Judul : bed bath bey
link : bed bath bey
bed bath bey
[music] it's now my, my, my pleasure and my honor to introduce today's speaker and mycolleague hayagreeva rao. hayagreeva is the ethel mcbean professor of organizational behavior and humanresources. here at the gsb he's also a professor of sociology at the schools of humanityand science. he's also the academic director of afantastic online program. our first online program the stanfordinnovation and entrepreneurship program.
this is a program that we built jointlywith out engineering school. it has 12 modules, they built half a dozenwe built half a dozen. and it's available online to to people whowho don't have the time or the ability to attend one of ourin person classes. and the attraction we've gotten from ithas, has significantly exceeded actually whatwe've, what we have planned, so. i, i thank hagi for his leadership inthat. he's also the corrector of the customfocus innovations executive program and the corrector of the advancedleadership program.
for asian american executives. he has been working on on on, on thechristian of scaling together with bob sutton, hehas a book on the topic, which i'm sure he's going totell you about, and i think you're in for areal treat. it's been a best seller, it's one of the top 10 books listed by forbes and thewashington post. please join me in giving a warm welcome tohuggie ralph. >> [sound].
>> good evening everybody. let me begin by first saying thank you certainly on behalf of my faculty,colleagues it's just impressive to see so many out of staters and so many people from overseasas well. coming here your affection for the gsp, ofcourse matters a lot to us, because we can, of course, continue tobuild on our traditions of excellence. and do and, and you know get to greaterheights. so i'm just actually begin this talk witha little bit of a confession.
i finished business school in india in1980. and the astonishing thing was as i was finishing business school in india thetextbooks that we actually followed i think the one infinance, if i recall, was by jim vanhorne. if you can believe it. you know the one in accounting was bychuck horngren. and we even had jeff pfeffer's a book onthe external control of organizations. and in 1980 when i graduated i had no ideathat all of them would eventually turn out tobe my colleagues.
it is truly sort of extraordinary in thatsense. and the other funny thing was in 1981 tompeters an alum of the gsb, of course, wrote a remarkable book called in searchof excellence and if at that time if you had told me i wouldhave actually written a book on excellence i would have said lookyou are out of your mind. but the amazing thing is that we did andtom has actually been a great supporter. and we're all very grateful to him. the books done well as garth mentioned. you know, it's nice to know sort to be onthe wall street journal
best seller list or whatever, but really,what touches your heart is we did a little thing on npr and there was thislittle there was this lady from rugle georgia who actually emailed me andsaid, you know, i heard you on npr. i ran down and bought the book. and we've created a little learning groupin rural georgia to scale excellence in ourschools, there. that sort of really melted my heart, youknow? and the reason it melted my heart is, mymom was a high school math teacher. you know.
she was, i guess, a tiger mom before that, that particular label kinda gotconstructed or invented. so, what i'm gonna do is you know, share a few ideas in the hope that we can have aconversation. i'd like to leave a bunch of time asidefor questions. in many ways this book and this whole adventure with bob sutton would never havebeen possible. without you know a program that[inaudible] to call the customer focus innovationprogram.
so what we do is we get 60 people, and themorning is all what we call clean models, social science, and the afternoon is allabout dirty hands, or design thinking. and the participants love it, so they come in the morning, and they come in onsunday. mornings you know, they're being exposedto social science and then, every afternoon,they have to solve an, a problem, a challenge, ifyou will, for some company or the other. and, seven years ago, we were doing thatwith jetblue. and, in fact the participants did somepretty amazing sort of things.
you know, you go and, you know, all the design thinking ideas that many of you i'msure, are. familiar with. so, at the end of the program, one of theparticipants asked us, how do i take this back and scale itin my company? so, i gave an answer, and my colleague bobgave an answer, and that evening as we weredebriefing in menlo park, we live there, over a nice glass of pinotnoir you know, i said, bob, what did you think ofyour answer?
and he said, what do you think of yours? and i said mine was truly sad, i said. it was really bad, you know. and i'm very disturbed about it because ijust couldn't say anything that really was, you know, interesting or challenging,or provocative or what have you. so i asked bob, what do you think, bob? and bob says, mine sucked too. [laugh]and so, you know, few more glasses later so we, we said, what are we gonna do aboutit?
and you know, we did something thatstanford professors do you know, with. great support and encouragement fromeverybody. and that is, if you really don't know something, and you're quite ignorantabout it, the best thing to do is to actually dreamup a course and offer it. [laugh] so, when you dream up a course and offerit, a couple of things happen. the first thing is, you are supermotivated, cuz you're clueless. you have to read a lot, you've gotta puttogether something.
and the students they, you know, theysense your excitement and your curiosity, they kind of engage with youand they teach you a lot. and, so we were kind of thinking, how do we teach people about how to scalesomething? and bob and i thought, let's give them achallenging, hard problem. so, we had 60 students, we could havegotten more but we capped it at 60. 30 from engineering and 30 from gsb. and then we put them into teams andrandomly assigned dorms and spokes team on the stanford campus to eachof these student teams.
and we give them one scaling challenge. scale to use of bicycle helmets bybicyclists on stanford campus. it's a very hard thing to do by the way. because, you know, there's memory loss andparticularly the undergrads are fully pursuant of their immortality andinvincibility [laugh] and it was very, very hard. but the students came up with like,amazing ideas. you know, so one team i remember they weregiven ev, they scandled a village with these youknow, grad students.
and so they kinda went to the gradstudents and try to persuade them and facts and, you know, here's whathappens when you fall of a bike. none of these things worked. and finally these people said wait aminute. we really ought a go to the elementaryschool on the stanford campus. because a lot of the kids of grad studentsactually go there. let's actually organize a campaign amongstthese kids. and what they did was, they actually gotthe kids to write little notes for their parents,saying hey, wear a helmet.
it was amazing. and by the way, the way we graded thecourse was, we knew what the base rate of helmet usage was, so the more you moved the needle, the better the grade,you know? which caused some interesting challenges,i might add. [laugh] and so there was another team thatwent to the men's soccer, team. so, they told him look, you know, if youfall off a bike, bad things can happen, you know, you're telling the men's soccerteam, they just said give me a break, you know.
but, then, they quickly realized, wait a minute, we need to communicate this tothem in an interesting way, and they came with the idea of a watermelon smashingcompetition. so they would go there, and tell thesepeople after they dropped a water melon, that's your brain when youfall off a bike. and people said, really. and they said, you know, you may thinkyou're immortal but if that happens to you, think of what's gonnahappen to your team.
and that got the attention of the men'ssoccer team. and very quickly you know, their penaltyif you will, if. any of the men's soccer team didn't wear a bike helmet, they'd actually fling thewatermelon at the kid. so, you know, that kind of deviantbehavior got punished. it sort of spread for awhile. and the second year we actually did a project with the mbas and the engineersagain, and this time they were actually working withfront
line engineers and soldiers and that kindof stuff. it's unbelievable. i mean when you sort of see we actuallyhad a general in the rapid deployment force who came to usand said hey can they help. and we said oh, what a great problem. and so we once again told our students togo figure these things out. and the sorts of things they came up withwere, like, simple but profound too. one, just one example. [cough]
so soldiers, of course, get redeployed,and you get rotated and you have replacements. and you have interesting transfer ofknowledge problem. how do you transfer knowledge fromoutgoing cohort of soldiers. to an incoming cohort. the u.s army's modus operandi is, youactually have a two hundred item questionnaire that youhave to fill out. and, you can imagine what people are goingto do when you give a two hundred itemquestionnaire and that's
given to you right after you land on theair base in the u.s. when you've come backfrom kandahar. you know, the response rate is abysmallylow. so, there's just no way you're going toget anything. plus it goes to some place in thepentagon. they do some statistical analyses and theydon't turn out to be much either. so our kids actually thought about forawhile and they came up with like a, one interesting idea that they came upwith for example was a look, why don't we. ask each soldier.
look, can you take ten minutes to talk tosomeone who is going to replace you. tell them three stories that actually areinteresting examples of your experience. and if you spend ten minutes, we'll makesure you get half an hour more web access when you talkto your family. but, by the way, it's a very high poweredincentive, as you can image. web access is a very scarce resource, yougive them half an hour, and you give these stories, and it's amazing whenthese people would tell stories to the incomingsoldiers. what they were doing is they were actuallygiving way
more data can, than can be accommodated bya regression model. you know you sort of say you know, sunnydays, sun on the arise, the enemies in the hills orwhat ever. just look at the number of variables thatare being incorporated in that story. and it is very officially packaged. so the army is actually trying it out. so any way after many years of trying itout we sort of thought you know. we've actually learned, and we intervieweda number of people we wrote case studies, we did a bunch ofthings.
so six years after the original question, we actually thought okay let's write thisbook. and first a word about excellence. you know, we could debate this to whatexcellence is you know. plato says this or socrates says this. if your interested in eastern philosophythe gita says this. but bob and i are social scientist and forus excellence simply mean people do the right job even when nobodyis looking over your shoulder. that's what excellence is.
because when you need people to monitorothers executing a task. that's not excellence. you know, for every ten people whom you hire, you can't have three copsmonitoring. there is no way you can scale if you dothat. cuz it's just you know, going to be hugein bureaucratic and expenses. so what's the problem of excellence? there are two dimensions to the problem. on the one hand.
you, if you're in a large corporation, oreven a small one, you might have a pocket ofgoodness, somewhere. and what your challenge as a manager is,or as an executive or leader is, is how do i actually constructa dominant chain of goodness? how do i spread goodness from one place tothe others? the challenge is when you do that you havevoltage loss. this dilution. there's kind of another dimension ofscaling up accidents and that is if you're a small organization and as you're adding
people how do you actually scale upaccidents. so, you know, wonders if you willhorizontal or cross divisional or cross business unit, theothers, kinda, much more temporal. there's also a little problem and thelittle problem is, the larger an organization gets, the more likely aresmart people, inside the organization, to becomedumber. and i don't say this to be coy. and what i mean by beep smart peoplebecoming dumber is, i don't mean to imply, obviously, that they become stupid,but i
certainly mean to imply that they becomesilent. and that's the problem in a largeorganization. if they become silent. you're basically in trouble. so lots of examples of scaling you know, a lovely little start up called pulse, youknow. they would always, they were two stanfordgrads, and they would call us and say, we have a scaling problem, i'dsay like, what kind of problem? there's just the two of us, we've hiredfour more people, we have scaling problem.
i'm like, you guys are all working in oneroom, how could that be a scaling problem? but sure enough you realize they needed to coordinate, they needed to set acommunicate. and on the other extreme, you have the institute for health improvement incambridge. they want to reduce debts caused bypreventable medical errors. 100,000 people a year die. how do you do that when you don't havecloud, when you don't have the quality? when you don't have resources, and youhave 3,500 hospitals?
so that gives you if you will, the breadthof the. scaling challenge. so, the, the main sort of point i wanna i'd suggest is. scaling up is deceptively easy butscrewing up is easier. and the term, the scientific term bob andi use is, we call it a clusterfug. and we can't of course use the actualword. and so what we did was we, suddenly remembered, wait a minute, normanmailer wrote about fug successfully, and so we thoughtlet's
actually exploit this literally lineage ofthe term, so we called it clusterfug, and writingthe case as we speak, on the navy seals, yeah i know, can you imagine a guy like meriding a case on navy seals. i know, even i've chuckled about theabsurdity of this. so, i spend [inaudible] we actually havetwo amazing students who are helping me writethis case. carter [unknown], who's a navy sealhimself, and gib lopez, who's actually an mba also. both of them are helping me write thiscase, and carter and i went and.
spent a chunk of time at coronado. in, but i was with the navy sealcommander. you know they use a lot of abbreviations. and there was one abbreviation that kindof baffled me. they called it pfm. and after awhile i sort of had pretty goodidea of what the f stood for. i'm sure all of you know that. it was the p and the m that i was actuallyutterly befuddled by. and later when i asked the commandingofficer, i said what is pfm?
he said, pure, you of course know what thef is, and the m, of course, means madness. he said, that's what happens to us when wetry and scale sometimes, pfm. now, if you really look at, 30 years ofsocial psychology, why is it that scaling ends upin a clusterfug? and this could be any decision. new product, new technologyimplementation, an acquisition, expansion overseas, whatever the articulation orrealization of scaling. it turns into a clusterfug because ofthree things. the first is.
sense of illusion. you actually have illusion because thesmart people have become silent. you're not getting any feedback and it'skind of like an echo chamber. impatience. since you don't get feedback, you getimpatient, and then of course, you have the smart people becoming silent,the incompetent are the only ones you rely on, and so when you put the threeof them together, you get a trifecta, and you havea spectacular kind of clusterfug. and you can quickly see, actually, manyorganizations, teams
actually taking the off-ramp to aclusterfug pretty quickly. now, you know, what do you do? so, i love reading history. the greek historian herodotus says. the persians had an interesting decisionprotocol to prevent illusion, impatience, and incompetence,and apparently what they would do is, they would debate an issue drunk oneday and the next day, they would re-debate thesame issue sober. [laugh] now, you might ask, what is thevalue of such a decision protocol, and,
put simply, when people are drunk, theyput a lot of information on the table. [laugh] but their ability to derive aconclusion is debatable. but, on the other hand if you're soberpeople are strategic. so they don't put a lot of information onthe table. however even from the thinnest [unknown]. you can certainly construct an influenceif you will. now, our recommendation is not that youfollow what the persians did. which is, you know, debate an issue drunkand, you know, then rebate it sober the next day, one ofthe things we
started to talk about in the book is toactually rely on something called a pre-mortem, which isthe opposite of a post-mortem. so, for instance, if the six of you werein the company and you were thinking of, like,expansion, or ipu, or whatever the decision outcome, i would randomlyassign three people to the failure condition and say, imagine the event hasoccurred, engage in time travel. and then write a story of how that failurehappened. write it just in one page. remember it has to be a story not a list,and the three people have
to write that anonymously andindependently cuz if you, they collaborate, you'll havecorrelated error. so on the other side, same thing, you givethem, of course,. the scenario of success. imagine the thing has succeeded. write a story of how that succeeded. and, what is amazing is, when you getthese stories. it is incredible what [inaudible] youknow? when you write, when people write a story.
what they're doing is, they're actuallygiving you, if you will. you know, a state transition model. you have various transition points, youactually can see the downward spiral of a debacle, you can see the upward spiral of success,and then you can quickly sort of debug. by way of example let me illustrate, ijust recently did a version of this with the leadership ofthe stanford medical school. they really wanna go big guns, createsomething called. stanford medicine, where basic science hasto actually work with
the clinical hospitals. and then, of course, the clinicalinvestigators and you gotta work with the hospitals. it's amazing the stories of failure peoplewrote. you know? one person knew about side-effects, theyactually sent it as a pdf file, but their electronic patientrecords don't accept pdfs. so the person who knew about the side-effects, sadly told this toeverybody.
but nobody knew that. another person said, "oh my god, we don'thave nurses to run these trials." . another person would say wait a minute. you know, we the hospital is taking careof somebody in a trial. this person doesn't know the basic signs. they've misdiagnosed the cause of failure. you have an action of a lawsuit. so things like this when they wereactually put on the
table the very smart people, of course, inthe medical school faculty. they quickly said, "oh my god, we'vereally got to get this going." and very quickly they actually thought oflots of possibilities and began to put a lot of checks inmeasure. so just an example, so, think of. premortems if you will, the contemporaryequivalent to what the persians kinda do. so, what's the core message? why is it people take the off-ramp to aclusterfuck? and the answer is, usually when you askpeople what is
scaling, they invariably think of it interms of footprint-related metrics. and, you know, it's perfectly sensible. they're very important. but the catch is. if you ask people what's desirable socialpsychology tells us they actually think about this in a veryabstract way. that's why they love the metrics. you know, so many customers, so many this,so many that, et cetera. they're not really thinking about themindset.
the mindset is actually what allows thefootprint to survive and thrive. one example, people really thinking aboutthe mindset. we recently had founders of a companycalled [unknown] they do mapping software,they got acquired bygoogle for a couple of billion. and he was actually in class and he wasgiving a fascinating example of focusing really on the mindset, and [inaudible] a number of other importanttruths. so they get 30 million you know, you,in anearlier phase of the revolution, they get 30 million dollars,
and the [unknown] are telling them,"scale!" you got 100 people, hire 400 more. and the employees also expect head countto go up, the company to grow, you know, allthese things. and the founders start to think about it,and they thought about it, and then after. they made a decision and then they calledall the employees of the firm, and they announceda hiring freeze. people were stunned. the employees said, are you out of your
mind, we just got 30 million dollars,they're telling you to scale, you know, is this a vote ofno confidence in the future of the firm? why a hiring freeze now? it seems insane. and the founders stood up and said look,we had 100 people we're gonna hire 400 smart people, all ofthem have great ideas. the problem is when you hire 400 smartpeople and they come to you, i have to say no 400times. because we have to focus.
we can't just kind of get everybody's ideainto the mix, i've got to say no. and i don't want to create 400 demotivatedpeople in one fell swoop. and he said, the smart people that we hire, they're smart, but they're also overconfident. they think they can hit the ground runningon day one. they can't. they need to take three weeks to getready. if i hire 400 people in one step, i've actually altered dramatically the ratio ofunproductive to productive people.
and then what am i going to do as a ceo? i'm going to spend time thinking about how to make the produ, unproductive peopleproductive. and i don't want to do that. the real number we need to move is onemetric, and that is, the three month user re userretention metric. we have users, there's a lot of churn, andthe real test of our ability is, how many of them do we retainover a three month period? and so, one of the employees then said,well, what's my job?
and you know he said, well, your job issimple. in fact, the job of everyone of us, isvery simple. everyday, we've got to ask, what did i dotoday, to move that number? think about it. and this story is also particularlyresonant with me, because what it sorta [inaudible] to you is,scaling is actually about. when you scale an organization, you reallygot to figure out that it not just mindless addition,it's actually smart subtraction. in fact, when you're in the business ofscaling
excellence, you really are like a race cardriver. you've got to figure out when toaccelerate but, also, when to break and when to change thegears. and that's what they figured out. and of course the company succeededsubsequently. and you quickly realize if peoplemindlessly and individuals, you're definitely taking the on ramp to aclusterfug. so, you know, i'm going to sorta of moveon. and the problem of course with scaling isif you don't do this well it's dilution.
here's a confessional, a little memo fromhoward schultz of starbucks saying as we move from 13 hundred to 13 thousand, we ofcourse lost sight of the customer. and the constant is, in all the excellentorganizations we visited, the executives invariably see theglass as half empty. and nowhere did we find somebody who sorta says you know, we're doing great and youknow. you, you just don't find that at all, theyall assume, my god, we gotta do all these things all the time, and that's kind of that relentlessrestlessness.
so, i'm gonna give you a couple of biglessons, maybe a choice point or two, and a quick overview of someof the main principles. we, obviously won't be able to dive in toodeep, because we, we, we end at 6:15, and i wanna makesure. there's some time for q and a. so one of the big lessons is scaling upexcellence isn't just an air war. it's actually a ground war. so what are the characteristics of an airwar? when you think of an air war what do youthink about?
what's that? >> [inaudible]>> high level. abstract. no details. absolutely. what else? sorry, top down absolutely. huh? >> collateral damage.
>> collateral damage. >> uh,indeed collateral damage canactually be very high there's a fascinating historicaccount of the british bombing of germany and theyfound out that what is it? [laugh] twenty percent of the ordnancelanded within five miles of the intended target. [laugh] yeah, the rest of it was like allover the place. that tells you something about collateraldamage. what about a ground war?
were [crosstalk] very personal? >> more intense? >> more intense. >> actual engagement. so, when we go to companies, you canquickly find out who is thinking about. and in thinking in just an air war is notenough. i'm not saying air wars are bad obviouslyyou need a little bit of both. but what you want to think about is youreally want to focus on the ground war. and the reason is the following, becauseremember when
you think air war, you're only thinkingvery abstractly. when you thinking ground war you'rethinking very, very concretely. so for us the compelling distinction wefound is when i go to companies i ask people hey how do you think you need toget all these people to do something? and the people who are in an air war mode instantly leap to incentive basedreasoning. we just need to figure out how to reward them if they do or punish them if theydon't. that's is that's all we need to do.
but the people who think of it in a groundwar term, they actually sort of think, you know,motivation isn't the problem, tools are the problem. the question is what kinda tools can wegive these people to make them succeed. very, very different. so, i'll give you a little illustration of how powerfully somebody thought of aground war literally. in the context of a war and that was ageneral called matthew ridgeway. now this was actually in the korean war,which is
for some strange reason not as you know,given the kind of attention, because some of the mostbrilliant acts of leadership really in the 20th centuryoccurred, in the, korean war. so imagine. you know, [unknown] is there in tokyo andyou know, he's consumed by sensible delusion and delusionand then truman fires him. and you're ridgeway. you're 42 years old. you've never fought in korea.
you're in charge of like 500,000 men. who simply being beaten, bruised,battered, basically by the chinese first army. you can't go and get youtube's. you can't give people a speech and sayjeez let's kinda go and do something, you can't doall that. but look at what ridgeway does. the first thing he does is, he actually,you know, he says. i, i need to understand the topography, ofkorea.
so he goes to on a bomber and there's a navigator and he tells the navigator can itake your seat? i want to see what the topography is. and he realizes oh my god, you know, lotsof hills and ridges and rivers and then he meets these regimental commanders and hesays do you know the name of the river close by toyou? i said, do you know how the deep the wateris? no.
you're fired, because you're puttingpeople in harm's way, you have no idea. apparently what the pentagon had done is,there were a lot of people in the pentagon who apparently weren't given leadership opportunities in worldwar two. they moved a lot of files and somebodythought, you know. we really need to kind of give themleadership opportunities so let's send them to koreaas regimental commanders. when the heaviest thing you've moved if afile between two desks, it's not exactly the right kind of personto send to korea.
and so the things that he sorta does to give you a sense, no speeches, none ofthese things. but when he travels, no fighter escort,most shot at general, in the 20th century. he says, my men are getting shot. why do i need a fighter escort? every time he goes and visits troops, it'sthe cold korean winter. the only thing he has in his jeep, amplesupply of gloves. he gives gloves to soldiers and say, youcan hold your carbine. so he goes to a forward operating post.
and he looks at a map. and, you know, blue pins us, and redcommunist forces. so if you're ridgeway. you got 20 men in the room. what's the question you're going to ask? what's the diagnostic question you'regoing to ask? imagine. it's clear. you know where the us forces are, wherethe communist forces are.
there are 20 men sitting there. you're the general. what's the, if you had one question to askto diagnose what is going on, what would thatquestion be? hm? come on, somebody take a guess. come on. >> [inaudible]>> what do you need, that's good. which direction, that's good.
wind. how many people? they're all very legitimate questions, butthe question he asks is. how old is the information on which thepen positions are based on? and the men look at him and say, they're15 days old. what does that tell you? obviously, it's obsolete. but it also tells you the men are afraid. they don't want to go on a foot patrol.
because they're afraid of gettingambushed. that's the reason you're not getting highquality information. and, that's when he tells people, look. i need information that's three hoursfresh, we're going to support you in this way, we're going to dothis, etcetera. and, it's these same people who wererefusing to go, as it were, on a foot patrol, the very samepeople push. the chinese and not going in communistforces to the 36th parallel. there's actually a lovely story of a crazymarine general
whose unit is surrounded on all sides andthe men tell him, the enemy is in front of us,behind us, to the left, to the right, what do wedo? and this guy is slightly maniacal, so hesays, wonderful. they've got nowhere to run. [laugh]you've gotta be crazy to think like that. they actually punched through thecorridor. so here's you know, this is what one mckinsey consultant told us, the realthing about the
ground war is, you gotta get a thousandpeople to move one foot forward at the same time. in an air war, obviously, one person cango a thousand feet forward. so, the second thing is, scaling is a lotabout felt accountability. everybody talks of people feeling a senseof ownership, and that's true. you certainly need to feel you own theplace. but if that's the only thing you do, thatcan be dangerous. you can quickly begin to think ofyourself. it can get very swiftly egocentric.
and so what we find out about felt accountability is you really need not onlythe feeling that you own the place, but youalso need the feeling the place owns you. so you need actually the pull of obligation, and then the push ofcommittment. both of those are needed. and in the, one of the scps, there are acouple of scpers sitting here as well. i was actually talking about thisinteresting case study. anybody remembers this?
what's the hotel on the, what's thebuilding on the left? i gave the game away by. the taj, exactly. november 26 you had terrorists attackingthe hotel. and, you know? you had employees like these young ladieson the left, doing things. and, what's amazing is, you know? when you have a terrorist attack, what doyou expect employees to do? head for the exit, cuz they know where theexits are.
and that day there was no command andcontrol, the general manager is rushing in from somewhere, canonly get to the lobby. his family is being consumed by fire inthe 35th or the 36th floor. nobody's giving orders, nothing,everything is in disarray. well, what was amazing is young ladieslike this who were actually overseeing the reception forthe outgoing ceo of unilever. you know, in a conference room quicklybarricaded the room, made sure that people were hydrated, the wiveswere separated from the husbands. cooks escorted people down the stairs.
a lot of cooks lost their lives. because they gave their bodies as shields. and later, when they asked some of theseemployees, you guys were heroic. they said no, we actually were doing ourjob. what's their job? their job actually is not to be anambassador of the taj to a customer, but to actually be anadvocate of the customer. the taj actually hires very carefully,though. you obviously can't do that in the unitedstates for legal reasons.
but one of the things they do is the never hire people who are usually from the majorcities of india. they always hire people who are from thesmaller cities of india. because the tradition of [inaudible] beinggod is actually quite vibrant. in the bigger cities, modernization hasactually kind of assaulted all of those norms. they do a whole bunch of other things. as i was in this sep class, talking about, there's one of these executives made avery sensible point.
and he said, professor robert, surely thisis a case of crisis. you obviously expect your employees torally around this. you know, isn't the real test what the tajdoes every day? and i was about to say something. we actually had another participant whowas a senior executive at [inaudible] guy calledjohn thomas. he said, let me tell you about myexperience. he said, i flew with my wife and threeyear old to india. jet lag tired, the kid's crying, we're allornery,
the taj sends a car to pick us up, on route the baby is crying and i'm thinking,oh, my god, we gotta go through checking,check-in delay and everything. they come to the taj and somebody meets them and says, you've already been checkedin. they take them to the room. they open the doors and you know there's alittle flask of milk and a bunch of cookies for thethree year old. so john looks at them and says, how didyou know?
and the guy says, when you got in the caryour baby was crying, the driver called the front deskand said, baby on board, jet lagged. expedite. >> [laugh] >> now isn't that an amazing case ofaccountability, felt accountability? that's what felt accountability is. and by the way, the driver isn't an employee of the taj, but a contractualindividual. so here's this one woman called fifi.
and when we did a study of wyeth. you know the one of the wyatt plants, whatdid they do? they actually thought of equipment changes as breaks. so, mentally, when you categorizesomething as a break, you're kind of like, loose and easy, and you don't,there's no sense of urgency. but these people sort of said "wait aminute, it's not a break" that we've got organize ourselvesas though it's a pit stop. just that person and her team members andlook at
sort of what they did, in terms of, youknow, not just the number of syringe production,but how they cut down the changeover time and the likeand so forth. he's a quick example of a united pilot. saying i used to do all these things. and now, what do i do? i gave up. and i'm an eagle scout, and i'm anentrepreneur. so i'm gonna be very quick here, and talkabout a couple of ideas.
so, the most important choice point iswhen you scale. do you want more, or do you want them tobe better? so that's kind of one important choicepoint. the second choice point is, and this isthe one i want to spend a little more time on. do you want to be a catholic, or do youwant to be a buddhist? i went to jesuit school in india. i actually meditated and extracted fromst. francis of asisi every morning. so, you know, even though i was born ahindu.
so, you know, certainly jesuit educationhas an enormous imprint on you. but what do we mean by catholic versusbuddhist? see, if you're a catholic, it's likeyou're, like in-n-out burger. you've got a specific design or likeintel's copy exactly model. but, if you're a buddhist, what you do is,you don't allow a thousand flowers to bloom but you give them good guard rails, in the form of principles orprotocols. when you can actually allow for localrealization. and one of the heroes in our book is chipconnelly, who
actually just has joined arab b and b astheir customer relations person. he used to own a chain of hotels calledjoie de vivre. that he sold. and then all over the place, like thephoenix hotel to the rex hotel, and their range in price from 100bucks a night to 700 bucks. so when we are asked, chip, how do we go about building these hotels, look athis buddhist protocol. he asks his employees, any time you cometo me with a request for funds to start a new hotel, you needto answer four questions.
question one: what magazine does yourcustomer read? if it's the new yorker, i know exactly whothey are. you don't need to tell me anything else. question two: what are the five adjectivesyou would use to describe the experience? so, if it's the new yorker, is sophisticated one of the adjectives, forexample? if that's indeed an adjective, how willthe customer quickly sense that through his or her fivesenses? nobody needs to tell the person this is asophisticated place.
you walk in, you instantly know, you know? maybe they're playing a particular kind ofmusic, through a sense of ear, or you see the bar's like alibrary, whatever. and the next question is, how do the employees know this through their fivesenses? you know, what kind of uniform are youwearing? you know from that onwards. so, when he's getting people to look athow detailed he's making people to thinkthrough a problem.
seemingly simple protocol, but in everyhotel you actually have to come up with different answers to thesame five questions. so, i wanna sort of give us a little bitof time. there are a number principles. i won't talk about all of them. i'm actually gonna talk about a couple ofthings. the first one is, when you scale, most of us think that structure is the way tocoordinate. and that certainly is useful.
but smart leaders very quickly realize stories are actually very powerfulcoordination devices. sometimes they're a substitute forstructure, and other times they're a complement to structure, and the kind of so storiescan be very powerful. so one lovely experiment done by lauraboroditsky, who used to be in the psych department here, she and hergraduate student peggy thebadore did a brilliantexperiment. so they took a large group of people likeyou, randomly put you into two conditions.
and gave you one page of information thatwas completely identical about a fictionalcity called adison in california. so they gave you about, information aboutthe crime rate in california. income, education, all these things. there was just one difference. people on this side of the room, they were told that crime is lurking like a beast incalifornia. and this side of the room, they said crimeis lurking like a virus in addison city. and they ask people, what do you think thecity of addison oughta do?
so what do you do to a beast? what do you do to a beast? kill it. cage it. slay it. all of that. and so when you ask people what should thecity of addison do? what, what did they say? more cops, more jails, tougherenforcement.
and by the way, this is independent ofpolitical persuasion. you could be a democrat, who hates allthis, but the moment you've got this like beast framing there in your head, you're saying more jails, more this, morethat. what do you do to a virus? treat it? you immaculate, you cure it. so, what should the city of addison do? what the city of addison ought to do
is more counseling, more rehabilitation,and even if you were a fervent republican who was againstgovernment expenditures, you said we need to do all of this. it's amazing, so ben leda and hercolleagues said, you know, maybe we are using words like praying, what ifjust use beast and virus. same result. so they would use the word beast or virusin the first two, three lines. and they said, what happens if we do it atthe very end? no effect.
but when you actually have a small wordlike that, very quickly what the story is doing is, all of us havemental butlers in our heads. and these word cues. you instantly initiate the story tellingin your head, and they reflect, of course, in yourpolitical preferences. sometimes stories can be even more, imean, the stories or cause can be even moredramatic. and one of the things we chronicle in ourbook is about british petroleum where the head of the retailunit wanted to go after shell.
and his storyline, was slam the clam. so who's the clam? obvious, shell. so i went to him and i said, phil, comeon, this sounds kinda hokey. i mean, you just say slam the clam andstuff happens? come on, tell me how, how does stuffhappen here? and he said, you know, i give you that,you might think it's a simple little thing, but it's extraordinarily how powerful theeffect is.
first, he said, i didn't need a slide toget that idea across. i just told people, slam the clam. they got it, it stuck. so it survived the car journey home. most of us listen to leaders who when theymake big announcements, you simply sit in your car, you drive home,you've forgotten what they said. so this guy said, look, it survived thecar trip home. and i said, okay, i grant you that. what else did it do?
and he said, you know, people would sit ina meeting, and they would say, where should we putnew gas stations? ultimately, it would all come down to onething. does it slam the clam? where do we put advertising dollars? where do we put r&d dollars, where do wehire people, anything you did, you just had one question to actually focuson, and that's the power of the story. so, that's something you really ought tothink about. the second important point i want to makeis, i want to focus on number four.
called cut cognitive load. so, one of the most famous articles inpsychology is by george midler, it's called the magicnumber 7, plus or minus 2. and, the idea is that there are seventhings we can actually keep in our head. and, that's the, that's what we can keepin working memory, that's what we can actuallyprocess in terms of attention. you actually increase the load to be onseven. what happens is, willpower is amazinglyeroded. so here is a lovely experiment done by roybormeester and his colleagues.
they took a large group like you andrandomly put you into two conditions. so both groups, of course, they don't knowthat the other group is there, they're given ten math problems, and you have toanswer them, in exactly the same sequence. unbeknownst to the participants in theexperiment you can only answer four of the ten. in the ten minutes given to you. so what they experimenters do, is they actually tell people in this one group,hey, before you solve the ten math problems,
just write a two paragraph essay, sometopic. hypothetically, let's assume it's the cityof new york, write an essay without, you know, but make sure you don't usewords that actually have the letters. z, or x, which is easy, cuz a lot of english words don't have this particularconjoining of the letters. this group, the [unknown] cognitive load,they say, write a story about new york, but don't use the letter n andthe letter a. that's tough, and then please solve themath problems. so everybody does that and theexperimenters go back to these people and
say, you know what, instead of likegetting each of your answer sheets, physically verifying it and giving you adollar for each correct answer, why don't we scream the answers, and you tellus how many you got right. we'll give you a buck. but this is where the experimenters becomevery clever. so, they actually make sure in each roomthere is a paper shredder. so, they tell the participants take yourphysical answer, shred it. nobody will know how many you got right,and tell us how many you think you got right, and we'llpay you a buck.
they do this for this group as well. now this group that was asked to write anessay without using the letters n and a, how many answers do you thinkthey claim they got right? and remember, you can only get four right. the average number they claim they gotright was eight. [laugh] and what about that group there? 4.75, now, because you randomly assignedpeople, it's not the dishonest people over here and honestpeople over there. you know, you've randomized with respectto inclination towards honesty.
what happened in this room, what happenedwas they increased load, they put temptationin their way. what funded the temptation [inaudible] thepaper shredder. and once you use the shredder you can tell me how many answers you think you gotright. i'm gonna give you a buck for everycorrect answer. for four extra bucks, basically theycheated and they lied. and, why is that? it's not that people are saying, oh my godthis, this was a hard
task that's why i'm trying to get moremoney out of you or something like. when you, what is it you need to actuallyresist temptation? will power, that's exactly right. most of us think will power is here, brainpower is here, but they're actually much moreinterdependent than we ever have realized. and, what do we do in companies? to our employees. what do we do in terms of their brain, thedemands on their brain power? is that going up or going down?
going up! and then what do we do? we sit and stand in front of them and sayuse your will power. you think they have any to use? obviously not. and what a lot of executives therefore. the smart ones do is they're constantly thinking about, how do you actually lowerload? how do you actually make sure that thewillpower kicks in?
because, remember, excellentorganizations, you don't need to tell people what to do. they know what the right thing is, they'lldo it. so, and for you as you think about yourorga, i was just actually at a company who's namei can't mention. and i asked one of their executives, heywhen you had to order laptops in your company, how manypeople had to approve it? he looked at me and said 18 approvals. i'm thinking to myself 18 approvals.
that's a lot of approvals for a laptoppurchase. what are most people gonna do? they'll give up [unknown] screw thelaptop, i'm not gonna go and get like 18 approvals to do this,and that giving up is the worst kind of problemyou, you can have, so you really want to thinkcarefully about cognitive load. the last thing i wanna touch on is, mostpeople think that going from good to, i mean, going towards excellenceis a journey where, you know, you just pour in goodinto a system.
and that'll get you to great. it turns out that the real thing is bad isway more powerful than good. in fact, a basic finding, in, psychologyis bad dominates good all the time. see if i ask you, what are your mostpowerful memories. bad memories rush to the forefront. what is the most formative relationships,bad memories kind of quickly bad relationships surge tothe forefront. in one study of marital couples marriedcouples they found that for every nasty thing you said toyour partner
you had to at least say a minimum of five good things to offset it you know, so youcan imagine. now the point is, if you don't focus onthe bad and only focus on the good, you're introuble. so, the best experiment on bad versus goodwas actually done by philip zimbardo here, that's thebroken windows experiment. it wasn't some cop in new york who figuredthis out. philip zimbardo did a brilliant experimentin 1969. he took a vw, put it in brooklyn,
broke the windows of the car, and startedfilming. and guess what? if you see the film, you actually have afather with two kids, he sees the broken windows,quickly steals the radio. and within an hour the car's like wipedclean, ransacked. it's become a playground. so, zimbardo says, you know, maybe we outta try the experiment in palo alto,after all stanford university, you know, maybepeople are
better, maybe people are civic minded, etcetera. so, he keeps the card, people are kind ofpuzzled they say hey, who's card is it. but zimbardo says, let me break thewindows here too. and, you know, he breaks the windows heretoo. what do you think happens in palo alto? exactly the same thing that happened inbrooklyn. what do broken windows tell people? that's exactly it. nobody cares, and that's what you've gotto really focus on first.
so, there's a lovely case of this chineserefrigerator company called haier. lousy refrigerators, they were making bad,lousy refrigerators. what would most companies do? let's get black belt, blue belt, greenbelt, whatever it is, and beat you them, you know, etcetera. you over invest in them. but this guy said wait a minute. max doing all of this statistical qualitycontrol in tqm, when the bathrooms the workers areusing are terrible.
that's the broken window i need to fix. cuz what do the lousy bathrooms tell theworkers? nobody cares. and how can you ask them to care for thecustomer? so, that's where he started. so, the real question for you is, you'vegot to always ask yourself, what is yourbroken window? and one other piece of interestingresearch. there are a number of you know, women inthe group here,
and i know the win folks just came in lateas well. particularly for women executives, thescaling challenge is always more likely to be from bad togreat. and the reason is, because of somethingthat is called a glass cliff. it turns out that a glass cliff isobviously very different than a glass ceiling and soon. terms, i'm sure all of you are familiarwith. the problem with the glass cliff is, womenare actually given leadership opportunities where thechance of failure is very high.
so, you know, their given actually,precarious jobs. stuff to sort of succeed in these jobs,one reason is, the men have declined them. you know, which is one reason. you know, they, they, the get, these jobs,and the woman accept them, because their not sure theymight get a second chance. the results are experimental evidence thatsuggests that people, when they ask women to be leaders in situationswhere the performance is poor, usually when youthink of crises they actually gravitate towards a more femalestereotype of a manager.
that's the experimental kind of evidence,the long and short of it is, i was, of course, giving a talk to a group of three hundred womenexecutives of course i was telling them, them this and they said of course we know this youknow, we, we they call it the glass cliff we've known this for a while you know, but the challenges you alwayshave to figure out how is it you're going tostart from the bad to the great. we have i thought i'd give a little moretime for questions, but i'm a professor.
i got caught up in my own, the littlesharing of the book idea, so pardon me for that, but if wehave time, let's take a couple of questions, and then i'll make aquick closing remark, and then will you know, you can head on to wine and foodand merriment and enjoyment. so, any questions that you might like toask? >> what was about the implications fromfeature expert obviously, that might be viewed as a model, and aboutscaling assets? >> yeah. the fact that she was actually saying i am
not going to wait for anybody, i'll do itmyself. that's what real excellence is. so, you, the people who are excellent,that's kind of what they do. let me, one of the heroes in our book, andthank you for asking that question. one of the heroes in our book is a guycalled doug deets. doug is a senior executive in ge healthcare. you go to his office he's got plaques, awards, but that one day went to apediatric, radiology section and you had young kidsafflicted
with cancer who actually were, had to havemris. and doug sees these young childrenscreaming with fear. nearly all of them had to be sedated. and he says, is this the kind of producti'm making? that's scaring young kids, and they haveto be sedated. parents have to entrust their seven year old to somebody, who's going to sedatethem. so what did dough do? doug was one of the first, edition of ourcustomer focused innovation program.
the guy goes back spends time in, in daycare and kindergarten, observingyoung children. and then it dawns on him and he says, ohmy god i've been stupid. i've been developing an engineerssolution, to a childs problem of anxiety. i need to develop a childs solution, tothe childs problem of anxiety. and what does he do? in the new line of, things that he's comeup with, and he had to fight against a lot ofpeople in ge. who were saying, do we need to do this?
do we not? why and where for? so, in the new line of products, what'samazing is, if you go to an mri, one of the mri, the, the front end of themri, actually looks like a space capsule. [sound] so, you tell a young kid, hey. you're going inside a space capsule. they've forgotten it's narrow. that it's actually claustrophobic. you're kind of saying, you know?
okay, that sounds cool. so what, you know? another one. there's an x-ray machine that shakes alot. and, so what they've done is they'veredesigned it, as, like, the rapids. so, they tell people, hey, you're in araft. it's going to go over the rapids. it's gonna shake like hell. but, it's gonna be fun and, you know, theydo all these little things and
it's extraordinary what happened tosedation rates, they actually fell down by a large margin. customer satisfaction went up and, of course, you know, processing timesimproved, then got promotions, and as we werewriting this book bob and i. were on the phone with doug. and, you know, doug, what's gonna happento you, now that you've become a vp? and he says, no i just declined thepromotion. i said what?
why did you decline the promotion? and doug said well, i came home one day. and my wife had just come back from amammogram. and like most males, i sorta asked her,honey, is everything okay? was it okay? and doug's wife is definitely a pistol. so, she kinda quickly realized that dougneeded some serious user feedback to put itmildly. [laugh] so she looked at doug and said.
come, lemmi show you what it feels like. [laugh] so, yeah, i know, i know. poor doug, you know? unsuspecting doug. he's actually led into the garage. she opens the garage door, and says, doug,put your thing out. i'll show you what it feels like. [laugh] and doug says, oh my god. and he, he looks at her, and says, messagereceived.
i actually, i'm gonna focus on this. and that's what he's been doing for thelast year and a half. what's he been doing? coming out with a new range ofmammography, environments, where you actually can[unknown]. and the images are much better. you don't need to face the excruciatingpain, pressure, all of these kinds of things that a lot of peoplego through. so, when i think of that young schoolteacher, it's like,
you don't wait for people to tell you whatto do. you do sort of what you need to. yes, please. right there. >> you talked a little earlier about thepremortem. >> as an idea, where people kinda writedown how they would fail. >> mm-hm, or how they would succeed, too. >> yes, how do you prevent the people who write the failure condition fromfreaking out?
[laugh] or you know, maybe fulfilling theprophecy themselves? >> so, that's a reasonable question. so what you want is you want to stimulatetheir imagination, but you don't want to arouse their anxiety is kind ofthe other way you put it. so, what you have to do is you actually have to give them reasonablydetailed instructions that serve as kind of mental scaffolding,so they're not gonna go off the deep end. so, what you want is you actually, so for instance, you would say, here's a storythat you, when
you fail, this is the kind of story that's gonna feature in the pages of nyt orwhatever magazine. and then you say, write a story. and when you write a story, you actuallygive them a page limit. you know, so you, you don't need to go onand on. and you tell them you can write the storyany way you want. you can write it as a memo. you can write it as whatever, but it hasto be a story and not a list. and usually what people do is they
actually give pretty granular kinds ofaccounts. now, i found that the stories of failure. or actually little more concrete thanstories of success, and the reason you need to do both is, what causes failure may bequite different from what causes success. it's not like there are five factors, highon everything success, low on everythingfailure you know, it doesn't work like that, you know life's toocomplicated to work like that, and, so you have to take thosethings.
you then have to put them together, andthen you have to analyze what the common themesare. i usually like to after reading when iwork with companies, i usually gather all these premortem, and ipresent them not in the sense of, like you know, so many peoplesaid this or that, but you present them more in the visual ofa downward spiral. and then an upward spiral, people say oh my god, that's where we need to startearning. >> do the same people write both stories?
>> no, you have to actually, cuz onceyou've imagined one thing,. it's hard to reimagine the other, so you,you got to make it random and not say, well, let me go to the optimist and let me go tothe pessimist and that kind of thing. you don't want to do all that. just make it random, one page, andanonymous, independent, they actually ask, you can actually get a lotof useful things. by the way, you might wanna do that ifyou're thinking of taking new jobs, too. so you should really do a pre-mortem ofthe job and ask the person hiring you,
or if you're the person hiring somebody,you know, you can actually do this sort ofinterchanging. cuz that's really what you wanna negotiateabout, not simply how much you're gonna be paid. cuz everything's about the runway. is it a short runway, is it a long runway,how to prepare the runway. one more question please. then we're gonna be, out of time. yep.
>> i read marty seligman's book whichreferences the five to one ratio for couples, and he also talks about how there's a three toone ratio for organizations, of positivecomments to negative comments. so i've implemented this in myorganization, but one of the questions that i've always had is, ifpeople in meetings sit around telling each other howfantastic they are, then there can be a demotivating effect of basicallybeing average or okay. so is there any experimental evidence ofa, of a top side to that
ratio of when organizations are toopositive and the employees don't feel enoughmotivation? >> that's an excellent question. i think what you don't want to do is, youknow, seligman's view, of course, is it's all about positive thinking and positiveemotion and there's certainly quite a bit to kinda recommendabout that. but if you only give people informationabout what is it that they're good at and you don't talkabout their failure. clearly, you're giving them biased
information, and paradoxically, that verybias in the information, causes people todiscount the usefulness of it. so, what you want to give people is, theway i sort of think about it is, it's not feed backbut feed forward. so it's not like what did you do well orwhat did you do poorly. what is it you need to do in the future? cuz, you know, a little bit of therearview glass is a useful thing, but you gotta get them to think what is itthey're going to do in the future? and you know, i think when you think ofthe negative things.
here the important thing, of course, isnot to fall prey to the fundamentalattribution adder. >> [inaudible]the number is too high? >> uh-huh. >> [inaudible]>> yeah. i don't think that there's kind of upper bound if that's kind of what you'rereferring to. our interest of course, in the bad beingstronger than good is i sort of think of that as a nicestudy.
that tells you how potent bad is and whyyou need to pay attention to it. i think a lot of it is also quite honestlythe kinds of people you hire. i was actually with the ceo recently and isaid, what's your secret of hiring? how do you scale excellence? and he says, well, i try and actually lookfor people who are tougher on themselves thani am on them. i don't need to micromanage it. then i can put an arm around them, youknow. now the catch is, how do you find thosepeople?
and sometimes you can find them in theoddest of spaces. one lovely story i heard from a hedge fundexecutive in new york was i said, how do you hire people at,the, the entry level? you get to like math jocks and you know,so forth, or whizzes and so on and so forth, or how do you goabout doing it? he says, no, we actually aim to hireathletes from the best university. i said, athletes, like, exactly what doesthat tell you? he looked at me and he said, the greatthing about athletes from the best universities, is they may not be
eventually succeed in terms of aprofessional career. but they have two amazing characteristics. the first thing is, these athletes have anincredible capacity to do repetitive work. hundred meter swim. do it again. they can do that for a long period oftime. the other, the, the other thing he said,which was like a brilliant line was, he said they're themost open to feedback. they can be coached.
cuz what do you do if you're an athlete? you get the hundred meters. and then you sorta look at the coach andsay, what did i do wrong? where'd i, what do i need to fix? so he said, that's why we need to, we hirepeople like that. cuz. they're actually craving that feedback. but they keep, keep on doing therepetitive work. and they're not all thinking they wanna beyou know
running the company you know, imminentlysoon or whatever it is. so it's 6:15. my hope is this has been useful to you. and you wanna actually think of. you know, taking some of these ideas aboutscaling excellence. so as you scale them and if you have achance to read the book, by all means please write to me withquestions, stories, and the like. that's how at least i get educated. but i want to close with one little story
hopefully that kind of inspires you alittle bit. and that is some, sometimes when you have to scale excellence, you don't haveresources. you don't have money, you don't have thepeople, you don't have all these things, how do you dothat? and a great example of that is this chainhere called cost plus world market, in california, when the ceo tookover the company, the stock was i think at a buck and a half a share, whenhe sold it to bed, bath and beyond i think he sold itfor twenty two dollars a share.
so i asked barry, i said, barry, how didyou turn this place around? i said, did you pay out big bonuses? he said, can't do that, no money to dothat. did you hire great people? can't do that. did you train them? he said, how can i train people? we sell, you know, jam from sweden tohandy crafts from kenya to whatever, nobody can have all theproduct knowledge, it's too complicated.
so, i said, what did you do? and he said like, well, we figured outthat the real task is to get people to buy more than one thing whenthey walk into a store. so i said, what did you do? did you have an incentive program? the more people bought, the more peoplegot paid? he said no. i said, what if you do, then? he said, we told our employees to do twothings.
when a customer walks in, smile. and the second thing is, when a customerwalks in, give him or her a basket. that's it. that's what you need to do. because the moment you give them a basket,the probability of them purchasing more than one producthas shot up substantially. then i'm gonna just buy one thing. they'll probably buy three things. it was the most effective thing that wedid, he said.
and my question to you, as you're thinkingabout scaling up excellence in your own respective organizations is,what's the equal end of your basket? so think about that. it's 6:15, i'm out of time. thank you so very much. [sound] thank you.