. And that is step five -- testing the solution to see what's most effective.

Alison Beard: So in all of the work that you've done across organizations, have you found that there are particular strategies that really work the best?

John Beshears: It turns out that bypassing both systems is often highly effective. And the reason is, when you follow that strategy, you are requiring no particular action on the part of the individuals involved. So that's a very powerful way to influence outcomes. But in many situations it's not feasible to bypass both systems, and then the organization is choosing between triggering System 1, and engaging System 2.

And one factor that is important to consider in that choice is the extent to which you want the individuals involved to be exerting a lot of cognitive energy. So in some cases, maybe you don't want them to be exerting any cognitive energy at all, because that detracts from their ability to pay attention to other important aspects of the situation. And actually, Francesca, your work with fundraisers was a case where actually you didn't want them to have to exert any cognitive energy, right?

Francesca Gino: Yeah. So that was a project in collaboration with Adam Grant of Wharton, where we were trying to understand how to motivate fund raisers at a US public university. And this is a job where you get rejected a lot, and keeping the motivation high is difficult. So what we decided to do is to trigger System 1 -- try to make their emotions a little bit stronger. And we did that by asking a director to walk around and visit the different centers, and basically express gratitude for the work that the employees were doing.

And so we were able to make emotions higher by triggering System 1. Now in that situation, you could imagine choosing a different strategy. For example, you could choose to trigger System 2. And so you would ask the employees to spend more time thinking about their calls before actually making the calls. Or you could imagine increasing accountability for the outcomes of those calls. But we decided not to have people spend a lot of energy in cognitive resources by triggering System 2, because that might be a case where you detract from their effort and their motivation.

And by triggering System 1 instead – raising the emotions –we were able to increase their effort and their motivation, and they ended up performing much better in the weeks after by making many more voluntary calls.

Alison Beard:So a simple thank you is pretty powerful.

Francesca Gino: Yes, it is.

Alison Beard:Terrific. Well, this is really interesting research. Thank you so much for sharing it with us.

Francesca Gino:Thank you for having us.



本篇文章主題決策