COVERSTORY

How FailureBreeds Success

Everyone fears failure. But breakthroughsdepend on it The best companies enibracetheir mistakes and iearn from them.BY JENA MCGREGOR

EVER HEARD OF CHOGLIT? HOW ABOUT OK SODA OR SURGE?Long after " New Coke" became nearly synonymous with inno-vation failure, these products joined Coca-Cola Co.'s graveyardof beverage busts.

Choglit, in case you blinked and missed it, was a chocolate-flavored milk drink test-marketed with Nestlé in 2002. OK Soda,unveiled in 1994, tried to capture Generation X with edgy mar-keting. The "OK Manifesto," parts of which were printed oncans in an attempt at hipster irony, asked : "What's the point ofOK Soda?" It turned out customers wondered the same thing.And while Surge did well initially, tliis me-too Mountain Dewlater did anything but. Sales hcgan drying up after five years.

Given that history, failure hardly seems like a subject Chair-man and CEO E. Neville Isdell would want to trot out in front ofinvestors. But Isdell did just that, deliberately airing the topic atCoke's annual meeting in April. "You will see some failures," hetold the crowd. "As we take more risks, this is something wemust accept as part of the regeneration process."

Warning Coke investors that the company might experiencesome ftops is a little like warning Atlantans they might experi-ence afternoon thunderstorms in July. But Isdell thinks it's vi-tal. He wants Coke to take bigger risks, and to do that, he knowshe needs to convince employees and shareholders that he will

42 I BusinessWeek I July 10. 2006

tolerate the failures that vidll inevitably result. That's the onlyway to change Coke's traditionally risk-averse culture. And giv-en the importance of this goal, there's no podium too big forsending the signal. "Using [the annual meeting] occasion ele-vates the statement to another order of importance," Isdell saidin an interview with BusinessWeek.

CLOSETO BLASPHEMYWHILE FEW CEOS are as candid about the potential for failureas Isdell, many are wrestling vnth the same problem, trying toget their organizations to cozy up to the risk-taking that inno-vation requires. A warning: It's not going to be an easy shift. Al-ter years of cost-cutting initiatives and grovting job insecurity,most employees don't exactly feel like putting themselves onthe line. Add to that the heightened expectations by manage-ment on individual performance, and it's easy to see why somany opt to play it safe.

Indeed, for a generation of managers weaned on the rigors ofSix Sigma error-elimination programs, embracing failure-gasp!—is close to blasphemy. Stefan H. Thonike, a professor atHarvard Business School and author oí Experimentation Mat-ters, says that when he talks to business groups, '̂ I try to be

COVERSTORY

•«I9 N O- - J

; !&

W — i^

Famous

Need inspiration?Many companieshave founasuccess in theashes of theirmemorablemisses:

provocative and say: 'Failure is not abad thing.' I always havelots of people staring at me, [thinking] 'Have you lost yourmind?' That's O.K. It gets their attention. [Failure] is so impor-tant to the experimental process."

That it is. Crucial, in fact. After all, thaf s why true, break-through innovation—an imperative in today's globally com-petitive world, in which product cycles are shorter than ever—is so extraordinarily hard. It requires well-honed organizationsbuilt for efficiency and speed to do what feels unnatural: Ex-plore. Experiment. Foul up, sometimes. Then repeat.

Granted, not all failures are praiseworthy. Some flops are justthat: bad ideas. The eVilla, Sony Corp.'s $500 " Inter-net appliance." The Pontiac Aztek, General MotorsCorp.'s ugly duckling "crossover" SUV. For goodmeasure, well throw in our own industry's spectacu-larly useless flop: the CueCat A marketer's dream, thedevice, which was launched in 2000 (when else?),scanned bar codes from magazine and newspaperads, directing readers to Web sites so they wouldn'thave to go to the troubleto type in the URL.

But intelligent fail-ures—those that happenearly and inexpensivelyand that contribute newinsights about your cus-tomers—should be morethan just tolerable. Theyshould be encouraged."Figuring out how tomaster this process offailing fast and failingcheap and fumbling to-ward success is probablythe most importantthing companies have toget good at," says ScottAnthony, the managingdirector at consulting firm Innosight.

"Getting good" at failure, however, doesn't meancreating anarchy out of organization. It means lead-ers—not just on a podium at the annual meeting, butin the trenches, every day—who create an environ-ment safe for taking risks and who share stories oftheir own mistakes, it means bringing in outsidersunattached to a project's past. It means carving outtime to reflect on failure, not just success.

FAILURE PARTIES: PERHAPS MOST IMPORTANT, it means designing ways toi measure performance that balance accountability viâth the free-' dom to make mistakes. People may fear failure, but they fear theI consequences of it even more. "The performance culture real-i ly is in deep conflict with the learning culture," says Paul J. H.I Schoemaker, CEO of consulting firm Decision Strategies Inter-¡ national Inc. " It's an unusual executive who can balance these."\ Some oi^anizations have tried to measure performance in a

way that accounts for these opposing pressures. At IBM Re-search, engineers are evaluated on both one- and three-year timeframes. The one-year term determines the bonus, while thethree-year petiod decides rank and salary. The longer frame canhelp neutralize a year of setbacks. "A three-year evaluation cyclesends an important message to our researchers, demonstratingour commitment to investing in the early, risky stages of innova-

tion," says Armando Garcia, ^vice-president for technicalstrategy and worldwide op-erations at IBM Research.

XeroxMode/A Copier(the "Ox Box"), 1949

The first manually operatedxerographic printer, theModel A was slow, messy,and hard to use. Businesseswere unconvinced andlargely stuck with carbonpaper. But 10 years later,Xerox launched the fullyautomated Model 914,transforming modernoffice work,

FordEdsel,1957

Known as the Titanic of tlieindustry, this overhyped,oversized, and overpricedcar had abysmal sales.Customers even scoffed atthe stodgy name. After justover 2,800 of the 1960models were churned out,the Edsel was history.Chastened, Ford did itsresearch and tuned intocustomers'call for stylishatfordability, launching thelegendary Ford Mustangin 1964.

Huía Burger,1962

Founder Ray Kroc testedthis cheese-topped grilledpineapple on a bun forChicagoans who avoidedeating meat on Fridays,But customers abstained.The company learnedmeatless didn't have tomean wacky, and the nextyear a franchise ownercame up with a tastieralternative forhamburger-free Fridays;theRiet-O-Fish, nowaMcDonald's classic.

In addition to making sure performance evaluations take along-term view, managers should also think about celebratingsmart failures. (Those who repeat their mistakes, of course,should hardly be rewarded.) Thomas D. Kuczmarski, a Chicagonew-product development consultant, even proposes "failureparties" as a way of recognizing that it's part of the creativeprocess. "What most companies do is put a wall around a fail-ure as if ifs radioactive," says Kuczmarski.

Intuit Inc., based in Mountain View, Calif., recently celebrat-ed an adventurous marketing campaign that failed. The com-pany had never targeted young tax filers before, and in early2005 it tried to reach them through an ill-fated attempt to com-

"Most companies...puta wall around a fail44 ! BusinessWeek I July 10, 2006

bine tax-filing drudgery with hip-hop style. Through a Web sitecalled RockYourRefund.com, Intuit offered young people dis-

[ counts to travel site Expédia Inc. and retailer Best Buy Co. andthe ability to deposit tax refunds directly into prepaid Visa cardsissued by hip-hop mogul Russell Simmons.

But even hip-hop stars can't make 1040s cool enough to getyoung adults excited about taxes. "We did very few returns"through the site, says Rick Jensen, vice-president for productmanagement at Intuifs consumer tax group. "It was almost arounding error." Through a postmortem process, the team thatdeveloped the campaign documented its insights, such as the

now finding fans among large companies. Eggers hopes fu-ture conferences -will feature even more presentations on fail-ures. She also plans to distribute "narrative storytelling book-lets" about failed projects so people can "feel the pain."

Unlike Intuit, most companies don't spend enough time andresources looking backward, says Chris TVimble, a professor atthe TYick School of Business at Dartmouth College and co-authorof Í0 Rüksßjr Strategic Innovators. That's a mistake. "How do youleam if you don't examine the past?" asks Trimble.

General Electric Co. is trying to do just that. The company,which is well-known for sharing best practices across its many

SonyBetamax,1975Sony kept tight controlsover the license for itsvideo recordingtechnology while its rival,JVC, shared its VHSformat. Now, Sony facesanother format war, pittingits Blu-ray next-generation DVD againstToshiba's HD-DVD. Thistime, Sony has assembleda powerful cadre ofbackers. Will it win?

The first commercialpersonal computerfeaturing a graphical userinterface (GUI), Lisa didn'tsell because it was sluggishand expensive. But Lisa'sGUI helped to inspireApple's famously user-friendly product line, fromthe crisp Macintoshinterface of the iMac to thesleek and simple iPod.

Coca-ColaNew Coke,1985

Coca-Cola rolled out a well-researched, taste-testednew formula to energize itslethargic brand. Butafirestorm of consumerprotest led to New Coke'sdemise after 79 days, whenthe original was broughtback with great fanfare.Today, New Coke is acelebrated chapter inCoke lore.

PfizerSiídenafíl,1991

First tested on humans in1991, Sildenafil didn't proveeffective for ifs initialindication: angina, or chestpain. After patients reportederections as a side effect,Pfizer began testing thecompound for erectiledysfunction. In 1998, Viagrabecame the first drug totreat the condition, and theblockbuster has been ahousehold name ever since.

Kellogg'sBreakfast Mates.1998

This cereal-and-milkcombo product was lost onconsumers who found thenon refrigerated milkunappetizing. Theinconvenient conveniencefood was not nearly asappealing as the real thing(above). More mindful ofits harried customers'needs, Kellogg hasexpanded its line of handybreakfast bars.

fact that Gen Yers don't visit destination Web sites that feel toomuch like advertising. Then, on a stage at the Dolce HayesMansion in San Jose, Calif., last October in front of some 200Intuit marketers, the team received an award from Intuit Chair-man Scott Cook. "Ifs only a failure if we fail to get the learn-ing," says Cook.

In addition to postmortems, Intuit has begun plucking in-sights from its flops through sessions that focus on failure.Jana Eggers, who heads up Intuit's Innovation Lab, held thefirst such "When Learning Hurts" session recently. There, sherecounted the story of QuickBase, a software application thatfailed for its initial market, small business customers, but is

units, has recently begun formally discussing failures, too. LastSeptember the company set up a two-hour conference call formanagers of eight "imagination breakthroughs" that didn'tlive up to expectations and were being shelved, or "retired," inGE'S parlance. ("Imagination breakthroughs"—IBs—are newbusinesses or products that have potential sales of $100 mil-lion within three to five years.)

Such discussions can be nerve-racking, especially in compa-nies where failure has traditionally been met with tough con-sequences. That was the case at GE, which is now three yearsinto the effort spearheaded by Chairman and CEO Jefifrey R. Im-melt to make innovation the new mantra at the $150 billion be-

as if it's radioactive," says one consultantJulv 10. 2006 I BusinessWeek I 45

COVER STORY

Helen Greiner-Co-founder and chairman

¡Robot

When we first started workingwith the military in the early1990s, lied the design of anamphibious mine-clearingrobot, called Ariel. It walked

in the suri and acted like a crab. It was a "12-degree of freedom" walking robot, meaningthere are 12 motors to drive 12 joints. At thetime, it was the most advanced walkingrobot in the world. We built technologies thatwe knew from the labs. Buf there was a verystrong missing component of the user'sneeds. It couldn't walk far enough, itcouldn't carry the payload it would need tocarry, and it was too complex.

When we demonstrated Ariel, thereaction was: "iRobot is doing some reallyinnovative and futuristic stuff, but it doesn'tsolve a current need." It started theperception of our company along a line

My favorite mistake..MI wouldn't choose now: We do innovation forinnovation's sake. Now I believe theperception of iRobot is we build practicaland affordable robots that help people.

After Ariel, I was invited to an ArmyRangers demonstration and broughtanother prototype to solicit input fromsoldiers. That input led to the design of our

PackBot-we've delivered 500 of them sofar—a bomb-disposal robot that has beencredited with saving the lives of dozens ofsoldiers. Innovation in our field can lead usin the wrong direction to the most exotic

creation. 1 learned to talk to users and getinput before designing. Don't think that youare going to design something that has all ofthe bells and whistles, then reduce the costssometime later.

hemoth. "I had some offline conversations with some of the IBleaders reassuring them that this was not a call where they weregoing to get their pink slips," says Patia McGrath, a GE mar-keting director who helped put together the call. "The notion oftaking big swings, and that it's O.K. to miss the swing, is some-thing that's quite new with Jeff."

Some companies have gone even further, taking a compre-

hensive look at all their previous failures. That was the case atCorning Inc., which found itself teetering on the brink ofbankruptcy after the once red-hot market for its optical fibercollapsed during the telecom bust. Following that debacle,then-Coming GEO James R. Houghton asked Joseph A.Miller Jr., executive vice-^president and chief technology offi-cer, to produce an in-depth review of the company's 150-year

history of innovation, documentingboth failures and successes.

One of the failed products Coming in-vestigated was the DNA microarray, orchip, which the company began devel-oping in 1998. Genomics research washeating up at the time, with Dr. J. CraigVenter launching Celera Genomics thatyear. Coming, which makes laboratorysciences equipment, saw an opportunity.Its DNA chip was designed to print all28,000 human genes onto a set of slidesthat could be used by researchers. By2000, Coming had invested $100 mil-Hon in the project and announced apartnership with Massachusetts Insti-tute of Technology.

"WE WERE LATE"BUT WHILE CORNING was trying tolaunch the chip, another company,Afíymetrix Inc., commercialized one."They had the dominant design onmicrochips, and they were the firstout," says Peter F. Volanakis, nowComing's chief operating officer. "Wewere late." Quality problems plaguedthe project, and customers had notbeen brought in early. With Corning ina freefall financially, the DNA chip waskilled in 2001.

Still, the experience ojíened Comingup to a whole new market. "We had dis-covered the marketplace of drug discov-ery," says Miller. By combining its intro-duction to the drug research marketwith another failed business, photonics,which manipulates data using lightwaves, it created Epic, a revolutionarytechnoiogy for drug testing that it will launch this fall. By usinglight waves instead of fluorescent dyes. Epic promises to accel-erate dramatically the process of testing potential drugs andimprove its accuracy.

One key difference? This time, 18 pharmaceutical companieshave tested Epic before the launch. By 2010 to 2012, Jeff"Mooney, who led Epic's development, projects that Epic sales

mark that you couldn't get outunless you buffed it withsandpaper. It was a classic caseoí ¡ust not asking the rightquestions up front.

This one was my mistake. 1 letthe need for speed overwhelmdoing enough up-front marketresearch and testing. It was a $20million mistake. We caught it afterabout three months. Customerswould complain, At first, you go

Jeffrey R. Immelt-Chairman and CEO

General Electric

ni992,1 was running all thecommercial operations for the plasticsbusiness at GE. We had a productcalled Nuvel, which was a sheetproduct that would go over wood to fry

to create the poor man's Corian countertop.Turns out the thing just didn't work. Any timeyou dropped a coin on it, if would leave a

through this denial phase: "Youdon't understand" the productand stuff like that.

It made me leam aboutlistening better. I'm moredisciplined on the up-front stuffnow than I was then. I wanted to

do something big and exciting, and I wantedto do it now rather than waif a year. I fessedup: stepped into it, made up for the financialhit we had to take on it [by exceeding salestargets on other products], and made goodto the customers, I think that t got better atunderstanding the need for research andmore thought up front, so you don't have toredo things.

You're never allowed in GE to make thesame mistake twice. You're allowed to makethe mistake once. If you try something and itfails but you went about if the fight way andyou learned irom it, that's not a bad thing.

could reach $ 100 million to $300 million a year, and more than$500 million annually long-term.

As Coming learned ÍTom the DNA chip and with Epic, get-ting potential users in before a project goes too far helps toprove the market for it. But outside perspectives can also helpneutralize emotions and biases about failing product lines, saysDuke University Fuqua School of Business professor William

SHARE YOUR THOUGHTS.Sun believes sharing is the way to create better ideas. That's why we've teamed up with

BusinessWeek to offer you an opportunity to share your comments. Join the conversation about

this week's Cover Story at businessweek.com/coverstory.

micfosystems

COVERSTORY

Boulding. In research published in the April issue ofûie Journalof Marketing, Boulding and his colleagues contradict the com-mon notion that teams cling to a project because they want tosave face or salvage the "sunk costs." Rather, the problem Iswith the objectivity oí the people involved. "Even if you're noton the hook in terms of financial embarrassment or psycholog-ical embarrassment," says Boulding, "you did form beliefs, andthat causes you to warp new feedback."

That's why W.L. Gore & Associates Inc. in Newark, Del.,makers of the waterproof fabric Gore-Tex, recognizes out-siders—people within Gore but not on the product develop-ment team—who make the call on projects that need to bepulled. When Brad Jones led Gore's Industrial Products Div.,which makes sealants and filtration systems, he handed out"Sharp Shooter" trophies to these outside managers when aproject was effectively killed. These marksmen, so to speak,fi^ed from the trappings of familiarity, can identiiy potentialsnags that the team may have overlooked. "We're eñusive inour thanks for that contribution," says Jones. "We ask them towrite up what they learned from it, and how we could havemade the decision [to kill the project] faster."

FIND YOUR OWN FLAWSTHE MINDSET THAT Gore looks for in these outsiders—theability to home in on uncertainties—requires employees to re-frame their thinking. Most people naturally seek positive out-comes and set about trying to prove that an experiment works.

But designers, inventors, and scientists, all models for com-

FAILING FORWARDMore Mistakes: In an expanded slide show, more leaders sharestories about their failures and what they learned from them

Learning on the Front Lines: How the U.S. Army's after-actionreview process can help corporations learn from their mistakes

The Story Behind the Story: For a podcast interview with JManagement Editor Jena McGregor by Executive Editor John 1A. Byrne, go to businessweek.ccmÄearch/podcasting.htm J

http:/Awíw.busír)e5sweek.comA3rtras

panies struggling to be more creative, take the opposite tack.They try to prove themselves wrong. That focus on potentialflaws makes failure, and the lessons that come with it, happenearlier. Amy Edmondson, a professor at Harvard Business Schoolwdio has studied how oiganizations leam from failure, says man-agers would do well to think more like scientists. "Failure pro-vides more learning' in a stricdy logical or technical sense" thansuccess, she says. "Ifs a principle of the scientific method thatyou can only disconfirm, never confirm, a hypothesis."

Failure's capacity to teach is exactly why venture capitalistsoften look for managers to run startups whose resumes includeexperience with a flop. Gordon McCallum, CEO for RichardBranson's Virgin Management Ltd., can point to managerswithin Virgin who might have been overlooked by other com-panies because of failures in their careers. He's also quick to

PLAYBOOK: BEST-PRACTICE IDEAS

A Formula for Failureit's innovation's great paradox: Success-that is, true breakthroughs-usuaily comes through failure.How to help your team get comfortable with taking risks and learning from their mistakes.

Formalize ForumsFor FailureTo keep failures and the valuablelessons they offer from gettingswept under the rug. carve out timefor reflection. GE has recentlybegun sharing lessons Irom failuresby bringing together managerswhose "Imagination Breakthrough"efforts are put on the sheli.

MoveGoalposts

ii'ovation requires flexibility inmeeting goals, since earlypredictions are often little morethan educated guesses. Intuit'sScott Cook even suggests thatteams developing new productsignore forecasts in the early days."For every one of our failures, wehad spreadsheets that lookedawesome," he says.

SharePersonal StoriesIf employees hear leadersdiscussing their own failures.they'll feel more comfortabletalking about their own. But it snot just the CEO's job-front-l ineleaders are eyen more important.says Harvard Business Schoolprofessor Amy Edmondson. "Thatperson needs to be inviting,curious, and the first to say:'I made a mistake.'"

Bring inOutsidersOutsiders can help neutralize theemotions and biases that prop upa ñop. Customers can be the mostvaluable. After its DNA chip failed,Corning brought pharmacompanies in early to test its newdmg-discovery technology. Epic.

Prove YourseifWrong, Not RightThe tendency for developmentteams is to look for supporting,rather than countervailing,evidence. "You have to reframewhat you're seeking in the earlydays," says Innosight's ScottAnthony. "You're not reallyseeking proof that you have theright answer. It's more abouttesting to prove yourself wrong."

CelebrateSmart FailuresManagers should designperformance-managementsystems that reward risk-takingand foster a long-term view.But they should also celebratefailures that teach somethingnew, energizing people to try againand offering them closure.

48 BusinessWeek I July 10, 2006

COVERSTORY

note that eiTors on the job, as long as they aren't repeated, arenot only supported, but valued.

One example: Virgin Atlantic Airways Ltd.'s J2000 seats, a$67 million investment made in 2000 to create new sleeperseats that reclined at an angle for the airline's "upper-class"seats. Although sleeper seats had long existed in first class, air-lines had not yet adopted them for business class. Virgin wasthe first to announce it would be offering "a bed in business,"says Joe Feny, Virgin's head of design, who led the design of theJ2000 seats. Within a year, however. Virgin's idea was one-upped by its chief competitor, British Airways PLC, which rolledout a truly flat bed. While customers were initially enthusiasticabout the J2000, some complained about sliding and di.scom-

fort. In the end, says McCallum, it "was wildly unsuccessful.Everybody acknowledged that it was not as good a product asour principal competitors'." jarees Ferry: "We were an also-ran, which didn't really sit well with us."

But Ferry didn't get the ax. In fact. Virgin entrusted him totake on another extraordinary risk, committing a huge $127million to an overhaul of the airline's upper-class seats yearsbefore tlie traditional product life cycle would have ended.And the company stuck by its investment even after Septem-ber 11. The new version, launched in 2003, has been a solidsuccess. Called the "upper-class suite," Ferry's makeovermade a design leap beyond merely being flat. Flight atten-dants flip over the back and seat cushions to make the bed, al-

E. Neville Isdell-Chairman and CEO

Coca-Cola

Iwas [Cuke's] division president forCentral Europe between "85 and '88. Wewere doing a restructuring to get 116bottlers to exchange their equity forownership in a single bottler in

Germany. We took the plan to Atlanta andpresented it to the key people we wereresponsible to and got the support that weneeded. We went back and went on a roadshow around Germany for three days. We

reckoned we would get [support] fromaround 70% of the bottlers, and about 30%would reiect us, but that would be enough toget critical mass, End of Day Three, we werecelebrating because we had a maior bottier,who we thought would not agree, say, "Itmakes sense."

Then a call came in from the internationalpresident-at about U o'clock at night-saying: "Stop. You're going to take this offercompletely off the table."

What I had underestimated was thedegree to which there were multipleconstituencies that needed to be coveredin order to ensure that the back-door, oldrelationships that some ot the bottlers had

builf up could not stop us. Phone callscame in to headquarters from bottlersaround the world, asking: "What does thismean for me?" I had not thought thatthrough well enough.

I like telling that story because I failed,and the lesson I learned is you have to lookoutside the narrow view you have of wherethe areas of influence exist. Be sure thatyou have them all covered, Now I think in adifferent way about where obiections mightcome from. If you're going to createchange, you're going to have people whoyou're not going to convince. You canneutralize some of them, but I had notdone that,

50 I BusinessWeek I July 10. 2006

COVERSTORY

lowing for different foam consistencies for sitting and sleep-ing. While Ferry hoped the new seats would eventually im-prove Virgin's business-class market share by 1%, they've al-ready exceeded that goal.

TREMORS AND THRILLSA COMPANY'S REACTION in the face of intelligent failures,whether it's Virgin reinvesting in a pricey design revamp or In-tuit giving an award to its marketing team, can send tremors orthrills through a culture. If top executives are accepting, peoplewill embrace risk. But if managers react harshly, people will re-treat from it.

That's something Eric Brinker, JetBlue Airways Corp.'s di-rector of brand management and customer experience, wantsto make sure doesn't happen. Last year, Jet-Blue, which tries to limit its in-flight snackmix to keep costs low and reduce complexity,began hearing that some of its customerswanted a healthier choice. Brinker and his

Pete Carroll-Head Footbail Coach,

University of Southern Caíifornia

When I left my job as defensivecoordinator of the SanFrancisco 49ers, I had twoopportunities, at the St. LouisRams and the New England

Patriots. The Rams had had problems, andexpectations would be low. On the otherhand, the Patriots had iust come off losingthe Super Bowl. What's more, [former headcoach] Bill Parcells had iust left the team.

I went to the harder situation. Mymistake was not to evaluate the impact ofthe guy I was following-one of the mostheralded coaches in the NFL, a guy withgreat charisma and great stature. Thechallenge of it—all the pressure of trying tomaintain the program-blinded me to howdifficult it would be. Even going 27>21 overthree years. I still couldn't keep up withexpectations, and I lost my job because wedidn't win enough,

I thought I could handle anything,thought I could follow Bill Parcells. I felt like,Yeah, this'll be good., .111 win them over.

My instincts-to a fault—are always to goto the high-pressureopportunity. I didn't reallyassess the situation dearlyenough, I've learned to reallyweigh things and not be sodriven by my gut. Sometimes it gets us introuble, that feeling that you can handleanything. You have to dig deeply and not iustbe emotionally driven into a situation. Greatcompetitors sometimes fail prey to theirown drive to see what they can endure.

team decided to replace a popular but hardly healthy mix ofDoritos chips called Munchie Mix. "It's the ultimate junkfood," says Brinker.

The junk food fans revolted. "The tribe had spoken, andthese guys wanted Munchie Mix," Brinker says. "People wrotereally spirited letters, saying: This is the only reason I flew Jet-Blue!'" Brinker realized he would have to reverse course, but hedidn't want his team to think change wasn't encouraged.

So he decided to make fun of himself to keep the reactionlighthearted. On the company's intranet, Brinker started up a"Save the Munchie Mix" campaign that read: "Some pinheadin marketing decided to get rid of the Munchie Mix!" He invit-ed employees to write in poems and stories about why thesnack should return to JetBlue. The point? By keeping thingsfun, he hoped employees wouldn't hesitate to make their owncreative decisions. "If we don't have people willing to risksomething, then we'll really end up like our competitors." Andthat, of course, would be a failure indeed. •

-With William C. Synionds in Boston, Dean Foust in Atlanta,and Diane Brady and Moira Herbst in New York

52 I BusinessWeek I July 10, 2006

Copyright of BusinessWeek is the property of Bloomberg, L.P. and its content may not be copied or emailed to

multiple sites or posted to a listserv without the copyright holder's express written permission. However, users

may print, download, or email articles for individual use.