1 00:00:03,000 --> 00:00:07,499 RSAnimate 2 00:00:07,639 --> 00:00:10,265 www.theRSA.org 3 00:00:11,000 --> 00:00:14,022 Dan Pink, Drive 4 00:00:14,162 --> 00:00:17,182 Our motivations are unbelievably interesting, I mean... 5 00:00:17,322 --> 00:00:19,380 I've been working on this for a few years 6 00:00:19,520 --> 00:00:23,610 and I just find the topic still so amazingly engaging and interesting 7 00:00:23,750 --> 00:00:25,439 so I want to tell you about that. 8 00:00:25,579 --> 00:00:29,749 The science is really surprising. The science is a little bit freaky. 9 00:00:29,889 --> 00:00:32,026 OK? We are not as 10 00:00:32,166 --> 00:00:35,603 endlessly manipulable and as predictable as you would think. 11 00:00:35,743 --> 00:00:39,589 There's a whole set of unbelievably interesting studies. I want to give you two 12 00:00:39,729 --> 00:00:42,609 that call into question this idea that if you reward something 13 00:00:42,750 --> 00:00:44,515 you get more of the behavior you want. 14 00:00:44,655 --> 00:00:47,130 If you punish something, you get less of it. 15 00:00:47,270 --> 00:00:49,216 So let's go from London 16 00:00:49,356 --> 00:00:51,721 to the main streets of Cambridge, Massachusetts 17 00:00:51,861 --> 00:00:54,294 the northeastern part of the United States and let's talk about a 18 00:00:54,434 --> 00:00:57,281 study done at MIT Massachusetts Institute of Technology. 19 00:00:57,421 --> 00:00:59,756 Here's what they did: They took a whole group of students 20 00:00:59,896 --> 00:01:02,904 and they gave them a set of challenges. Things like 21 00:01:03,044 --> 00:01:05,312 memorizing strings of digits 22 00:01:05,452 --> 00:01:08,358 solving word puzzles, other kinds of spacial puzzles 23 00:01:08,498 --> 00:01:10,790 even physical tasks like throwing a ball through a hoop. 24 00:01:10,930 --> 00:01:12,860 OK, they gave them these challenges and they said 25 00:01:13,000 --> 00:01:15,006 to incentivize their performance 26 00:01:15,146 --> 00:01:17,604 they gave them 3 levels of rewards. OK? 27 00:01:17,744 --> 00:01:21,128 So if you did pretty well, you got a small monetary reward. 28 00:01:21,268 --> 00:01:24,497 If you did medium well, you got a medium monetary reward. 29 00:01:24,600 --> 00:01:27,214 And if you did really well, if you were one of the top performers 30 00:01:27,350 --> 00:01:29,223 you got a large cash prize. 31 00:01:29,390 --> 00:01:31,739 Ok, we've seen this movie before. 32 00:01:31,879 --> 00:01:35,090 This is essentially a typical motivation scheme within organizations 33 00:01:35,200 --> 00:01:37,800 right? We reward the very top performers 34 00:01:37,915 --> 00:01:41,610 we ignore the low performers and other folks in the middle. 35 00:01:41,750 --> 00:01:45,558 Ok, you get a little bit. So what happens? 36 00:01:45,698 --> 00:01:47,687 They do the test. They have these incentives. 37 00:01:47,827 --> 00:01:49,807 Here's what they found out. 1. 38 00:01:49,947 --> 00:01:54,000 As long as the task involved only mechanical skill 39 00:01:54,296 --> 00:01:56,136 bonuses worked as they would be expected 40 00:01:56,276 --> 00:01:58,535 the higher the pay, the better their performance. 41 00:01:58,675 --> 00:02:01,176 Ok, that makes sense, but here's what happens. 42 00:02:01,316 --> 00:02:06,400 But once the task calls for even rudimentary cognitive skill 43 00:02:06,560 --> 00:02:12,230 a larger reward led to poorer performance. Now this is strange, right? 44 00:02:12,370 --> 00:02:15,703 A larger reward led to poorer performance. How can that possibly be? 45 00:02:15,843 --> 00:02:17,650 Now what's interesting about this is that 46 00:02:17,810 --> 00:02:21,126 these folks here who did this are all economists: 47 00:02:21,266 --> 00:02:24,708 2 at MIT, 1 at the University of Chicago, 1 at Carnegie Melanie 48 00:02:24,848 --> 00:02:26,800 the top tier of the economics profession. 49 00:02:26,940 --> 00:02:29,650 And they're reaching this conclusion that seems contrary to 50 00:02:29,795 --> 00:02:31,768 what a lot of us learned in economics 51 00:02:31,908 --> 00:02:35,229 which is that the higher the reward, the better the performance. 52 00:02:35,369 --> 00:02:39,347 And they're saying that once you get above rudimentary cognitive skill 53 00:02:39,953 --> 00:02:41,964 it's the other way around 54 00:02:42,104 --> 00:02:45,658 which seems like the idea that these rewards don't work that way 55 00:02:45,798 --> 00:02:48,596 seems vaguely Left-Wing and Socialist, doesn't it? 56 00:02:48,736 --> 00:02:51,750 It's this kind of weird Socialist conspiracy. 57 00:02:52,601 --> 00:02:55,880 For those of you who have these conspiracy theories I want to point out 58 00:02:56,020 --> 00:02:57,965 the notoriously left-wing socialist group 59 00:02:58,105 --> 00:03:01,804 that financed the research: The Federal Reserve Bank. 60 00:03:01,944 --> 00:03:03,965 So this the mainstream of the mainstream 61 00:03:04,105 --> 00:03:06,650 coming to a conclusion that's quite surprising 62 00:03:06,790 --> 00:03:09,050 seems to defy the laws of behavioral physics. 63 00:03:09,190 --> 00:03:11,743 So this is strange, a strange funny. So what do they do? They say... 64 00:03:11,883 --> 00:03:14,287 This is freaky. Let's go test it somewhere else. 65 00:03:14,427 --> 00:03:16,834 Maybe that 50 dollars or 60 dollars prize 66 00:03:16,974 --> 00:03:19,976 isn't sufficiently motivating for an MIT student, right? 67 00:03:20,116 --> 00:03:22,410 So let's go to a place where 50 dollars is actually 68 00:03:22,550 --> 00:03:24,251 more significant relatively. 69 00:03:24,391 --> 00:03:27,189 So we take the experiment, we're going to Madurai, India. 70 00:03:27,329 --> 00:03:29,936 Rural India, where 50 dollars, 60 dollars 71 00:03:30,076 --> 00:03:32,512 whatever the number was, is actually a significant sum of money. 72 00:03:32,652 --> 00:03:35,617 So they replicated the experiment in India roughly as follows: 73 00:03:35,757 --> 00:03:39,658 Small rewards, the equivalent of 2 week's salary. 74 00:03:40,341 --> 00:03:43,769 I'm sorry, I mean low performance [received] 2 week's salary. 75 00:03:43,909 --> 00:03:46,500 Medium performance [received] about a month's salary. 76 00:03:47,857 --> 00:03:50,299 High performance [received] about 2 month's salary. 77 00:03:50,439 --> 00:03:52,447 Ok, so these are real good incentives 78 00:03:52,587 --> 00:03:54,850 so you're going to get a different result here. 79 00:03:54,990 --> 00:03:58,705 What happened though, was that the people offered the medium reward 80 00:03:58,845 --> 00:04:01,647 did no better than the people offered the small reward 81 00:04:01,787 --> 00:04:04,496 but this time around, the people offered the top reward 82 00:04:04,636 --> 00:04:07,629 they did worst of all. Higher incentives led to worse performance. 83 00:04:07,769 --> 00:04:10,708 What's interesting about this is that it actually isn't all that anomalous. 84 00:04:10,848 --> 00:04:14,330 This has been replicated over and over and over again by psychologists 85 00:04:14,470 --> 00:04:19,860 by sociologists and by economists, over and over and over again. 86 00:04:20,000 --> 00:04:23,197 For simple, straight-forward tasks, those kinds of incentives 87 00:04:23,337 --> 00:04:26,608 if you do this then you get that, they're great! 88 00:04:26,748 --> 00:04:28,642 With tasks that are an algorithmic 89 00:04:28,782 --> 00:04:32,038 set of rules where you have to just follow along and get a right answer 90 00:04:32,178 --> 00:04:35,700 "If-then" rewards, carrots and sticks, outstanding! 91 00:04:36,750 --> 00:04:39,168 But when the task gets more complicated 92 00:04:39,308 --> 00:04:41,866 when it requires some conceptual, creative thinking 93 00:04:42,006 --> 00:04:46,046 those kind of motivators demonstrably don't work. 94 00:04:46,186 --> 00:04:50,463 Fact: Money is a motivator, at work. But in a slightly strange way 95 00:04:50,603 --> 00:04:53,568 if you don't pay people enough they won't be motivated. 96 00:04:53,708 --> 00:04:55,812 What's curious about, there's another paradox here 97 00:04:55,952 --> 00:04:58,098 which is the best use of money as a motivator 98 00:04:58,238 --> 00:05:01,650 is to pay people enough to take the issue of money off the table. 99 00:05:01,790 --> 00:05:03,972 Pay people enough, so they are not thinking about money 100 00:05:04,112 --> 00:05:05,779 and they're thinking about the work. 101 00:05:05,919 --> 00:05:08,376 Now once you do that, it turns out there are 3 factors 102 00:05:08,516 --> 00:05:11,750 that the science shows, lead to better performance 103 00:05:12,122 --> 00:05:13,966 not to mention, personal satisfaction: 104 00:05:14,106 --> 00:05:16,461 autonomy, mastery, and purpose. 105 00:05:16,601 --> 00:05:18,989 Autonomy is our desire to be self-directed: 106 00:05:19,129 --> 00:05:20,667 to direct our own lives. 107 00:05:20,807 --> 00:05:23,922 Now in many ways, traditional methods of management run afoul of that. 108 00:05:24,062 --> 00:05:27,729 Management is great if you want compliance, but if you want engagement 109 00:05:27,869 --> 00:05:29,575 which is what we want in the workforce today 110 00:05:29,715 --> 00:05:32,049 as people are doing more complicated, sophisticated things 111 00:05:32,189 --> 00:05:34,101 self-direction is better. 112 00:05:34,541 --> 00:05:36,312 Let me give you some examples of this 113 00:05:36,452 --> 00:05:38,365 of the most radical forms of self-direction 114 00:05:38,505 --> 00:05:40,353 in the workplace, that lead to good results. 115 00:05:40,493 --> 00:05:42,759 Let's start with this company right here, Atlassian 116 00:05:42,899 --> 00:05:45,024 an Australian company. It's a software company 117 00:05:45,164 --> 00:05:46,899 and they do something really cool. 118 00:05:47,039 --> 00:05:50,943 Once a quarter on Thursday afternoon, they say to their developers 119 00:05:51,083 --> 00:05:53,940 "For the next 24 hours, you can work on anything you want. 120 00:05:54,080 --> 00:05:57,055 You can work at it the way you want. You can work at it with whomever you want. 121 00:05:57,195 --> 00:05:59,607 All we ask is that you show the results to the company 122 00:05:59,747 --> 00:06:01,404 at the end of those 24 hours." 123 00:06:01,544 --> 00:06:04,336 and this fun kind of meeting, not a star chamber session but 124 00:06:04,476 --> 00:06:08,285 this fun meeting with beer and cake and fun and other things like that. 125 00:06:08,425 --> 00:06:13,220 It turns out that one day of pure undiluted autonomy 126 00:06:13,360 --> 00:06:16,190 has led to a whole array of fixes for existing software 127 00:06:16,330 --> 00:06:18,877 a whole array of ideas for new products 128 00:06:19,017 --> 00:06:21,418 that otherwise have never emerged. One day. 129 00:06:21,558 --> 00:06:24,197 Now this is not an "if-then" incentive. This is not the sort of thing 130 00:06:24,337 --> 00:06:26,908 that I would have done 3 years ago before I knew this research. 131 00:06:27,048 --> 00:06:29,833 I would have said "You want people to be creative and innovative?" 132 00:06:29,973 --> 00:06:31,904 Give them a fricken innovation bonus. 133 00:06:32,044 --> 00:06:34,714 If you could do something cool, I'll give you 2,500 dollars. 134 00:06:34,854 --> 00:06:37,102 They're not doing this at all. They're essentially saying 135 00:06:37,242 --> 00:06:39,105 you probably want to do something interesting. 136 00:06:39,245 --> 00:06:41,082 Let me just get out of your way. 137 00:06:41,222 --> 00:06:43,918 One day of autonomy produces things that never emerge. 138 00:06:44,058 --> 00:06:47,277 Now let's talk about mastery. Mastery is our urge to get better at stuff. 139 00:06:47,417 --> 00:06:48,989 We like to get better at stuff. 140 00:06:49,129 --> 00:06:52,346 This is why people play musical instruments on the weekend. 141 00:06:52,486 --> 00:06:54,842 You have all these people who're acting in ways that 142 00:06:54,982 --> 00:06:56,641 seem irrational economically. 143 00:06:56,781 --> 00:06:58,718 They play musical instruments on weekends, why? 144 00:06:58,858 --> 00:07:00,758 It's not going get them a mate. It's not going to make them any money. 145 00:07:00,898 --> 00:07:03,996 Why are they doing it? Because it's fun. 146 00:07:04,136 --> 00:07:06,223 Because you get better at it, and that's satisfying. 147 00:07:06,363 --> 00:07:07,987 Go back in time a little bit. 148 00:07:08,127 --> 00:07:11,128 I imagine this: If I went to my first economic's professor 149 00:07:11,268 --> 00:07:13,289 a woman named Mary Alice Shulman. 150 00:07:13,429 --> 00:07:16,816 And I went to her in 1983, and said 151 00:07:17,083 --> 00:07:20,179 "Professor Shulman, can I talk to you after class for a moment?" "Yeah." 152 00:07:20,319 --> 00:07:23,175 "I've got this inkling. I've got this idea for a business model. 153 00:07:23,315 --> 00:07:25,057 I just want to run it past to you. 154 00:07:25,197 --> 00:07:26,953 Here's how it would work: 155 00:07:27,546 --> 00:07:30,655 You get a bunch of people around the world 156 00:07:30,795 --> 00:07:32,822 who are doing highly skilled work 157 00:07:32,962 --> 00:07:35,333 but they're willing to do it for free 158 00:07:35,473 --> 00:07:38,500 and volunteer their time 20, sometimes 30 hours a week." 159 00:07:39,894 --> 00:07:41,911 Ok, she's looking at you somewhat skeptically there. 160 00:07:42,051 --> 00:07:43,705 "Oh, but I'm not done. 161 00:07:43,845 --> 00:07:48,603 And then, what they create, they give it away, rather than sell it. 162 00:07:49,055 --> 00:07:50,716 It's going to be huge." 163 00:07:51,500 --> 00:07:54,174 And she truly would have thought I was insane. 164 00:07:54,314 --> 00:07:55,912 All right, you seem to fly in the face of so many things 165 00:07:56,052 --> 00:07:58,381 but what do you have? You have Linux, powering 166 00:07:58,521 --> 00:08:01,251 1 out of 4 corporate servers and Fortune 500 companies. 167 00:08:01,391 --> 00:08:05,939 Apache, powering more than the majority of web servers. 168 00:08:06,079 --> 00:08:09,260 Wikipedia...What's going on? Why are people doing this? 169 00:08:09,400 --> 00:08:12,282 Why are these people, many of whom are technically sophisticated 170 00:08:12,422 --> 00:08:15,435 highly skilled people who have jobs, ok? 171 00:08:15,575 --> 00:08:18,326 They have jobs! They're working at jobs for pay 172 00:08:18,466 --> 00:08:22,413 doing challenging, sophisticated, technological work. 173 00:08:22,553 --> 00:08:25,871 And yet, during their limited discretionary time 174 00:08:26,550 --> 00:08:29,587 they do equally, if not more, technically sophisticated work 175 00:08:29,727 --> 00:08:33,455 not for their employer, but for someone else for free! 176 00:08:34,230 --> 00:08:36,302 That's a strange economic behavior. 177 00:08:36,442 --> 00:08:38,599 Economists who look into it "Why are they doing this?" 178 00:08:38,739 --> 00:08:41,868 It's overwhelmingly clear: Challenge in mastery 179 00:08:42,008 --> 00:08:44,172 along with making a contribution, that's it. 180 00:08:44,312 --> 00:08:46,726 What you see more and more is a rise of what you might call 181 00:08:46,866 --> 00:08:50,410 the purpose motive. It's that more and more organizations 182 00:08:50,550 --> 00:08:52,871 want to have some kind of transcendent purpose 183 00:08:53,011 --> 00:08:54,892 partly because it makes coming to work better 184 00:08:55,032 --> 00:08:58,000 partly because that's the way to get better talent. 185 00:08:58,640 --> 00:09:00,831 And what we're seeing now is, in some ways 186 00:09:00,971 --> 00:09:04,710 when the profit motive becomes unmoored from the purpose motive 187 00:09:04,850 --> 00:09:07,662 bad things happen. Bad things ethically sometimes 188 00:09:07,802 --> 00:09:11,130 but also bad things just like, not good stuff: 189 00:09:11,270 --> 00:09:13,239 like crappy products 190 00:09:13,379 --> 00:09:17,447 like lame services, like uninspiring places to work. 191 00:09:17,587 --> 00:09:20,204 That when the profit motive is paramount 192 00:09:20,344 --> 00:09:23,578 or when it becomes completely unhitched from the purpose motive 193 00:09:23,718 --> 00:09:26,169 people don't do great things. 194 00:09:26,309 --> 00:09:28,411 More and more organizations are realizing this 195 00:09:28,551 --> 00:09:30,501 and sort of disturbing the categories between 196 00:09:30,641 --> 00:09:32,242 what's profit and what's purpose. 197 00:09:32,382 --> 00:09:35,579 And I think that actually heralds something interesting. 198 00:09:35,719 --> 00:09:38,004 And I think that the companies, organizations that are flourishing 199 00:09:38,144 --> 00:09:40,464 whether they're profit, for-profit or somewhere in-between 200 00:09:40,604 --> 00:09:44,014 are animated by this purpose. Let me give you a couple of examples. 201 00:09:44,154 --> 00:09:46,332 Here's the founder of Skype. 202 00:09:46,800 --> 00:09:48,988 He says our goal is to be disruptive but in the cause of 203 00:09:49,128 --> 00:09:51,280 making the world a better place. 204 00:09:51,420 --> 00:09:53,119 Pretty good purpose. 205 00:09:53,259 --> 00:09:55,012 Here's Steve Jobs. 206 00:09:55,152 --> 00:09:57,133 "I want to put a Ding in the universe." 207 00:09:57,273 --> 00:09:59,651 All right? That's the kind of thing that might get you up 208 00:09:59,791 --> 00:10:02,680 in the morning, racing to go to work. So I think that 209 00:10:02,820 --> 00:10:07,082 we are purpose maximizers, not only profit-maximizers. 210 00:10:07,222 --> 00:10:10,858 I think that the science shows that we care about mastery very, very deeply. 211 00:10:10,998 --> 00:10:13,270 And the science shows that we want to be self-directed. 212 00:10:13,410 --> 00:10:16,242 And I think that the big take-away here is that 213 00:10:16,382 --> 00:10:18,667 if we start treating people like people 214 00:10:18,807 --> 00:10:21,376 and not assuming that they're simply horses 215 00:10:21,516 --> 00:10:24,246 you know, slower, smaller, better-smelling horses 216 00:10:24,386 --> 00:10:26,334 if we get past this kind of ideology of 217 00:10:26,474 --> 00:10:28,473 "carrots and sticks" and look at the science 218 00:10:28,613 --> 00:10:30,497 I think we can actually build organizations 219 00:10:30,637 --> 00:10:33,538 and work lives that make us better off but I also think 220 00:10:33,678 --> 00:10:38,986 they have the promise to make our world just a little bit better.