CONTENTS rman o Aciscomtspanets . [rnopvcrin a Parr ont: THE McKINSEY WAY OFTHINKING ABOUT BUSINESS PROBLEMS 1 1, Building the Solution 3 2. Developing an Approach as 43. 80/20 and Other Rules to Live By B Part To: THE McKINSEY WAY OF WORKING ‘TO SOLVE BUSINESS PROBLEMS 47 4, Selling « Study 5. Assembling 2 Team 6. Managing Hieearchy 7. Doing Research 8 9 Teesanbote ar copyeahe Conducting Interviews ‘pati ernaael” Tangara ae epi dl pacer ocr 2 Brainstorming a parr THREE: THE McKINSEY WAY OF SELLING SOLUTIONS. 18 10, Making Presentations us 11 Displaying Daca with Chases 13 (Copyright 1999 Ethan M. Ras. Click here for Tess of Us. "ae ay Ri Ti of ie Mapa Cn YoY es tha BU ea aed Bp yeast lommen a Sale?12, Managing Internal Communications 19 13, Working with liens 7 ‘arr Four: SURVIVING AT McKINSEY 13 14, Find Your Own Mentor 1a 15. Surviving on the Road M45 16. Take These Thece Things with You Wherever You Go 149 17. A Good Assistant Is Lifeline 133 18, Recruiting MeKinsey Syle 157 19, IF You Want Life, Lay Down Some Res 16 arr rive: LIFE AFTER McKINSEY 10 20, The Most Valuable Leson 168 21, Memories of McKinsey 175 box 19 ‘Ba BI eR gia cs of Wi Tp Spe Coy ox Yaw as ‘Bat QOS ler a Pras eo 99 po BEE ctf Sle >cep apm sere ‘era rode! an tngora oma pore pane, EICeE AUC e tan onbste arcs cepyeane 0125. INTRODUCTION ink ofthis book as a sampler. You, the reader, need noe start atthe beginning and work your way through to the end. If you want to, feel fee; otherwise, scan the table of contents and read whatever items interest you mos ABOUT McKINSEY “To provide some background for thote of you unfamiliae with MeKinsey 8 Company, offer a few words abour the organization thar is members past and preset refer toas “the Fim.” Since founding in 1923, MeKinsey 8 Co. has become the world’s most successful strategic consulting firm. I curently has 75 offices (and counting) around the world and employs some 4500 profesional Ie may nor be the largest strategy firm inthe world—some of te big accountng firms have larger practices—bur itis certainly ke most prestigious. McKinsey consults to most ofthe Fortune 100, ss well as to many US. sate and federal agencies and forcign governments. MeKinsey ta brand name in inernatonal busines cirles, Several senior MeKinsey partners have risen to imernationa prominence in their own right, Lowell Bryan advised the Senate Benking Committee during the savings and loan cvsis. Kenichi ‘Ohmae ho recently Ife he Firm) wites books on management (Copyrghe 1999 Exhan M. asi Clik here for Terms of Use ‘ae an Wag gs Som a op Srp Cotas apa Yar Baas mai te enc laa a Dp Spies st ShnsSn Bodeie rrntace can Sayers oma sn pois eo paca joe Us ures spent cop enna © 18, Mere Pein! Bok Ou. Tae on arco Pactipderth and fururology that are bestsellers in Japan, Herb Hensler advised former German Chancellor Helmst Kohl on business and eco: omic matters. Even more visible are some of McKinsey’ “alumni,” who have gone on to senior postions around the worl “Tom Peters, management guru and coauthor of In Search of Excl. lence; Harvey Golub, president of American Express; and Adair ‘Turner, president of the Confederation of British Industries, name but thre. ‘To maincan its preeminent postion (and to earn its high fees) the Fim seeks out the cream of each yea’ crop of busines schoo! graduates It ures them with igh salaries, the prospect of a rapid ssedhrough McKinsey’s mevtocratic hierarchy, and the chance = single withthe elite of the business world. In retuen, the Fiem demands roral devotion to client service, submission toa grueling Schedule that can include weeks or months away fom home and family, and only the highest-quality work. For those who meet ‘MeKinsey’s standards, promotion can be rapid. Those who fll shoct soon find themlves a the latter end of he Firm’ strict pl fey of “up or out.” Like any stong organization, the Firm has « powerful corpo: sate culture based on shared values andl common experiences. Every ‘MeKinsey-ite goes through the same rigorous taining progear’s snd suffers through the same long nights inthe office, To outsider, this can make the Frm seem monolithic and forbidding one recent, book on management consuking likened McKinsey to the Jesuits In truth, a8 I hope thie book will communicate, McKinsey is 28 Jhuman a ics ermpoyees are. ‘The Frm has cs own jargon too. Its fl of acronyms: EM, ED, DCS, ITP, ELT, BPR, and so forth. MeKinseyites call hee ssign ments or projects engagement.” On an engagement, a McKinsey team will search for the *key divers” in ceir quest ro “add value” Like most azgon, much ofthis is simply verbal shorthand somefroin si of it, however, ence understood, can be a5 useful to business people outside the Firm a it is to MeKinsey:iteschemseles In sum, like any powerfl organization, MeKinsey is complex. 1 hope tis book wil life atleast a corner ofthe vel of the McKinsey mystique and allow you, the reader, a peek into the inner workings ofa very successful company. ABOUTTHIS BOOK have divided this book inc five pars In the first three, explain bow MeKinseyites think about business problems and how they ‘work a solving them, Parts One ehrough Thee are che “meat” of the book, and I hope that you wil find in them something new «© help you in your business care. In Part Fou, I give some lessons -MeKinsey-ites have leaned for coping with the stresses of life xt the Firm. Anyone who works hard in today’s business world should learn something useful—or atleast amusing—there. Finally, are Five briefly recounes some lessons learned at che Firm and shares afew memories of McKinsey. wrote this book with the goal of communicating new and we {al skills ro everyone who wants to be more efficient and effective in business, Anyone whe wants to lesa proven methods for approaching busines problems in a srucured, fact-based way wil find them in these pages. And anyone looking for survival cps in the jungle of today’s business world can benefit fom doing things the MeKinsey way. In addition, this book will give any executive who works with management consultants, whether from McKinsey ot elsewher, | ‘some insight into how these strange beings think. Although each | ‘a ae nina Wye gs fam Wa Tp Sagi Como TY Yor Bas soi tt ea is Oe op a Eek Spates dgTHE McKINSEY WAY OF THINKING ABOUT BUSINESS PROBLEMSBUILDING THE SOLUTION Like all things McKinsey, che Firm's problem-solving process has three major ateibuees. When team members meet forthe firse time to discuss their clienr’s problem, they know that theie solution wil be ‘Fact based ‘+ Rigidly structured ‘+ Hypothesisdriven” In this chapter, you will learn exactly what these attributes ‘mean and how you can apply them in your business. CCopyighe 1999 Ethan M, Rasiel. Clik het for Terms of Use. io aaices een sor perma NT cree : ti : : i reece) I ‘ ‘hing At Ber Pls FACTS ARE FRIENDLY Facts are the bricks with which you will lay a path to your solution and build pillars to support it. Don't fear the facts. Problem solving atthe Firm begins with facts. On the first day of an engagement ll members ofthe eam comb through stacks of auccles and internal research dacuments to gather enough facts to illuminate their piece ofthe problem forthe frst eam mecting, Having drawn up an initial hypothesis forthe problem, the team then races ro gather the facts necessary (when put chrough the appropriate analyses) ro support or refute it Ac the start of your time at MeKinsey, gathering and analy2- ing facts is your aon d'te, As one former SEM" observed: ‘When you strip away alr of he high-minded language with which MeKinsey dresses up its problem-solving proces, comes down to very carefl, high-quality analysis of the components of the problem combined with an aggresive aude toward fac gathering. ‘Why are facts so important to the way McKinsey does bus ess? There are two reasons. First, facts compensate for lack of aut instinct (see * -. But Every lien I Unique” in Chapter 2). ‘Most McKinsey ites are generalists, They know a litle about a lot ‘of things. As chey gain experience and move through the ranks, ‘they may come to know a lot about alot of things. Even a this point, however, they will sill know less about, say, inventorycoumces ied "Sanngra tome an poi tp "ie leysapboon de Conta, onan 193, sare ans eres ig te Satan 5 ‘management practices for perishable foodstuffs than the folks who have been running che distribution operations of Stop & ‘Shop for the last 10 years. Gut instinct might el those folks the solution co an inventory management problem in 10 seconds {although they sil would be wise ro check the facts); MeKinsey will goto the fats fis. Second facts bridge the credibility gap. When she joins the Firm the rypical associate” (a least in the United States) will have sraduated near the top of her college clas, spent rwo or three years ‘working fora large company, then received her MBA from atop business school. She wil be in hee mid- co late ewenties. On hie first engagement she may have to present her analysis tothe CEO ‘of a Fortune 50 company, who will or give much credence 0 ‘what some newly minted, 27-year-old MBA has to say—unless se has an overwhelming weight of facts to back her up. Tiss jst 4 eve fora junior executive presenting a proposal his boss. Despite (or possibly because of) the power of facts, many busi- nesspeople fear them. Perhaps they are afraid tha if they ook roo fit chose bilan ideas into your topline issues. Chances are you can. iil fal es fal, Orher Issues wil help you stay MECE. SOLVETHE PROBLEM ATTHE FIRST MEETING—THE INITIAL HYPOTHESIS Solving a complex problem is ike embarking ox a long fow= The initial hypothesis i your problem-solving map. ‘The intial hypothesis (TH), the thied pillar ofthe MeKinsey prob- lemsolving proces, is the most dificult explain, To make che explanation easier for you (and me), I wil break his section ito three parts: [Ba Ee Wy Wai te Tes a a Top a Coma pT Yor Baa ‘Bait Met rr 98 pinto Sermmo S00 apeaig Stain ° + Defining the initial hyporhesis + Generating the inital hypothesis. ‘+ eating the iil hypothesis DEFININGTHE INITIAL HYPOTHESIS ‘The esence ofthe initial hypothesis is “Figure ou the solution to the problem before you star.” This seems counterintuitive, yet yeu do iall the time, Suppose you have o drive oa restaurant in apart of town yeu don’t know, You know you have to make the third lft off Smith Steet and then ake the fis right it's just after that commer. Yeu know how to getto Smith Steet; youl jus follow your directions fom there. Congratulations, you hae an inital hypothesis, Solving busines problems is more complicated than finding 2 reseaurant, bute intial hypothesis works the same way Iria road smap,albeithasly sketched, ro take you fom problem to solution. If your THis corzcs, then solving the problem means lig inte deta ofthe map through factual analysis, ‘ers ern o Acme Widgets from the last section. You and your team must find a way to inerease sales atthe widget Business unt [Aer you've brainstormed using your knowledge of the widget bus nes, bur before you've spent alr of te gathering and analyirg Ae fats, you tig conse up wide the fllowing upline IE esr Sectional fen gnome sea potas oa puata econ Eat inherent aied i ‘Wecan increase widget sles by: * Changing the way we sel our widgets co etal oul + mproving the way we market our widges to consumers. + Reicing the nit cos of our widgets. opsfah © 188, W-awiitPetesonl Rsk rp, Toe AAs [wil show in the nex section, you would hen take each iste ddvin to another level or two of deal ro determine which analyses you need in order to prove or disprove each hypothesis [Remember that a hypothesis is merely a theory to be proved» “ining et Bins Pane or disproved. Iris not she answer: If your IH is correct, then, afew months down the rod, it wil be the first side in your present tion. Ifierurns outro be wrong hen, by proving it wrong, you will have enough information ro move toward the cight answer By putting your IH down on paper, and determining how you can prove or disprove i, you have setup a raed map that you can fol= low to an eventual proved solution GENERATING THE INITIAL HYPOTHESIS | “The IH emerges rom che combination of facts and strcture, Thee- fore, as the fist step in generating an TH, you must start with ke facts Remember, however, that you don’t want todo alot of dig- fing acound for information before you know where to dig. Ore former McKinsey SEM had a good approach for generating Ts: Ar the start of an engagement, I would just ery to digest as ‘much of our fat base as possible. would sit down with the trade publications in that industry for an hous oF two—not so much to gather facts a to absorb something ofthe ft- vor ofthat industry: what the jargon is, what the current indusery issues are I woud especially seek out peopein the Firm who knew about this particular industey. That was the uickert, most efficient way co get upto speed. noma sins poms ca pea ESTEE GTE Eject decors ‘When generating an initial hypothesis, you don’t need alle facts, jst enough to havea good overview of the industry and te problem. Ifthe problem i in your own business, ou may already have the fctsin your head. Tha’ great, but facts ae not enough ‘You have o apply structure to them. 4 To structure your IH begin by breaking the problem into is components—the Key drivers (see “Find the Key Drives," in ‘Chapter 3). Next,’make an actionable recommendation regarding cach driver, This extemely important, Suppose your business, conse 188, mera Seattle rect en ‘dt ney Ry Ug ec oe ah Tp Sag Crop Yo Ta ae Seca at eat mt Se eta Sdn‘en Sonpemios I Petesaa! sk Gro. Toso ingens prc pate oy 108, W-amse EITERLUTS ae lyeohenn se copytg Fa aprocct aig Satan 1 profits ate greatly affected by the weathers in fact, it isthe key determinant of profits in a given quarter, “We have to pray for ‘good weather" snot an actionable recommendation. On the other hand, “We must reduce our vulnerability co changes in the \weather” i an actionable top-line recommendation, For your next step/you must take each top-line recommenda sion and break it down to the level of issues. Ifa given recommen dation is correct, what issues does it raise? Consider the likey answers to each issue. Then go down another level. For each issue, ‘what analyses would you need ro make ro prove or dispeove your hypothesis? With a lle experience, anda lot of debate with your team, you should ger a good sense of what is provable ard ‘what isnot. This wll help you avoid blind alley, Tn te Acme Widgets problem, suppose your team decided tnt the key drivers were the sales force, the consumer marketing ta gy, and production costs. You then eame up with list of action= able top-line recommendations as your iil hypothesis: ‘We can increase widget sales by + Changing the way we sell our widgets to retail outs. + Improving the way we market our widgets to consumers. + Redlucng the unit cost of ou widgers. {er hein with closer lnk atthe sales force Ws onzanized geo _raphically (Northeast, Mid-Atlantic, Southeast, et.) and sells pxi- marily to three types of retail outlets: superstore, department stores and specialty stores. The team believes ‘ought ro be organized by customer type—tha's one issue. ‘What analyses could prove or disprove that belief? You could break out the sales by customer type foreach region, If pene atthe sles forse tion of superstores inthe Northeast i higher than in any othoe region and higher than forthe other types of retail urls, find ‘out why. When you tlk tothe Northeast sales reps, you migit——— 2 ihn Abt ene Pen find that they have a better fel for superstores than any other sales team, What if they were pur in charge of al superstores sctoss the country and achieved the same penetration? What would that mean for widget sales? “The end prot of ths exerci is what McKinsey cals the see tree, In other words, you start with your intial hypothesis and ‘ranch outa each se, The rel looks like dhe figure below. ‘When you've competed your isue tee, you have your prob- Jem-solving map. That’ the easy part. The dificult pat will come ‘when you have ro dig deep to prove your hypothesis. gon geton en none a a < ae omer mmr < aC BE eS se comane ony 9108, yet mee fe gi E WN ft, L— rates into ytm < ‘SS i i, age Te wr Co hi Fr Bae ei i tt ai ek 9) 9 Mises iano Sepace onnaht 108. Fiat Prosi Dok Group. ado a re eee etal es Ace omen sp la Goce Usa loyer option ar copy age Staton B ‘TESTINGTHE INITIAL HYPOTHESIS Before you take your problem-solving map out on the road, you want (forgive che mixed metaphor) to kick che tres on it. Test it. Is ie the best possible hypothesis you could devise, given what you know about the industry and your client or company? Have you thought about all the iues? Have you considered all he rivers of eh problem? Are all your recommendations acsionsbe and provable? ‘When I discussed generating an TH, 1 used the phease “your team” rather than “you.” My experience at che Fitm (an that ofthe many McKinsey lumi Interviewed for ths book) aught ime that Hs produced by teams are much stronger than those produced by individuals, Why? Mot of us ate poor critics of ou own thinking. We need others to pick apart our ideas. A team of thre or four bright individuals is an excellent vehicle for that. So when your team meets to come up with an TH let a ehow- sand flowers bloom. Everyone should have his or her own ides and initial hypotheses. Everyone should be prepared to push teammate's thinking and test each new ides. Ifyou are the team leade, you should try tobe the thought leader oo. Try to take diferent approach from whatever has ust been said. Ask, “What ifwe change this? What if we push chat? How about looking ait this way2? The process involves shooting a cereain asioune of bull. That's OK, have fun—as long as it pushes your :hinkin, (For more ideas and techniques to push your team’s thinking, ste ‘Chapter 9.)a Thinking Abt ine rae ‘THE PROBLEM IS NOT ALWAYS THE PROBLEM Sometimes a business problem will land on your desk and you will be told to salve it. Fair enough, But before you ¢2 ‘rushing off in any particular direction, make sure you're solt- ing he right problem—it may not be the one you were giver. A MeKinsey alumnus with 4 scientific background sold me that business problem solving is organic and complex, like medicine. A patient will come into a doctor’ office and say that he chinks he as the flu, He will ell the doctor about his symptom scratchy throat, achy head, and eunny nose. The doctor will not immed axely ust the patients conclsion. She wll take the patent’ hi tory, ask some probing questions, and chen make her diagnosis. “The patient may have the fl, ors cold, or something more serous, but the doctor will noe rely on the patent ro diagnose himself | ‘At McKinsey, we found that cients were often no better st iagnosing themselves chan a doctor's patients, Sometimes, prob- Jems would come tous in extremely vague formulations. In my fist study at the Firm, ou eam’ mission was to elp a New York Jnvestment bank “increase proftabiliry"—the business equivalect, ‘ofa patient elling the doctor “I don’ fee! well.” In another ese, 1 MeKinseytesm went in o evaluate expansion opportune fr division of a manufacrating company. After afew weeks of gat ring and analyzing data, the eeam eealized chat what the division needed was not expansion; it was closure or sell-of ‘The only way to figure out if the problem you have been given isthe real problem isto dig deeper. Get facts. Ask questions. Poke around. I usually does not take very long to figure our if you ae heading down the right path, and the time you ake up front wil ‘el ee a (Boe i Ue Meet rd Se 9.916 Inova Tayi te ees oi i Seo Hi To Yor Ba mo tapesntact i | iz 1 i i i EXERC RE ek decioyto tpn ere ” tore than make up for itself in ime you don’ waste further down the line ‘Whar do you do when you are convinced you are working oa the wrong problem? When a doctor thinks that » patients mince symproms mask something more serious she wil rel her patient, Me, Jones, Lean treat your headache, bt Think i’ a symptom of something more serious and I'd like to do further tests.” In the same way, you should go ack to your elient, or your boss—who- ver it was chat asked for your input in the fist place—and say, You asked me to look at problem X, but the eal impace on ove performance will come from solving problem Y. Now Ican sohe problem X, if that's what you eally want, but I think is in ove interest to focus on ¥." Ifyou have the data to back you wp, the clieatcan ether accepe your recommendation or ell you to stay en the original problem, but you wil have fulfilled your responsibly to act in the client's best interests, DON’T REINVENT THE WHEEL (PART 1) ‘Most business problems resemble each other more than they differ. This means that with a small number of proble-sole- ing techniques, you can answer a broad range of questions. These techniques may be somewhere in your organization, either written down or inthe heads of your fellow employee. If not, use your experience to develop your own too! kit. McKinsey, ike every other consulting firm, hs developed a sun ber of problem-solving methods and given them faney names Analysis of Value Added, Business Process Redesign, Product-Ma: i ae ay Wy teen oe Wai Tp Song Csi Ys Ya as Dice OW CS Nr al es St Fe ae ett Slammen a Sab Sloe 3peers rarer ere i ee | Sr eee 4 ‘changes occurring in each of the four categories, What impact— ? ‘Whatever business you're in, this framework will not only help t ee ere : ‘Try. Kesounds simple, bu it is a powerful way co simulate your I 3 a thinking about strategic business problems. ‘Those framework cclly help a the ing proces. For instance, when Iwas a second-year associate, joined a team that was helping a major Wall Steet investment bank reorganize its information technology department. All he | executives atthe bank wanted IT reorganized, but norifit meant | any changes inthe way heir computers were supported. The I department was teal mess, with 600 employees, 2 dozen different | subdepartments, and a web of eporting rlaconships thar made iy head spin and the rest ofthe team) haedly knew where co begin. For- ‘ofthe problem sar ‘i ey Wang te Tees ofl ep Sr Comte Ha To Yar Ds ‘Baie i heer rrr p10 Se Seyler Sade?eR Sn Somatostatin parcel pats comb © 108. awit Prosi! Bak Group ado ESTEE a posters copie | Deming an dpc * tunately the Firm had recently developed a new paradigm, called Business Process Redesign, that gave us a starting point. The Fie ‘was stil coming to grips with BPR, and during the study we broke new ground forthe Firm. It was hard work, but BPR (along with 1 lot of sweat, Chines food, and late nights) helped us help che clint force through che reorganization, IF we had not ad atleast 2 starting amework to help us focus our efforts, we might never have lef the ground. . BUT EVERY CLIENT IS UNIQUE (NO COOKIE-CUTTER SOLUTIONS) That there are many similarities between business problems does not mean that similar problems bave similar solutions. You have to validate your intial hypothesis (or your gut) swith fact-based analysis. Ths will put you in a much beter positon to get your ideas accepted. Ill you have i «hammer, then every problem looks like anal. | Critics of McKinsey (and management consulting in general s1y that the Firm bass ie solutions on the most current managers? fad—the favorite ool in its intellectual toolbox. ‘Acthe Firm, at least thsis unerve. Fact-based analysis as prac- iced by McKinsey equires hard proof before any recommends tions are made to clients. Jason Klein, «former SEM at MeKinsey and now publisher of Field & Streams and Outdoor Life, puts it like this People think that he Firm—and management consultans ia sgeneral—have a precanned answer. That iscertanly not the [a EW iy Wai eT Top Sra Cota Ho Tos Yar a ‘Ba Wee ere rm 9012 ‘Splint coms cwemma ST ape0 Thing be Rien ln case at MeKinseys iit were, then the Fiem would not be as successful a iti “The tools may be che same from problem ro prablem, ‘bur you ave to apply them. For instance, in my experience, in 8 ou of 10 pricing problems the answer tuns out be “raise your pries." If you do the facr-based analyss— demand curves, breakeven calculations, budgets—enough times, you see almost invariably that firms should be ras- ing their prices, But if you automatically say that's the answer, you'll get into trouble, because you'll run into an instance where the answer i rally lower your prices.” cee Asa corollary to avoiding cookie-cutter solutions, be earefl about blindly trusting your gut. As you gan experience in business, as you see and solve more and more problems, you wil get 2 fir idea of what works in your industry and what doesn’, Although your gut will often be eight, take aint from former President Re gan “Trust and verify” As a McKinsey alumnus who now works a8 merchant banker pot ite A sharp manager with lot of busines experience ean oem reach the same conclusions as McKinsey—and in a moch shorter ime—by gut instinct, but most executives aren' that good. Hecause McKinsey focuses zo intenly on a problem, ‘often produces a more robust solution than even the best executive can. Most executives will miss a few things beause they don't ake the ime—they usually doa'e have the ime, DEE.US a lye shenian scope comght © 128. yews Proteins Back Group ado Soma arot den ingen ass So, even though your initial instinct may be—and probably ‘tig, take enough time to verify your gut with facts ‘Sie 1 age Tehsil i Tp Sra Con i ap or Vor Ba iit hi hel rd See 920 gists bSanennai SereDrip peck a DON'T MAKETHE FACTS FIT YOUR SOLUTION Avoid the temptation to view your inital bypothesis a the answer and the problem-solving process as an exercise in proving the IH. Keep an open and flexible mind. Don't let a strong initial hypothesis become an excuse for mer tal inflexibility. (On an engagement for 2 major insurance company, the EM. assured his eam andthe clien tha the key restoring the client’ profitably was eliminating “leakage"—the acceptance of cus ‘omer claims without adjustmene. He asigned a new associate © determine the rat of leakage in fe insurance claims over the pre vious chee years. As any good McKinsey associate would, this young maa applied himself sielessly and diligently o his task. Ee combed through stacks of claims in search of leakage. The resus ‘very itl leakage, far less than the EM had predicted. Rather than take these data for what they were—an indication that he needed ro reexamine his hypothesis—the EM simply told the young associate to look again, thi ime in auto insurance land then in macine insurance ... and in business insurane "Nowhere were the expected rats of leakage to be found, (One day, the EM was sting in the team’s onsite HQ lookirg slightly despondent. The team's main contact atthe client poked hishead around the doorpost, “What's the mater, Nick?” he sail Not enough leakage for you?” The moral of chi itl tale is that, no matter how beiliant, insightful, and original you may feel you initial hypothesis ro be, | you must always be prepared to accept thatthe facts may prove I ‘a Bs 1 iy Wg es fom a Tp Sri Comb aT Yo as Iii ht Merril renee et 19.92 iis dyer So dea2 ‘Thing About are rn you wrong. If they do, adjust to the facts: Dont try to pound them into your framework ike squaee pegs into round holes. How do you avoid that trap? The MeKinsey way sro take an ‘occasional step back from the continual grind of fact gathering and snalysis and to ask yourself what you have learned over the pest, week (or two weeks, or whatever). How does the new informa sion fit into your intial hypothesis? Ife doesa’t, how might it change that hypothesis? Doing these litle realiey checks now and then could save you from chasing down Blind ales. ‘Asa final note I should add that che story, although tr, cep resents the exeeption eather dhan the rule at MeKinsey at last in my experience And that EM has long since lef the Firm, ‘MAKE SURE YOUR SOLUTION FITS YOUR CLIENT The most briliant solution, backed up by libraries of dara and promising billions in extra profits, is useless if your cient or business can’t implement it. Know your client. Know the | organization’ strengths, weaknesses, and capabilites—what management can and cannot do, Tailor your solutions with these factors in mind. AA former McKinsey EM who now works on Wall Street tli this story 1 4 i H it i Hi fi H i Hi ‘We were doing a cos cutng rudy fora large financial ins tution. We discovered that they were inthe midst of inking all their offices—they had several hundred around che ‘world—by satelite. This project had begun several yeaes ‘i ty Wang be Tes oie ep Sree Ca Ha Yor Yar Bs ‘Basie nce as Praag 02 [i yet ana SaatDowling on peck 8 previously, and they had managed so far to roll tout 19 bout half thei offices, ‘We determined that with currently availabe technology ‘they could do the same thing for a fraction of che cost using conventional phone lines. By our calculations, chey would ‘have saved $170 milion ona present value bass ‘We took our findings to the senior manager overseing the engagement, the man who had brought us into begin ‘with, and be said “Well that’ tersife, We appreciate that ie could have saved us afew hundred million dollars, but ‘we've already stared doven this oad and, politically just 100 risky. You have to realize that we only havea certain ‘amount of energy inthe organization and frankly, we need ideas that are bigger than this.” ‘Avone level its unbelievable that he didnt accept oo idea, But on another level, we were coming up with other ‘sggestion that would sve the organization half billion ot billion dollars. So this was, fnor exactly chicken eed, jst 4 medium-size payoff for them. I's ratonal response, IFT «an do only thre ching, I'l do the three bigeest. port reer Bock ro, Toe on src McKinsey hires people wih stellar academic records and enforces a rigorous discipline in analyzing problems and struc turing solutions. Because ofthis, a McKinsey ites (especially «| ‘ew hize’s) first instinct isco go hell for Leather after the very best solution. Unfortunately, when academic ideals meet business realities, business realities usually win. Businesses ace fll of real people, with real strengths and weaknesses and limitations. These peop can do only so much with the finite resources available in their ‘organizations. Some chings they just cannot do, whether for poi ical reasons, lack of resources or inbily i ‘ é : 4 i i 3 tf z onrtahowm, oer ora ‘ee i ey Wn Ug yeu ie i ip So Csr ep Yon Trae "it ee hat ron Com 92 ‘Stmrop combo ¥owenan Coos Semperu ‘hiking ot Bates rane ‘As a consultant, you bear the responsibilty for knowing te limitations of you clint; if yous clients your own employer—oe your own business—that responsibilty is doubled. Knowing chow limitations, you must make suee that any recommendations you make ft within chem, SOMETIMES YOU HAVETO LET THE SOLUTION COMETO You The McKinsey rules of problem solving, like all rules, have their exceptions. You will not be able to form an initial hypothesis every time. Sometimes, the client will not know what the problem is, just that there is a problem. Other times, the scope of your project will be so large—or so sague—that starting with an IH will be worthless. Sil other times, you will be breaking new ground and nothing in your experience will point to a solution. Don't panic! If you get ‘your facts together and do your analyses, the solution will come to you. Hamish MeDermot, former EM atthe Fim, counted tis toy Twas on a study where we were tying to improve the per formance ofthe foreign exchange busines ofa major bank; we were supposed to educe costs in the back-office opera- tion by 30 percent: Thad nothing at shat stage thar T could point 1, no inital hypothesis about how to get those costs ‘ut of the organization, Frankly, we knew ver litle about how the back-office processing worked poCC“‘(_CC___C(‘éY ‘lB sy Wang Tes ei Top Sr Co Hap Tos Yar Baa thao bit ual Pd tg cesar Sdnopygh © 18, Hecrantt Preteen tsk rove, Tao lon sro Seem oracle genome sn spre pate ECE Reve cheated cota Dewlpig a dpe a had to interview the woman in charge of back-office procesing along with her senior staff. She dint mean to be ‘unpleasant, although itcertainy fl that way, but she came right out and said, “You've never done this before and you kknow nothing about this business. One of wo things is sing co happen. Either you'll come back with something wwe don't agree with, which will necessarily be wrong, oF you'll listen ro us and come back with what we alzeady know in which ease you've added no value, We understand that you're hee and that you're working on this, but from ‘our point of view it’s a total waste of ous time and che ‘bank's money” Sill she gave us che data we asked for. rturned out that fone product, which represented about 5 percent of thie business, was producing about $0 percent oftheir costs. We ‘were able o fx that. Theyd had no idea this was gong on In subsequent stages ofthe engagement, we were able r0 ‘extend this analysis to other par ofthe busines and we es ily exceeded our targets “The moral of ths story is that an intial hypothesisis nota pre= requisite fr succesful problem solving: Having one will help oege- ‘ize and forward your thinking, but if you can't come up with ons, don’t despaie. Any MeKiasey-ite wil tell you that no business problem is immune to the power of fact-based analysis. Pt together enough fats, combine chem with some creative thinking, and you will ome up with a solution. Se a Seco itt teat Pao ta Se oe tbe dyed26 iki Bie Pn SOME PROBLEMS YOU JUST CAN'T SOLVE SOLVETHEM ANYWAY Eventually, you will run into a brick wall thats rougher than ‘your head. Don't keep pounding: it has no effect om he wall ‘and does your bead no good. My unofficial mentor at McKinsey asked me to work with him en ‘what promised ro be a fun and exciting study. The lent, « major financial institution inthe midst of reorganizing its investment management business, faced severe challenges of heroic dimes: sions thousands of employees, billions of dollars. The McKinsey team included my mentor and my favorite EM. le seemed the per- fect recipe for an enjoyable, challenging McKinsey engagement | “The secipe might have bees right, bu theres left a bad tse in our mouth. Factions within the cline’ senior management pe vented us from doing our job. Dara we asked for arrived lat, oF in an unusable form or nota all People we needed to interview refused to speak with us, The members of che client team vigor- covsly pursued their own agendas a the expense of aching soli- tion, We spent several uncomfortable months on thi eudy and, inthe end, had ro make what recommendations we could, “declare victory." and get out. ‘My team’s experience was hardly unique inthe history of the Firm, The road of problem solving i often stewn with obstacle, Data to prove your hypothesis may be missing or bad. Sometimes businesses realize too late that chey havea problems; by the time MeKinsey, oz anyone, addzesses the problem, the busines is already doomed. ‘The biggest obstacle—he troll guarding the bridges politi como Neymsoieaton | a 3 t f | ony 1998, merret Eres ‘Sie Tey Wage Teheran Tp Sra Comd ap Tor Yor Ba RS he Mel Pda ee 920 pinto 50 Soasa ‘eet “ten nga toma sr poms putea ennfah 9108, wean Prost Bak rap. Yodo le rc EUCEETE Stee ceaer decoys Dowling an peck v ‘The frst thing to understand about polites—and how it can help ‘ot hinder you from doing your job-—is sha businesses ate fllof real people, When you lok at the boxes onan organization chat, you ae really looking at people. When you move those boxes around, you change someone's life. As one former MeKinsey EM. remarked, "Sometimes change management means just that— changing management.” When members of a McKinsey team go into a client, they carry change with them. Some at che client will welcome the bringers of change as white knights riding to the reseue; others will see MeKinsey as an invading army to either flee or dive oxt, depending on their power in the organization. As one former MeKinsey-ite put it, “le was a rate engagement when chece wasn’ a least one sector ofthe clint organization that di rot want us theze and did not want us to come up with areal answer In mont cases, when senior management brings in McKinsey, ‘enough players in che organization will cooperate willingly and MeKinsey can be effective. A few malcontents may grumble even caus trouble, but in the end they will be ether converted ro McKinsey cause or bypassed. Sometimes, however, one pow:r- fol faction in an organization callsin MeKinsey against the wishes of another powerful faction. That's when trouble arses, a5 We found out, ‘You have several options to pick from when confronted by a problem cht seems too difcul ro salve. [Redefine the problem. You ean tll your cient thatthe probs lem is not X, it's ¥. This is especially useful when you know that solving Y will add lot of value, where a eying to wrestle wth X would costa lot of ime and resources for litle result. Ifyou make this switch very erly on, you show great business judgments Ifyou do ic after several weeks” work, you risk being accused of cop out ‘Baal Bi Wa de Tre Dep Sree Cr Tor ar Ba ‘Bini incerta rare ‘Ep yb SopPy Twig et ies Pen “Tweak your way toa solution. Sometimes you will come up witha great solution that you know the clint organization ean- rot implement. This is especially true with organizational change—it is easy to devise an optimal organization, but you use ally have to deal with the personnel resources that the clieat aleady has. When that happens to you, expand your time horizon, Don’ worry about implementing your solution immediately. As people leave the organization, you ean “tweak” your way to your ‘optimum over time. ‘Work through the polities. Even political problems ate sole ble, Most people in busines aze rational, a least in their busines ‘conduct. They react ro incentives. Therefore, when you face poli ical opposition, i usually means chat your solution has negative implications for someone in che organization. So politics is just people acting in thee own interests. ‘To work through the pois, ou must think about how your solution afet the players in an organization. You must then build «consensus for change that rakes account ofthe diferent incen- tives and organizational factors driving the politics. Consensis building may require you to change your solution to make it acceptable. Do i. Remember that politics isthe art ofthe posi- ble, and ies no good devising the ideal solution ifthe len refuses como igure irs paseo pacar ep oprah © 188, Ween Proteasant sk rou. Tae deco Pane. ‘al sy Wing te Tas om Tp rag Cot pT Yor as Ba Ht neem Pan Sk op Fs ee yewelle Slammer Sa Slee2 Thang Aiur anes bln DON'T BOILTHE OCEAN | Work smarter, not harder. There's a lot of data out there relating 10 your problem, and alot of analyses you could do Ire most of them. McKinsey gathers enough acts 9 prove or disprove a hypothe sis or support or refute an analsie—and only enough face. This the fip side of fact-based analysis in a business site | sion, Anything more ia wast of time and effort when both ax precious commodities. Thad this leson brought hore tome lt ne night while was deafsing a “fact pack” ona clients competion. had gathered ‘mountain of data and was trying to wring outa few new insights from i My EM, Vik, walked into my ofce, briefcase and coat hand, and asked ow my work was goin, Lol him it was going | well but thought I could pull gether afew more chars. He Picked up my draft leafed through it and sud “Ean, is eleven (clock. The lent wil lve ti, No one wl be able to abso ‘more than you have ber, Calta day. Dont ol the cea.” We shared aca home “Don't bol the ocean” means doo try to analyze everything Be elective igre ou he BTSHEREY what you ae doing, Know ‘when you have done enough then top: Orkerwe, you wll spend 2 lor of tine and effort for very Be eur, lice boling he ace to get a handful fs "a a iy Wy eign of ae WaT Sas Ca ap Ys Yar Ba Be ei itt eat gaan toe og Se teeta SelesCg UTStar ce conve 10120 on Other Re Liey 3 FIND THE KEY DRIVERS ‘Many factors affect your business. Focus om the most impor- tant ones—the key drives. In any MeKinsey team meeting where problem solving ison te agenda, someone will use che inelegant phrase “key divers,” as in, “Vik, I think these ar the key drivers ofthis issue.” In other words, there may be a 100 different factors affecting the sales of ‘our widgets—weather, consumer confidence, aw matevalpices— but the chee most important ones are X,Y, and Z. We'll jgnore the res. Engineers earn something called the Square Law of Comps- tation, estates that for every component ofa system—for every audicional equation ina problem—the amount of computation requited o solve the system increases a least as fat asthe square ofthe numberof equations In other words, if the complexity of| your problem doubles, the time it takes to solve it quadruples— i tem contains millions of objects, all aving gravitational effects ‘on one another. When analyzing planetary motion, start by ignoring most ofthese objects." Focusing on the key divers means driling down co the core ofthe problem, rather chan picking the whole problem apart pie by piece, layer by layer. Then, you can epply thorough, fact-based analysis where it will do the most good and avoid going down blind alleys toless you make some simplifications. For example, ours ity an problem saving, ee GeridM. Weinberg, hn Pode a Gene Stems ‘ee Yor ey SoS) ‘Bia one ween ono Sat agp neon a To Yow es, ie iy otshaamann Stapfa Takin Aho Brine Preble Stacie sie heyday pow co cope Ueeves you de. retes efor, harp you om bot. BK ferret H | ‘THE ELEVATOR TEST Shreiisipeane = Caer pinta tina w tor “ee Phecik cghly that you can explain it larly and precisely to youre @D| anne dlen (or estomer or mestor)in 30 seconds Ifyou can ao gh alcte that, thn you understond what yore dong well enough to ‘al your solution Imagine it's time fr that big, end-of-engagement presentation, You and your team have been up until 2 sam, putting together your bse books,” making sure hae every # has been dotted and every ¢ crossed. You're all wearing your best suits and tying ro look on te ball, The senior executives of your Fortune 50 client, anxious 10 hear MeKinsey’s words of wisdom, are taking their places arourd the boardroom table onthe top floor ofthe corporate skyscraper, “The CEO strides into the room and says, "Sorry, folks. [ean' stay. ‘We havea crisis and I have to go meet with our lveyers." Then be sums to you and says, "Why don’ you ride down in the elevator ‘with me and tll me what you've found out?” The rie will take about 30 seconds, In that ime, can you rellthe CEO your solution? (Can you sll him you solution? That's the elevator test. “Many companies use the elevator test (or something similar because i an excellent way of making sure that thee executives tts eet re 7 Fy 7 2 any ‘a eH wn ag aso Wa Sra Coda eT Yor Bess eat epee ops ei ects Sdg iso parnace omyioh 918, Wom Prete ook Group. Toon on een Sori crater uratrne sos pres opted ESSER Oo ar yecaphanir se copy 10/20 en the Rae Lie By s time gets used efficiently. Procter & Gamble els its managers © write one-page memos. A Hollywood producer wil ela screen: write to “ive me the bullet” ona new seri if after 30 second, the produces likes what she's heard, the weiter wil get a chance ‘oval further, and maybe make a sale. Jason Klein insticuted ete elevator est when he took over as president of Field & Steam: _My sales force could nor explain the magazine to cusromers ‘Our adverisement space was shrinking. Then I rained my cemtte eles force onthe elevator test. I challenged them ro explain the magazine to mein 30 seconds, I became a val able rool for them, and our ad base has grown every year How do you encapsulate six months’ work in 30 seconds? Sear withthe issues tha your team addeesed. The clint wan to know the recommendations for each issue and the payott If you have alot of recommendations, stick tothe three most mportant—the ones with the biggest payoffs. Don't worry about the supporting data; you can talk about that when you have more time For example, your analysis shows that a manufacturing clicet can't sell enough widgets Because is sales fore is organized by te tory when it should be oxpanized by buyer category. You have tote of data illuctraing thts analytes of salgpoople by buyer pH, buyer interviews, fed visits o retail and wholesale outlets, and so forth. When you'r on that elevator side, ust tell he CEO, “We ‘hink you can boost sales of widgets by 5O percent in three yeas Ifyou reorganize your sales force by buyer category. We can take bout the details later Good ack withthe lawyers." ‘ae Bae gy Wg ech oe Wa pS Commi op Ys Ya bas ee Oi lit at seo ek Sesrapetnttacesmcn Sues Sepedprado pce cll ee pcee ESTER ATE yeah de cyt % ‘Thing out Brine Pale PLUCK THE LOW-HANGING FRUIT Sometimes in the middle of the problem-solving process, ‘opportunites arse to get an easy win, to make immediate improvements, even before the overall problem bas beou solved. Seize those opportunities! They create litle victories for you and your team. They boost morale and give you ‘added credibility by showing anybody who may be watch- ing that you're on the ball and mean business. | “Whenever possible, MeKinsey consultants put this doctrine into practice. Clients can ge very impatient fora resul during the sc ‘months (or more) that ig McKinsey engagement can last. Giving the client something practical before the end helps educe the pres sce on the eam, ‘Ar my stockbroker client, after we had derived a number of insights thanks to 80/20) from our analysis of sales and eading dats, we wanted to communicate our findings tothe senior man | agers ofthe institutional equities department. We et up a meeting with the deparemene head and the heads of al the business nis in the division: als, rading,reseazch, and soon, Since I had taken the lead in the acual analysis ofthe daa, goto present our findings. They hic this group of very experienced ‘Wall Stroot executives like hammer. The client had no idea just how inffcien its operation was. ‘The presentation had two important effects. Fest ieconvinced those executives who had nor been particularly keen on McKie sys presence in the fist place that they had a problem and we could help solve it. Second, because I had presented the findings, ‘heir opin of me rose quite sharply and my ob became alot ea Se aE ep oe OTT Tor ‘Srl tay stonenan tape{£0120 cn thr Pee ” Jer Before the meting, I as some smart-ass MBA poking around their business. After the meeting, I was someone who was work ing for them to solve thee problems. By plucking the low-hanging fui, by essing the temptation twhoard out information atl some bigend-of-srudy presentation, ‘we made out client more enthusiastic, our jobs eases, and ou selves happier ‘This rule is ceally about satisfying your customer ina long term relationship. Your customer could be the purchaser of your products, or i¢ could bea client for your services, or ie could be your boss. Whoever itis, it pays to keep im happy and let hit know that he is your top priority If you are on, say, a sofewate design project with a three-month lead time and you've put together a usable program that solves pat of the problem in ro weeks, show it ta your boss, Don's wai! Solving only part of problem can sill mean increased profits. Just don't lt anyhody think you've given up on a complete solution. Those lite wirs help you and your customers, Ted ae eece conan 9100. era Preston Sook Group. ado oe Seznis rntcl Te gaara sn spss pcs ‘enophstoc ae copeah erence MAKE A CHART EVERY DAY During the problem-solving process, you learm something new every day. Put it down on paper. 1 will belp you push Ihave erystalized it on the page, you won't forge it. Making a chare every day may strike you as somewhat anal-rtes- tive Iris. Then again, when you are trying to craft fats into sol tions thats nor so bad ‘your thinking. You may use it, or you may not, but once you << Fe ‘ee aig Wn Tenge ech Wap So Coo eV Var uaen—— te Ue ceca atone oe B95 Spits baoeenmn DoS Stpes” “ining et Biers In the course of atypical day at MeKinsey, you could stat with «ick brainstorming sesion a9 a.m, move on toa client inter view at 10, a factory tour at 11, and then a sandwich lanch with your diretor. You might follow this with more client inerviews, an end-of day team meeting, and then a quick trp down to Wha ton to participate ina eceuting semiae. In che midst of all his, itis very easy forthe faces to blend into one another lke pools of ‘much ina day. Thete is nothing quite so frustrating as looking back over the course oft day of week and realizing, not thar you haven't come up with any end products but that what you have ‘some up with is worthless in rem ofthe problem at hand, ‘As one former McKinsey EM told me, "Perhaps the most vale- able ching learned during my time atthe Firm wast think about the big pictare—to take a step back, figure out what Pm trying © achieve, and then look at whatever 'm doing and ask mysel, ‘Does this really mater?™™ JUST SAY, “I DON’T KNOW” ‘The Firm pounds the concept of professional integrity nto is associates from their first day om the job, and rightly so. One importane aspect of professional integrity is honesty —with ‘your clients, your team members, and yourself. Honesty Includes recognizing when you haven't got a clue. Admiting that is alot less costly than bluffing. was the morning ofan important progress meeting at ou clint, «4 Fortune 50 manufacruring company, The team and John, ott ED,* were going over the various sections of the presentation. ‘a Bs sy Wa mast Wa Tp Sr Como Ys Yar Daas RSH uted ran Sn ip [is Sees Staneonn Sd90/20 a Oa Rt Ley a had aleeady been through my piece of; [had been up uni 4 a.m. lcting it ready and I was exhausted, As the discussion moved 19 another section, one that I had nothing to do with and knew lid about, my brain stactd slipping into tac blsful place known 3s sleep I could hese the other members of the team discussing va 0 poins, but their words slipped away fom my mind lke water through a childs cupped fingers. Suddenly, my reverie evaporated as John asked me, "So, Ethan, ‘what do you think about Suzie point?" Momentary shock and {ear yielded to concentration a tied to recall what had just been said. Years of Ivy League and business school reflexes took over and Lcame out witha few lines of general agreement. OF course, ‘what I said mighe just as well have come out ofa horse's backside Thad told John, *'m nor realy sure—t haven looked at ths issue before,” I would have been fine. Even if Thad said, "Sor just lost it for @ minuze,” he would have understood afterall he had been through exactly the same experince, like every other MeKinsey.te, Instead, I red 0 fake it and ended up slipping 2 | ry own horsefathers. Ache end ofthe engagement, several weeks ates, the eam bad its final parey. We went out to TGI Friday's, ae alot of nachos, and drank alot of bet. Then the EM began presenting each of te seam members with presents of a rude andior hurmorous nat. For my gift he handed me litle desktop piture fame around tte following words, neatly printed in the MeKinsey offical font: “Just say, T don't knows “This is sage advice, and that picture frame remains on my desk to this day enti, Teynsohesie ae covreah eorauel Tonnes ose ol pater apt commen amt eres Fe Bn iy We se oe Wai Sos Can To Ts as te Ou Ut al eo ti 1 Sree et tikrme becom Sema“ ‘hadi it Ries ne DON’T ACCEPT “I HAVE NO IDEA” People always have an idea if you probe just abit. Aska few Pointed questions—yow'll be amazed at what they know. Combine that with some educated guessing, and you can be swell along the road to the solution. “or pemmcor lt you ask people a question about their business and they el you, *L have no ides done ust walle away in defeat. "Ihave no idea” isa BE PREPARED: WRITE AN INTERVIEW GUIDE When you go into an interview, be prepared. You may have only 30 minutes with a person whom you may never see again. Know what you're going to ask. \When I asked McKinsey alumni for their best advice on inte Views, every single one of chem said, "Write an inervew guide.” ‘Many people resent being interviewed, or at least begrudge you the time chat you are taking from their day. guide is your best, +00! for getting what you waat from inceviewees and for mab ing che bes use of your time—and this. ‘You must think on wa levels when constructing your eve. ust, and obviously, what are the questions co which you need answers? Write them all down in any order. Second, and more important, what do you really need from this interview? Whee ace you tying to achieve? Why are you talking to this person? Defining your purpose will help you put your questions inthe right order and phrase them correctly. Tr helps to know as much as possible about the interviewee in advance. Is she a prickly CEO who might bite your head off if you ask a sensitive question? Or is she a middle-level managee ‘whose pleas fr change in her organization have gone uneedec? Both might know the same piece of information, but you'd spproach each one differently. ‘At McKinsey we were taught that, as a rule, an interview should sart with general questions and move on to specific one. Don's dive right into a sensitive area like "Wha are your respon sibiliies2” of "How long have you been with the company?” Start with anodyne questions about, say, che industry overall ‘ne Re We WU eign oie it pangs Car pT Yar aes eo Ue eet atone 103.9 ‘Spitnreoetbinsnmn DoS See 8oer atu em tinge un pine pes Cems 108 Ment rotsen Bet rep. Toso ach oe Uc lstoee operation se copyeah © tng She Bes Pies ‘This will help the interviewee “warm up" and allow you w develop rapport. ‘When deciding on which questions ro ask, you might want © include some to which you know the answer This may sound ‘counterintuitive, but i's really very useful, On questions of fac, | asking a "singer" will give you some insights into the interviewee’ honesty and/or knowledge. For complex issues, you may think you “know the answer, but there may be more than one; you should find out as many as posible. | ‘Once you've written your guide, look at it and ask yourself, “Whar ae che three things I mose want to know by the end of te interview?" These are the things you will focus on when you po into the interviewees office, the thee things that you wil ry your hardest ro obeain before you leave. Sometimes you wont even get those answers sce “Diffcle Interviews” later inthe chapter, a | sometimes they'll ome easily. Anything moe is gravy. Finally, every interview guide should conclude with what Leall the protorypical McKinsey question. When you've asked all your questions, oF you're running out of time, put away your guide and ask the interviewee if there's anything ls he'd ike ro ell you 0 any question you forgot to ask. As often a not, the interviewee willsay no, bu every once ina while you'l strike paydit. Remen: ber that, chances are the people you interview know thee organ ations; their business units, or their departments beter than yeu o. They may know which problems ae eluding senior manages, ‘whos pushing which agenda, or where the bones are buried. Ard somecimes, if you're icky, they'll you a aE a pn OT or Sh Stayton ce sltonconnate Prosi! Bok Oooo edhe Seam eras tre ome tess a uatewenige Soetoro nding tion: a WHEN CONDUCTING INTERVIEWS, LISTEN AND GUIDE When you're picking people's brains, ak questions and ther let them do the talking. Most people like to talk, especially if you et thems know you're interested i wwhat they're saying. Keep the interview om track by breaking in whem necessary McKinsey consultant recive lot oftetning in interviewing tech- niques. The frst thing we were taught was “always lt the inte viewee know you are listening.” We did ths by filling the gaps in the interviewee’ conversation with verbal placeholders such ¢s “yes, oF *I see," and even ust “ub-hub” (this particular choice like to eal ehe MeKinsey grunt). Uh-huh may no seem like much, bor ie shows tha you're paying atenion (even when you'e ot}, and i gives the other person a chance ro gather his thoughts and catch his breath ‘We also learned ro communicate our interest through body la guage. When the interviewee was speaking, we leaned rowed her slighly. When she completed a sentence, we nodded. And we always took notes. Even if the interviewee was babbling {and this happened often enough), we had our notepads and pens our and ‘wrote things down. Like the MeKinsey grunt, note caking implied we were paying attention and kept us prepared in ease the inter viewee di say something imporant. ‘This technique could be carted 10 far, of course. Accordieg to one piece of Firm lore, two consultants went to interview a webusioielle et pet cnt pe ke atresia eden te Sd tc cee tr ar ‘a ae nein Rong eee Wa pa Cr You Yar aa thoi tt carat a te Sp rapt nese Bbsali STE Uo anys aphenin ce copy copmante 00. a Whig Se Bits Pm, high-level executive ata client, The engagement manager made his introduction, then started asking questions. The executive answered in detail Al ehrough this, the associate nodded and ierocted with "yes," “ubchub,” and “Isee" while taking nows furious pace ust ashe had been taught, but he never asked any questions imsel. The EM asked follow-up questions and the associate just kept on nodding and saying “ub-hub.” When the interview ended, che EM thanked the exscutive for his time and the two consultants got up to leave. As they were shaking hands, the executive pointed to che associate and asked the EM, “Does ke speak English?” ‘When McKinsey consultants conduct interviews, is because they want access the information, experience and aneedotes in other person's heads, Consultants ae chere eo listen, not total They need to semember that the other person hs 4 separate agenda, and needs ro be kepe on tack. The process can somecines prove dilficul,Lonce had to interview a purchasing manager from «client's plane in Mdaho. He knew all about che plant's supplies, customers, input requirements, and manufacturing proceses, bet all he cared about was fishing—Aly fishing. “Did I fly fish?” be ‘wanted to know. I should ery it 1 was ever in che Pocatello ares, he could set me up. You get the picture. I elk sbi bad geting hin ‘off his favorite subjost, but I as thereto get information, not ‘swap fish stoves. “The main thing to remember when trying to get informatien fom other is that chey need to feel you are listening and hae you'r intersted in whar they have to say. Use postive body la _muage and always tke notes. One final erick: If you want people say more chan they have if you hink they have left out something important but you'ee not sure what its, say nothing. Let the sileace hang, Nature abhors vacuum, and so do most peopl. ‘Chances are they wil tare calkng, just ro fill he gap. I chey have "ie a tay Ry Ug Teco ePaper ep Fon Tr as SAS GH it eet aed te Be SOM demotion atu Seope ii a eng tere o been giving you « prepared “script,” they will probably dropit, because the one thing they were not prepared for was silence. Tey iand se, I's suprisingly effective SEVEN TIPS FOR SUCCESSFUL INTERVIEWING Always think strategically when conducting an interview. You have a goal to reach and limited time to reach it, Here are seven tried-and-tested stratagems to help you get what ‘you want from an interviewee. 1. Have the interviewees boss set up the mecting. Going through the boss tells the interviewee thatthe interview is impor: tant, Hell be less likely to jerk you around ithe knows his boss wants him to talk ro you. 2. Inerview in pairs. 1° very dificult to conduct an effete interview on your own. You may be so busy tang notes chat it ‘becomes difficult co ask the right questions. You may miss nen verbal cues thatthe interviewee is giving. Sometimes, ii weil for a pair of interviewers r0 “tag team" —ewitch roles from gue tion poser to note taker during the session. The approach is espe cially effective when one ofthe interviewers has specific knowledge ‘om certain issues that will he covered. Furthermore, iis always ‘seul ro have wo diferent views of what actually happened in the interview. Jt make sure that whoever writes up the interview notes corroborates them with the other interviewer. 3, Listen; don’ lead. In most interviews, you are not looking for yes-orno answers to your questions. You want exhaustive answers—as much information as posible. The way to ee them ‘aa eigen Fw a Como FT Tar Ba ata pir. ann Sp“ king Se Bsns Pos isco listen, Talk as lle as possible, just enough 10 keep the ine | view on tack, Remember that che interviewee prabably knows a lor more about her business than you do, and most ofthe infor. mation she gives you wil be usefl one way or another. Here's another trick for keeping the information flowing. Ade open-ended questions. If you ask yesor-no or multiple choice ‘qvestons hat’ all you wil get. For example, suppose you wane find out when a store's busiest season is, You think itis either sum rer oe winter, but you'e nor sue. your busiest season summer or winter she mighe say summess she might say winters o she might say, "Acrually, i's spring,” in ‘which case, you've just highlighted your lack of knowledge about her business. IFyou ask hes, "What is your busiest season?” ste will give you the answer, and probably in more detail than if you tive her multiple choice—for example, “We're busiest in te spring, specifically at Easter” By asking the open-ended question, you get a much beter result 4. Pasaphrase, paraphrase, paraphrate, Before going ovt cn lmeviews, every MeKinsey constant strained to repeat back sub- jeor's answers in sighly diferent form. I cannot overstress how Jmpoctan this is. Most people do nor think or speak ina competey structured way. They ramble, they digress, they jumble important facts among erlevancies If you repeat cheir own words bacl 10 ‘them—idealy with some structure appied—then they an el yeu ‘whether you understood them coreecd: Paraphrasing also gives the | interviewer a chance to add information or amplify important poins 5. Use the indirect approach. An EM had on his team a new associat, fresh out of the Navy The two had put together a very clear interview guide and had agreed on a specific st of goals for an interview with a middlelevel manager a thie client, so the EM lec che associate take the lead, The associate proceeded to gil the manager agaressivly in order to get precisely what he wanted, in nl pao scp en pam you ask the store manager, “Is cok Grou. Tote =o ennngenaores ECE UU eee sneaesoconytoe copyeahe sme eet aero "el Bi Wa ing Tees any Sr Cote HT Yar Bes Resi it nce rn Sa 0 tas ay shana Sa dg 1enti Satomi ere eee, ETERS Re lyecaenn se copy nding eee ts ifit were an interrogation rather than an interview. As you might imagine, the interviewee was rattled; he became defensive and essentially refused to cooperate “The moral ofthis tale is "Be sensitive co the interviewee’ ee ings." Understand thatthe person may fel threatened. Don't dive right into the rough questions. Ifyou have to dance around she important issues for a fw minates, that’s OK, Take time to make the interviewee comfortable with you and the interview proces (for an in-depth discussion, see the next section, 6. Don't ask for too much. There for everything the interviewee has, Fits, you might ger it. When you write your interview guide, you narrow down your goals ons not to ase the wo or three most important questions. Ifyou then ask the interviews for the sum rotal of his knowledge ofthe widget indus 1, you may find yourself wading through a lor of information get what you really need if you even find i aval. Second, you want to stop short of the straw that breaks the ‘camels back. Remember, being interviewed, especially in the con- text of a business problem, is an uncomfortable experience fer many, Ifyou compound that discomfort by peessing roo hard, you ray find that the interviewee becomes uncooperative ot even hos tle You never know when you may want to come back otis pe sn for more information, a0 don' shut the door 7. Adopt the Columbo tactic. Ifyou watched TV inthe 1970s, you may remember Peter Falk's renchcoat wearing decective,Liew- tenant Columbo. After he finished quizzing a murder suspect bout her whereabouts onthe night in question, he would pick up his rumpled raincoat and head out the door, As he reached tke threshold and was about leave, he would run around, stick his finger wp to his temple, and say, “Excuse me, ma'am, bur there’ something I forgot ask.” This question invariably gave Columbo the answer he needed 0 figure out who di it. FRE Nay Wayne Ts ie Vrs ap rar Cot Tp ox Tr st