Kotter, John P. "Leading Change. Why transformation efforts fail", Harvard Business Review, January 2007, pp 92-107

THETESTS OFA LEADER 8HST SF HSN 1995

Editors Note: Gulding change may be the ultimate test of a leader - no busrness

survives over the long term if it can't reinvent itself' But' human nature betng

what it is, fundamental change is often resisted mighttly by the people it most

affects: those tn the trenches of the business'Thus' leading change is both

absotutely essential and incredibly difficult

Perhaps nobody understands the anatomY of organizatronal change better

than retired Harvard Business School professor John P Kotter'This article'

original ly pubtished in the spring of 1995' prevtewed Kotter 's 1996 book Leadtng

Change. l t outl ines ergnt cr i t icalsuccess factors - f rom establ ishtng a sense of

e\traordinarY urgencY, to creating short-term wins' to changtng the culture ("lhe

way we do things around 6srs") lt will feel famitiar when vou read it' in pan

becauseKot te r ' svocabu laryhasentered the lex iconand inpar tbecause i t

con ta ins thek indofhomet ru ths tha twerecogn ize , immedia te ly 'as i fwe 'd

always known them. A decade later, his work on leadtng change remains

def initive.

Leading Changewhy Transformati0n Efforts Fail

ILl

l t

Leaders who successful ly t ransform

in the r ight order) .

businesses do eight th ings r ight (and they do them

by John P. Kofter

i

I

vER THE PAST DECADE, I have watched more than loo

companies try to remake themselves into significantly

better competitors' They have included large organiza-

tions (Ford) and small ones (Landmark Communications)'

companies based in the United States (General Motors) and else-

where (sritish Airways), corporations that were on their knees

(Eastem Airlines), and companies that were earning good money

(erir,ol-lt4y.rs Squibb). These efforts have gone under many ban-

ners: total quality management, reengineering, rightsizing' re-

structuring, cultural change, and turnaround' But' in almost

au.ry .ur", the basic goal has been the same: to make fundamen-

,ul .hung., in how business is conducted in order to help cope

with a new, more challenging market environment'

A few of these corporate change efforts have been very suc-

cessful. A few have been utter failures' Most fall somewhere in be-

tween, with a distinct tilt toward the lower end of the scale' The

lessons that can be drawn are interesting and will probably be rel-

evant to even more organizations in the increasingly competitive

96 Harvard Business Review I January 2007 I nbr'org

o

:I

business environment of the comingdecade.

The most general lesson to belearned from the more successful casesis that the change process goes througha series of phases that, in total, usuallvrequi re a considerable length of t ime.Skipping steps creates only the illusionof speed and never produces a satisfy_ing result. A second very general lessonis that crit ical mistakes in any of thephases can have a devastating impact,slowing momentum and negating hard_won gains. perhaps because we haverelatively little experience in renewingorganizations, even very capable peopleoften make at least one big error.

Error 1: Not Establishing a GreatEnough Sense of UrgencyMost successful change efforts beginwhen some individuals or some groupsstart to look hard at a company's com_petitive situation, market position, tech_nological trends, and financial perfor_mance. They focus on the potentialrevenue drop when an important pat_

/7./'/< l L t )

ent expires, the flve-year trend in declin_ing margins in a core business, or anemerging market that everyone seemsto be ignoring. They then find ways tocommunicate this information broadlyand dramatically, especially with re_spect to crises, potential crises, or greatopportunities that are very timely. Thisfirst step is essential because iust get_ting a transformation program startedrequires the aggressive cooperation ofmany individuals. Without motivation,people won't help, and the effort goesnowhere.

Compared with other steps in thechange process, phase one can soundeasy. It is not. Well oV€r loyo of the com_panies I have watched fail in this firstphase. What are the reasons for thatfailure? Sometimes executives under_estimate how hard it can be to drivepeople out of their comfort zones.Sometimes they grossly overestimatehow successful they have already beenin increasing urgency. Sometimes theylac l< pat ience:"Enough wi th the pre l im-inaries; let's get on with it.,' In manycases, executives become paralyzed by

the downside possibilities. Thev worrvthat employees with seniority wilt be-come defensive, that morale will drop,that events wi l l sp in out of contro l , thatshort-term business results will be ieop_a rd i zed , t ha t t he s tock w i l l s i nk , andthat they wil l be blamed for creatinea crisis.

A paralyzed senior managementof ten comes f rom having too manymanagers and not enough leaders.Management 's mandate is to min i_mize risk and to keep the current sys_tem operating. Change, by definition,requires creating a new system, whichin turn always demands leadershio.Phase one in a renewal process typ i_cally goes nowhere until enough realleaders are promoted or hired intosenior-level jobs.

Transformations often begin, andbegin well, when an organization has anew head who is a good leader and whosees the need for a major change. If therenewal target is the entire company,the CEO is key. Ifchange is needed in adivision, the division general manageris l<ey. When these indiv iduals are io t

-._

nbr.org January 2007 i Harvard Business Revjew 97

THETESTSoFALEADERIgH 'sT t }F ' . { gR lLead ingChange :WhyTrans fo rma t i onE f fo r t sFa i |

new readers, great readers, or change termfutureoftheorganizationisputin outsideof thenormalhierarchybydef-

champions, phase one can be a huge jeopardy' inition' This can be awkward' but it is

challenge. when is the urgency rate high clearly necessary. If the existing hierar-

Bad business results are both a bless- .norrgnl From whai I have seen, the chy were working well' there would be

ing and a curse in the first phase. on the urrr*.. is when about 75ozo oi u .o*- no need for a major transformation' But

positive side, losing *o.t.y does catch lll^'

management is honestly con- since the current system is not working'

people,s attention. But it aiso gives less uin..a irrui uirri.r.r, as usuar is iotatty reform generally demands activity out-

maneuveringroom.withgoodbusiness .rrru.c"piuut". Anything r"r, L. pro side of formalboundaries,expectations'

results,the opposite lstru-e: Convincing d"*;;;;;ious proulems later on in and protocol'

people of the need fo' tiungt is much the process' A high sense of urgency within the

harder, but vou have more resources to rocess'

Tffffi;'IL:11fi-t::t"ill"JT#1il:lhelp make changes' Frror 2. Not Greating a Powerful But more is usually required' Someone

But whether the star t ing polnt ls ; - - '

good performance or ;::, ,f ,ffi; rnough

-G-;iding co-aritior needs to get these peopre together' help

successful cases I have witnessed, an Majorrenewalprogramsoftenstartwith them develop a shared assessment of

individual or a group always fac'itates just one or two people. In cases of suc- their company's problems and opportu-

a frank discussion of potentia'y un- ..rrfrrttrurrrformation"r*"r,,n.t."0 nities, and create a minimum level of

pteasant facts about n.'* .o-p.,i ion, .*il.;;; grows and ;;;; ;"t trust and communication' off-site re-

shrinking margins, decreasing market time. But wheneve*orn. -ini-.,- treats, for two or three days, are one

share, flat earnings, a lack of revenue *u$ ir rro, achieved earry in the effort, popurar vehicle for accomplishing this

growth, or other ,.t.u*i inarces of a .'o,t i"g -"rn worthwhile.h"nn*r. iurt. I have seen many groups of five to

declining competitive position. Because . ,f rr"or,.n said that -uiot iiu"g. ir 35 executives attend a series of these re-

there seems to be an armost universal i-porriuil unless the head of the orga- treats over a period of months'

human tendency to shoot the bearer of nization is an active supporter. what Companies that fail in phase two usu-

bad news, especiany if the head of the , "*,"ir.i"g "uout

go., fu.b.yona,trur. ally _underestimate

the difficulties of

organization is not a change champion, tr, *...rrr,'t transformationrG .nuit- producing change and thus the impor-

executives in these companies often .nui o. pr.rident or ai"irio" g*"r"r iance of a powerful guiding coalition'

rely on outsiders to bring unwanted in- -un#r, pr,,s another nut o' i' o' 'o Sometimes they have no history of

formation. wa, street analysts, custom- ry"rrl, io*. tog"ttr", una"a"*ioo u teamwork at the top and therefore un-

ers, and consurtants c", "ii

u. herpful shaied commitment to exceir* p.rio* dervalue the importance of this type of

in this regard. The purpose of all this ac- *""." through renewal rn .nv ""p.ti

coalition. Sometimes they expect the

tiviry in the words of one former CEo ";;int,

*r""1::::T.:*;;uriortttt team to be led bv a staff executive from

of a large European company, is ,,to company's most senior executives be- human resources' quality' or strategic

make the status quo seem more danger- aurrr" ,o*. people lort *on't buy in, at planning instead of a key rine manager'

ousthanlaunchingintotheunknown. r"urt*tatfirst.Butinthemostslccess-

No matter how capable or dedicated

In a few of the most successful cases, fJ-.ur"r, the coalition i' u*u' n"T lhe :taff head' groups without strong

a group has manufactured a crisis. one oo*"rrui - in terms

"r ti*t, i"r"t*u line leadership never achieve the power

CEOdeliberatelyengineeredthelargest tion and expertise, reputations, and thatisrequired'

accounting loss in ihe company's his- relationships' Effortt that don't have a powerful

tory, creating huge pressures from wall .

'-in uotrr^ r*a' and large organiza- enough guiding coalition can make ap-

street in the process. one division pres- tions, a successful guiding team may parent progress for a while' But' sooner

ident commissioned first-ever customer .onrlra or or-rry three to nu."p.opt" aui or later, the opposition gathers itself to-

satisfaction surveys, knowing full well lns

trie first year of a renewal effort. But gether and stops the change'

that the results would be terrible' He in*bigcompanies'the coalitionneedsto

:lH[rT.,:rTj;rtiri:* m il"#.::ii:i"i:tr#i';',T:::ili:: '.:,i::.Lackins avisionrisky. But there is arso risk in praying it anJbeyond. senior managers always I-n ev;rv successful transformation ef-

too safe: when the urgency rate is not form the core of tt" sr""pl ;uL sorne- fort that I have seen, the guiding coali-

pumpedupenough,thetransformation timeryorrnnauou'a*"'if""'u"p"-

tion develops a picture of the future

process cannot succeed, and the rong ##;.Jil*:\"J;::'-;;;;;;"' :xT;'#xiffy;1ilTJi:$ii'"1:

Now retired, John P r<onut*u' the Kono

--'U"tut"t the guiding coalit ion in- ers' and employees' A vision always

suke Marsushita professor of Leadershrp at crudesmemberswhouranotpurior*- goe.s beyond the numbers that are

Harvard Business School in Boston ttio' *utu"-ent' it tenJs to ope'ate typicallyfound in five-year plans' A vi

98 Harvard Business Revrew I January 2007 I hbr'org

sron says something that herps clarify works at it for three or five or eventhe direction in which an organization 12 months, something much betterneeds to move' Sometimes the first emerges through their tough anaryticaldraft comes mostry from a single in- thinling and a little dreaming. Eventu-dividual' It is usually a b-it bruiry, at all,astrategyforachievingthatvisionleast initially. But after the coaliiion ls also Oevetopea.

EIGHT STEPS TO TRANSFORMINGYOUR ORGANIZATION

ft EstaUtistring a Sense of UrgencyI . Examining market and competit ive realit ies

. ldentl fying and discussing crises, potential cr ises, or ma,or opportunit ies

Forming a Powerful Guiding Coali t ion' Assembling a group with enough power to read the change effort. Encouraging the group to work together as a team

In one midsize European company,the first pass at a vision contained two_thirds of the basic ideas that were in thefinal product. The concept of globalreach was in the initial version fromthe beginning. So was the idea of be_coming preeminent in certain busi_nesses. But one central idea in the finalversion- getting out of low value_addedactivities - came only after a series ofdiscussions over a period of severalmonths.

Without a sensible vision, a transfor_mation effort can easily dissolve intoa l ist of confusing and incompatibleproiects that can take the organizationin the wrong direction or nowhere atall. Without a sound vision, the reengi_neer ing pro iect in the account i iedepartment, the new 36o_degree perlformance appraisal from the human re_sources department, the plant,s qualitvprogram, the cultural change pio;..tin the sales force will not add up in ameaningful way.

In failed transformations, you oftenfind plenty ofplans, directives, and pro_grams but no vision. In one case, a com_pany gave out four_inch_thicl< note_books describing its change effort. Inmind-numbing detail, the books spelledout procedures, goals, methods, anddeadl ines. But nowhere was there aclear and compelling statement ofwhere all this was leading. Not surpris_ingly, most of the employees with whomI talked were either confused or alien_ated. The big, thick books did not rallvthem together or inspire change. tnfact, they probably had just ti,. Lppo_site effect.

. Creating a vision to help direct the change efforl

. Developing strategies for achieving that vision

Communicating the Vision' Using every vehicle possibre to communicate the new vision and strategies. Teaching new behaviors by the example of the guiding coal i t ion

Empowering Others to Act on i lVision

Creating a Vision

. Creating those improvements

don't f i t the vision

. Gett ing r id of obstacles to change'changing systems or structures that seriousry undermine the visron. Encouraging r isk taking and nontradit ional ideas, act iv*res, and actions

IPlanning for and Creating ShXrm Wins. Planning for visible perforrnance tmprovements

. Hir ing, promoting, and deveroprng employees who can

. Reinvigorating the process with new projects, themes,

v

Recognizing and rewarding emproyees invorved in the rrnprovements

Consolidating lmprovements and producing Sti l l More Change' Using increased credibir i ty to change systems, structures, and poricies thalIn a few of the less successful cases

that I have seen, management had asense of direction, but it was toocomplicated or blurry to be useful. Re_cently, I asked an executive in a midsizecompany to describe his vision and re_ceived in return a barely comprehensi_ble 3o-minute lecture. aur ied in ; ; ; ; ;swer were the basic elements of a soundvision. But they were buried _ deeplv.

A useful rule of thumb: lf you cin;tcommunicate the vision to someone infive minutes or less and get a reaction

rmptement the visronand change agents

Insti tut ional izing New Approaches'Art icurating the connections between the new behaviors and corporatesuccess

I

nDr.org January 2007 | Harvard Business Feview g9

THETESTS OFA LEADEH * tST * f $$* I Lead ing Change:WhyTrans format ion Ef fo r ts Fa i l

that signifies both understanding and

interest, you are not yet done with thisphase of the transformation process.

Error 4: Undercommunicating theVision by a Factor of Ten

l've seen three patterns with respect to

communication, all verY common. In

the first, a grollp actually does develop

a pretty good transformation vision and

then proceeds to communicate it bY

holding a single meeting or sending out

a single communication. Having used

about o.ooo1oz" of the yearly intracom-pany communication, the group is star-

tled when few peoPle seem to under-

stand the new approach' In the secondpattern, the head of the organization

spends a considerable amount of time

making speeches to employee groups'

but most people still don't get it (not

surpr is ing, s ince v is ion captures only

o.ooo5% of the total yearly communi-

ca t i on ) . I n t he t h i rd Pa t te rn , much

more effort goes into newsletters and

speeches, but some very visible senior

executives still behave in ways that are

ant i thet ica l to the v is ion. The net resul t

is that cynicism among the troops goes

up, while belief in the communicationgoes down.

Transformation is impossible unless

hundreds or thousands of people are

willing to help, often to the point of

making short-term sacrifi ces. Employ-

ees wil] not make sacrifices, even if they

are unhappy with the status quo, unless

they believe that useful change is possi-

ble. Without credible communication,

and a lot of it, the hearts and minds of

the Lroops are never caPtured'This fourth Phase is ParticularlY

challenging if the short-term sacrifices

include job losses. Gaining understand-

ing and support is tough when downsiz-

ing is a part of the vision. For this rea-

son, successful visions usually include

new growth possibilities and the com-

mitment to treat fairly anyone who is

laid off.Executives who communicate well

incorporate messages into their hour-

by-hour activities. In a routine discus-

sion about a business Problem, theY

talk about how proposed solutions fit(or don't flt) into the bigger picture.

In a regular performance appraisal,

they talk about how the emPloYee's

behavior helps or undermines the vi-

sion. In a review of a division's quarterly

performance, they talk not only about

the numbers but also about how the

division's executives are contributing

to the transformation. In a routine

Q&A with employees at a company

facility, they tie their answers back to

renewal goals.In more successful transformation

efforts, executives use all existing com-

munication channels to broadcast the

vision. They turn boring, unread com-pany newsletters into l ively articles

about the vision. They take ritualistic,

tedious quarterly management meet-

ings and turn them into exciting dis-

cussions of the transformation. They

throw out much of the comPanY'sgeneric management education and

replace it with courses that focus on

business problems and the new vision.

The guiding principle is simple: Use

every possible channel, especially those

that are being wasted on nonessential

information.Perhaps even more important, most

of the executives I have known in suc-

cessful cases of major change learn to

"wall< the talk." They consciously at-

tempt to become a living symbol of the

new corporate culture. This is often not

easy. A 6o-year-old plant manager who

has spent prec ious l i t t le t ime over 40years thinl<ing about customers will not

suddenly behave in a customer-orientedway. But I have witnessed just such aperson change, and change a great deal.

ln that case, a high level of urgency

helped. The fact that the man was a part

of the guiding coalition and the vision-

creation team also helped. So did all the

communication, which kept reminding

him ofthe desired behavior, and all the

feedback from his peers and subordi-

nates, which helped him see when he

was not engaging in that behavior.Communication comes in both words

and deeds, and the latter are often the

most powerful form. Nothing under-

mines change more than behavior bY

important individuals that is inconsis-

tent with their words.

WQ

100 Harvard Business Review January 2007 I hbr'org

Error 5: Not Removing Obstaclesto the New VisionSuccessful transformations begin to in-volve large numbers of people as theprocess prog,resses. Employees are em-boldened to try new approaches, to de-velop new ideas, and to provide leader-ship. The only constraint is that theactions fit within the broad parametersof the overall vision. The more peopleinvolved, the better the outcome.

To some degree, a guiding coalitionempowers others to take action simplyby successfully communicating thenew direction. But communication isnever sufficient by itself. Renewal alsorequires the removal of obstacles. Toooften, an employee understands thenew vision and wants to help make ithappen, but an elephant appears to beblocking the path. In some cases, theelephant is in the person's head, andthe challenge is to convince the indi-vidual that no external obstacle exists.But in most cases, the blockers arevery real.

Sometimes the obstacle is the orga-nizational structure: Narrow job cate-gories can seriously undermine effortsto increase productivity or mal<e itvery difficult even to thinl< about cus-tomers. Sometimes compensation orperformance-appraisal systems make

the new initiatives. He paid lip serviceto the process but did not change hisbehavior or encourage h is managers tochange. He did not reward the uncon-ventional ideas called for in the vision.He allowed human resource systems toremain intact even when they were

l f you can"t communicate the vlsion to someone in f iverninutes or less and get a reaction that signif ies both

understanding and interest, you are nCIt done.

people choose between the new visionand their own self-interest. Perhapsworst of all are bosses who refuse tochange and who make demands thatare inconsistent with the overall effort.

One company began its transforma-tion process with much publicity andactually made good progress throughthe fourth phase. Then the change ef-fort ground to a halt because the officerin charge of the company's largest divi-sion was allowed to undermine most of

clearly inconsistent with the new ideals.I thinl( the officer's motives were com-plex. To some degree, he did not believethe company needed major change. Tosome degree, he felt personally threat-ened by all the change. To some degree,he was afraid that he could not produceboth change and the expected operat-ing profit. But despite the fact that theybacked the renewal effort, the other of-ficers did virtually nothing to stop theone blocker. Again, the reasons were

FinancialTimes, zoo6

Iir\1r')sr,1

L I A F Tn'{r) ' i . : I i .

PE.; J,:,,)

I

F o r a t m o s t f i f t y y e a r s ,

e x e c u t i v e s f r o m a r o u n dt h e w o r l d h a v e i n t e g r a t e dThe Executive Program (TEP)

a t D a r d e n i n t o t h e i r c a r e e r s .Through fEP, execut ives seet h e i r b u s i n e s s e s f r o m a w h o l en e w p e r s p e c t i v e a n d u s e t h e i rn e w k n o w l e d g e t o l e a d t h e i ro r g a n i z a t i o n s t o p l a c e s t h e ym i g h t n e v e r h a v e e n v i s i o n e d .T h e t r a n s f o r m a t i o n a l I e a r n i n ge n v i r o n m e n t i s c u s t o m a r y ;t h e r e s u l t s g a i n e d w i l lbe ex t raord inary !

i l

I I l-,' ' : 7

I

Iq,

:tf.

E 't- fr$e U^Nlvensrrv

7 VIRGINIADARDEN ScrJooI- oF tsusr^*[ss

l r x E c u t ' l v u I D u c i T I o !

-.-

After a few years of hard work, managers may be ternpted to declare victory with the first

clear performance improvement. while celebrating a win is f ine' declaring the war won can

be catastroPhic'

THETESTS OF A LEADER *fi$T *T i lSTq

complex. The company had no history

of confronting problems like this' Some

people were afraid of the officer' The

CEO was concerned that he might lose

a talented executive. The net result was

disastrous. Lower-level managers con-

cluded that senior management had

lied to them about their commitment

to renewal, cYnicism grew, and the

whole effort collaPsed.ln the first half of a transformation,

no organization has the momentum'

power, or time to get rid of all obstacles'

But the big ones must be confronted

and removed. If the blocker is a person'

i t is important that he or she be t reated

fairly and in a way that is consistent

with the new vision. Action is essential,

both to empower others and to main-

tain the credibility of the change effort

as a whole.

Error 6: Not SYstematical lYPlanning for, and Creating,Short-Term Wins

Real transformation takes time, and a

renewal effort risks losing momentum

if there are no short-term goals to meet

and celebrate. Most people won't go on

the long march unless theY see com-

pelling evidence in tz to z4 months that

the lourney is producing expected re-

sul ts . Wi thout shor t - term wins, too

many people give up or actively join the

ranks of those PeoPle who have been

resisting change.One to two Years into a successful

transformation effort, you find quality

beginning to go up on certain indices or

the decline in net income stopping' You

find some successful new product intro-

ductions or an upward shift in market

share. You find an impressive productiv-

ity improvement or a statistically higher

customer satisfaction rating. But what-

ever the case, the win is unambiguous'

Lead i n g Change: WhyTrans format ion Ef fo r ts Fa i I

The result is not just a judgment call

that can be discounted by those oppos-

ing change.Creating short-term wins is different

from hoping for short-term wins' The

latter is passive, the former active' In a

successful transformation, managers ac-

tively look for ways to obtain clear per-

formance improvements, establish goals

in the yearly planning system, achieve

the objectives, and reward the people

involved with recognition, promotions,

and even money. For exampie, the guid-

ing coalition at a U.S. manufacturing

company produced a highly visible and

successful new product introduction

about zo months after the start of its re-

newal effort. The new product was se-

lected about six months into the effort

because it met multiple criteria: It

could be designed and launched in a

relatively short period, it could be han-

dled by a small team of PeoPle who

were devoted to the new vision, it had

upside potential, and the new product-

development team could operate out-

side the established departmental struc-

ture without practical problems. Little

was left to chance, and the win boosted

the credibil i ty of the renewal process'

Managers often comPlain about

being forced to produce shortterm wins'

but I 've found that pressure can be a

useful element in a change effort'When

it becomes clear to people that major

change wil ltake a long time, urgency lev-

els can drop. Commitments to produce

short-term wins help keep the urgency

level up and force detailed analytical

thinking that can clarify or revise visions'

Error 7: Declaring VictorY Too Soon

After a few years of hard work, manag-

ers may be tempted to declare victory

wi th the f i rs t c lear per formance im-

provement. While celebrating a win is

flne, declaring the war won can be cat-

astrophic. Until changes sink deeply

into a company's culture, a process that

can take five to ten Years' new aP-

proaches are fragile and subiect to

regression.ln the recent Past, I have watched a

dozen change efforts operate under the

reengineering theme' In all but two

cases, victory was declared and the ex-

pensive consultants were Paid and

thanked when the flrst major project

was completed after two to three years'

Within two more Years, the useful

changes that had been introduced

slowly disappeared. In two of the ten

cases, it 's hard to find any trace of the

reengineering work todaY.

Over the Past 20 Years, I 've seen the

same sor t of th ing haPPen to huge

qualigu projects, organizational devel-

opment efforts, and more' Typically' the

problems start early in the process: The

urgency level is not in tense enough,

the guid ing coal i t ion is not powerfu l

enough , and the v i s i on i s no t c l ea r

enough. But it is the premature victory

celebration that kills momentum' And

then the powerful forces associated

with tradition take over'

Ironically, it is often a combination

of change in i t ia tors and change res is-

tors that creates the premature victory

celebration. ln their enthusiasm over

a clear sign of progress' the initiators

go overboard. They are then joined by

resistors, who are quick to spot any

opportunity to stop change. After the

celebration is over, the resistors point to

the victory as a sign that the war has

been won and the troops should be sent

home. Weary troops allow themselves

to be convinced that theY won' Once

home, the foot soldiers are reluctant to

climb back on the ships. Soon there-

after, change comes to a halt, and tradi-

tion creeps back in.

102 Harvard Business Review i January 2007 hbr'org

Instead of declaring victory, leadersof successful efforts use the credibilityafforded by short-term wins to tackleeven bigger problems. They go after sys_tems and structures that are not consis_tent with the transformation vision andhave not been confronted before. Theypay great attention to who is promoted,who is hired, and how people are devel_oped. They include new reengineeringprojects that are even bigger in scopethan the initial ones. They understandthat renewal efforts take not monthsbut years. In fact, in one of the mostsuccessful transformations that I haveever seen, we quantif ied the amountof change that occurred each year overa seven-year period. On a scale of one(low) to ten (high), year one receiveda two, year two a four, year three athree, year four a seven, year five aneight, year six a four, and year seven atwo. The peak came in year five, fully36 months after the first set of visiblewins.

Error B: Not Anchoring Changesin the Gorporation's GultureIn the final analysis, change sticks whenit becomes "the way we do thingsaround here," when it seeps into thebloodstream of the corporate body.Until new behaviors are rooted in socialnorms and shared values, they are sub_ject to degradation as soon as the pres-sure for change is removed.

TWo factors are particularly impor_tant in institutionalizing change in cor_porate culture. The first is a consciousattempt to show people how the newapproaches, behaviors, and attitudeshave helped improve performance.When people are left on their own tomake the connections, they sometimescreate very inaccurate links. For exam_ple, because results improved whilecharismatic Harry was boss, the trooosl ink h is most ly id iosyncrat ic s ty le wi ththose results instead of seeing how theirown improved customer service andproductivity were instrumental. Helo_ing people see the r ight connect ions re-quires communication. Indeed, one

company was relentless, and it paid offenormously. Time was spent at everymajor management meeting to discusswhy performance was increasing. Thecompany newspaper ran article after ar_ticle showing how changes had boostedearnings.

The second factor is taking suffi_cient time to make sure that the nextgeneration of top management reallydoes personify the new approach. Ifthe requirements for promotion don,tchange, renewal rarely lasts. One badsuccession decision at the top of an or_ganization can undermine a decade ofhard work. poor succession decisionsare possible when boards of directorsare not an integral part of the renewaleffort. In at least three instances I haveseen, the champion for change was theretiring executive, and although hissuccessor was not a resistor, he wasnot a change champion. Because theboards did not understand the trans_formations in any detail, they couldnot see that their choices were not soodfits. The retiring executive in onelasetried unsuccessfulty to tall< his boardinto a less seasoned candidate who bet_ter personifled the transformation. Inthe other two cases, the CEOs did notresist the boards'choices, because theyfelt the transformation could not beundone by their successors. They werewrong. Within two years, signs of re_newal began to disappear at bothcompanies.

There are still more mistakes that Deo_ple mal<e, but these e ight are the b igones. I realize that in a short articleever),thing is made to sound a bit toosimplistic. In reality, even successfulchange efforts are messy and full ofsurprises. But just as a relatively sim_ple vision is needed to guide peoplethrough a major change, so a vision ofthe change process can reduce theerror rate. And fewer errors can spellthe d i f ference between success andfailure.

Repr in t R07O1J; HBR Onpo jn t 1710To order, see page 127

'Firing Back should be read by anyonewho has success on their agenda.'Donald Trump

FI

"l strongly recommend it."Clayton M. Christensen, authorThe lnnovatols Dilenma and The lnnovato/s S0/u'0,

"Filled with ideas and methods that canhelp arry comparry.,'Jong.YongYun,Vice Chairman and CE0.Samsung Electronics

Ava lable wherever books are sold

H A R V A R DB U S I N E S S

s c H o o L wwwftBsPress.org

P R E S S

-