The Effective Decision - Harvard Business Review

Effective executives do not make a great many decisions. ... They try to find the constants in a situation, to think through what is strategic and generic ... Youhave1freearticlesleftthismonth. Youarereadingyourlastfreearticleforthismonth. Subscribeforunlimitedaccess. Createanaccounttoread2more. Decisionmakingandproblemsolving TheEffectiveDecision Effectiveexecutivesdonotmakeagreatmanydecisions.Theyconcentrateonwhatisimportant.Theytrytomakethefewimportantdecisionsonthehighestlevelofconceptualunderstanding.Theytrytofindtheconstantsinasituation,tothinkthroughwhatisstrategicandgenericratherthanto“solveproblems.”Theyare,therefore,[…] by PeterF.Drucker by PeterF.Drucker FromtheMagazine(January1967) Tweet Post Share Save GetPDF BuyCopies Print Tweet Post Share Save GetPDF BuyCopies Print Leerenespañol Leremportuguês Effectiveexecutivesdonotmakeagreatmanydecisions.Theyconcentrateonwhatisimportant.Theytrytomakethefewimportantdecisionsonthehighestlevelofconceptualunderstanding.Theytrytofindtheconstantsinasituation,tothinkthroughwhatisstrategicandgenericratherthanto“solveproblems.”Theyare,therefore,notoverlyimpressedbyspeedindecisionmaking;rather,theyconsidervirtuosityinmanipulatingagreatmanyvariablesasymptomofsloppythinking.Theywanttoknowwhatthedecisionisallaboutandwhattheunderlyingrealitiesarewhichithastosatisfy.Theywantimpactratherthantechnique.Andtheywanttobesoundratherthanclever. Effectiveexecutivesknowwhenadecisionhastobebasedonprincipleandwhenitshouldbemadepragmatically,onthemeritsofthecase.Theyknowthetrickiestdecisionisthatbetweentherightandthewrongcompromise,andtheyhavelearnedtotellonefromtheother.Theyknowthatthemosttime-consumingstepintheprocessisnotmakingthedecisionbutputtingitintoeffect.Unlessadecisionhasdegeneratedintowork,itisnotadecision;itisatbestagoodintention.Thismeansthat,whiletheeffectivedecisionitselfisbasedonthehighestlevelofconceptualunderstanding,theactioncommitmentshouldbeascloseaspossibletothecapacitiesofthepeoplewhohavetocarryitout.Aboveall,effectiveexecutivesknowthatdecisionmakinghasitsownsystematicprocessanditsownclearlydefinedelements. SequentialSteps Theelementsdonotbythemselves“make”thedecisions.Indeed,everydecisionisarisk-takingjudgment.Butunlesstheseelementsarethesteppingstonesofthedecisionprocess,theexecutivewillnotarriveataright,andcertainlynotataneffective,decision.Therefore,inthisarticleIshalldescribethesequenceofstepsinvolvedinthedecision-makingprocess. 1.Classifyingtheproblem.Isitgeneric?Isitexceptionalandunique?Orisitthefirstmanifestationofanewgenusforwhicharulehasyettobedeveloped? 2.Definingtheproblem.Whatarewedealingwith? 3.Specifyingtheanswertotheproblem.Whatarethe“boundaryconditions”? 4.Decidingwhatis“right,”ratherthanwhatisacceptable,inordertomeettheboundaryconditions..Whatwillfullysatisfythespecificationsbeforeattentionisgiventothecompromises,adaptations,andconcessionsneededtomakethedecisionacceptable? 5.Buildingintothedecisiontheactiontocarryitout.Whatdoestheactioncommitmenthavetobe?Whohastoknowaboutit? 6.Testingthevalidityandeffectivenessofthedecisionagainsttheactualcourseofevents.Howisthedecisionbeingcarriedout?Aretheassumptionsonwhichitisbasedappropriateorobsolete? Letustakealookateachoftheseindividualelements. TheClassification Theeffectivedecisionmakerasks:Isthisasymptomofafundamentaldisorderorastrayevent?Thegenericalwayshastobeansweredthrougharule,aprinciple.Butthetrulyexceptionaleventcanonlybehandledassuchandasitcomes. Strictlyspeaking,theexecutivemightdistinguishamongfour,ratherthanbetweentwo,differenttypesofoccurrences. First,thereisthetrulygenericevent,ofwhichtheindividualoccurrenceisonlyasymptom.Mostofthe“problems”thatcomeupinthecourseoftheexecutive’sworkareofthisnature.Inventorydecisionsinabusiness,forinstance,arenot“decisions.”Theyareadaptations.Theproblemisgeneric.Thisisevenmorelikelytobetrueofoccurrenceswithinmanufacturingorganizations.Forexample: Aproductcontrolandengineeringgroupwilltypicallyhandlemanyhundredsofproblemsinthecourseofamonth.Yet,whenevertheseareanalyzed,thegreatmajorityprovetobejustsymptoms—andmanifestations—ofunderlyingbasicsituations.Theindividualprocesscontrolengineerorproductionengineerwhoworksinonepartoftheplantusuallycannotseethis.Hemighthaveafewproblemseachmonthwiththecouplingsinthepipesthatcarrysteamorhotliquids,andthat’sall. Onlywhenthetotalworkloadofthegroupoverseveralmonthsisanalyzeddoesthegenericproblemappear.Thenitisseenthattemperaturesorpressureshavebecometoogreatfortheexistingequipmentandthatthecouplingsholdingthevariouslinestogetherneedtoberedesignedforgreaterloads.Untilthisanalysisisdone,processcontrolwillspendatremendousamountoftimefixingleakswithoutevergettingcontrolofthesituation. Thesecondtypeofoccurrenceistheproblemwhich,whileauniqueeventfortheindividualinstitution,isactuallygeneric.Consider: Thecompanythatreceivesanoffertomergefromanother,largerone,willneverreceivesuchanofferagainifitaccepts.Thisisanonrecurrentsituationasfarastheindividualcompany,itsboardofdirectors,anditsmanagementareconcerned.Butitis,ofcourse,agenericsituationwhichoccursallthetime.Thinkingthroughwhethertoacceptortorejecttheofferrequiressomegeneralrules.Forthese,however,theexecutivehastolooktotheexperienceofothers. Nextthereisthetrulyexceptionaleventthattheexecutivemustdistinguish.Toillustrate: ThehugepowerfailurethatplungedintodarknessthewholeofNortheasternNorthAmericafromSt.LawrencetoWashingtoninNovember1965was,accordingtofirstexplanations,atrulyexceptionalsituation.Sowasthethalidomidetragedywhichledtothebirthofsomanydeformedbabiesintheearly1960s.Theprobabilityofeitheroftheseeventsoccurring,weweretold,wasoneintenmillionoroneinahundredmillion,andconcatenationsoftheseeventswereasunlikelyevertorecuragainasitisunlikely,forinstance,forthechaironwhichIsittodisintegrateintoitsconstituentatoms. Trulyuniqueeventsarerare,however.Wheneveroneappears,thedecisionmakerhastoask:Isthisatrueexceptionoronlythefirstmanifestationofanewgenus?Andthis—theearlymanifestationofanewgenericproblem—isthefourthandlastcategoryofeventswithwhichthedecisionprocessdeals.Thus: WeknownowthatboththeNortheasternpowerfailureandthethalidomidetragedywereonlythefirstoccurrencesofwhat,underconditionsofmodernpowertechnologyorofmodernpharmacology,arelikelytobecomefairlyfrequentoccurrencesunlessgenericsolutionsarefound. Alleventsbutthetrulyuniquerequireagenericsolution.Theyrequirearule,apolicy,oraprinciple.Oncetherightprinciplehasbeendeveloped,allmanifestationsofthesamegenericsituationcanbehandledpragmatically—thatis,byadaptationoftheruletotheconcretecircumstancesofthecase.Trulyuniqueevents,however,mustbetreatedindividually.Theexecutivecannotdeveloprulesfortheexceptional. Theeffectivedecisionmakerspendstimedeterminingwhichofthefourdifferentsituationsishappening.Thewrongdecisionwillbemadeifthesituationisclassifiedincorrectly. Byfarthemostcommonmistakeofthedecisionmakeristotreatagenericsituationasifitwereaseriesofuniqueevents—thatis,tobepragmaticwhenlackingthegenericunderstandingandprinciple.Theinevitableresultisfrustrationandfutility.Thiswasclearlyshown,Ithink,bythefailureofmostofthepolicies,bothdomesticandforeign,oftheKennedyAdministration.Consider: Forallthebrillianceofitsmembers,theAdministrationachievedfundamentallyonlyonesuccess,andthatwasintheCubanmissilecrisis.Otherwise,itachievedpracticallynothing.Themainreasonwassurelywhatitsmemberscalled“pragmatism”—namely,theAdministration’srefusaltodeveloprulesandprinciples,anditsinsistenceontrainingeverything“onitsmerits.”Yetitwascleartoeveryone,includingthemembersoftheAdministration,thatthebasicassumptionsonwhichitspoliciesrested—thevalidassumptionsoftheimmediatepostwaryears—hadbecomeincreasinglyunrealisticininternational,aswellasindomestic,affairsinthe1960’s. Equallycommonisthemistakeoftreatinganeweventasifitwerejustanotherexampleoftheoldproblemtowhich,therefore,theoldrulesshouldbeapplied: ThiswastheerrorthatsnowballedthelocalpowerfailureontheNewYork–OntarioborderintothegreatNortheasternblackout.Thepowerengineers,especiallyinNewYorkCity,appliedtherightruleforanormaloverload.Yettheirowninstrumentshadsignaledthatsomethingquiteextraordinarywasgoingonwhichcalledforexceptional,ratherthanstandard,countermeasures. Bycontrast,theonegreattriumphofPresidentKennedyintheCubanmissilecrisisrestedonacceptanceofthechallengetothinkthroughanextraordinary,exceptionaloccurrence.Assoonasheacceptedthis,hisowntremendousresourcesofintelligenceandcourageeffectivelycameintoplay. TheDefinition Onceaproblemhasbeenclassifiedasgenericorunique,itisusuallyfairlyeasytodefine.“Whatisthisallabout?”“Whatispertinenthere?”“Whatisthekeytothissituation?”Questionssuchasthesearefamiliar.Butonlythetrulyeffectivedecisionmakersareawarethatthedangerinthisstepisnotthewrongdefinition;itistheplausiblebutincompleteone.Forexample: TheAmericanautomobileindustryheldtoaplausiblebutincompletedefinitionoftheproblemofautomotivesafety.Itwasthislackofawareness—farmorethananyreluctancetospendmoneyonsafetyengineering—thateventually,in1966,broughttheindustryundersuddenandsharpCongressionalattackforitsunsafecarsandthenlefttheindustrytotallybewilderedbytheattack.Itsimplyisnottruethattheindustryhaspaidscantattentiontosafety. Onthecontrary,ithasworkedhardatsaferhighwayengineeringandatdrivertraining,believingthesetobethemajorareasforconcern.Thataccidentsarecausedbyunsaferoadsandunsafedriversisplausibleenough.Indeed,allotheragenciesconcernedwithautomotivesafety,fromthehighwaypolicetothehighschools,pickedthesametargetsfortheircampaigns.Thesecampaignshaveproducedresults.Thenumberofaccidentsonhighwaysbuiltforsafetyhasbeengreatlylessened.Similarly,safety-traineddrivershavebeeninvolvedinfarfeweraccidents. Butalthoughtheratioofaccidentsperthousandcarsorperthousandmilesdrivenhasbeengoingdown,thetotalnumberofaccidentsandtheseverityofthemhavekeptcreepingup.Itshouldthereforehavebecomeclearlongagothatsomethingwouldhavetobedoneaboutthesmallbutsignificantprobabilitythataccidentswilloccurdespitesafetylawsandsafetytraining. Thismeansthatfuturesafetycampaignswillhavetobesupplementedbyengineeringtomakeaccidentsthemselveslessdangerous.Whereascarshavebeenengineeredtobesafewhenusedcorrectly,theywillalsohavetobeengineeredforsafetywhenusedincorrectly. Thereisonlyonesafeguardagainstbecomingtheprisonerofanincompletedefinition:checkitagainandagainagainstalltheobservablefacts,andthrowoutadefinitionthemomentitfailstoencompassanyofthem. Effectivedecisionmakersalwaystestforsignsthatsomethingisatypicalorsomethingunusualishappening,alwaysasking:Doesthedefinitionexplaintheobservedevents,anddoesitexplainallofthem?Theyalwayswriteoutwhatthedefinitionisexpectedtomakehappen—forinstance,makeautomobileaccidentsdisappear—andthentestregularlytoseeifthisreallyhappens.Finally,theygobackandthinktheproblemthroughagainwhenevertheyseesomethingatypical,whentheyfindunexplainedphenomena,orwhenthecourseofeventsdeviates,evenindetails,fromexpectations. TheseareinessencetherulesHippocrateslaiddownformedicaldiagnosiswellover2,000yearsago.TheyaretherulesforscientificobservationfirstformulatedbyAristotleandthenreaffirmedbyGalileo300yearsago.These,inotherwords,areold,well-known,time-testedrules,whichanexecutivecanlearnandapplysystematically. TheSpecifications Thenextmajorelementinthedecisionprocessisdefiningclearspecificationsastowhatthedecisionhastoaccomplish.Whataretheobjectivesthedecisionhastoreach?Whataretheminimumgoalsithastoattain?Whataretheconditionsithastosatisfy?Insciencetheseareknownas“boundaryconditions.”Adecision,tobeeffective,needstosatisfytheboundaryconditions.Consider: “Canourneedsbesatisfied,”AlfredP.Sloan,Jr.presumablyaskedhimselfwhenhetookcommandofGeneralMotorsin1922,“byremovingtheautonomyofourdivisionheads?”Hisanswerwasclearlyinthenegative.Theboundaryconditionsofhisproblemdemandedstrengthandresponsibilityinthechiefoperatingpositions.Thiswasneededasmuchasunityandcontrolatthecenter.EveryonebeforeSloanhadseentheproblemasoneofpersonalities—tobesolvedthroughastruggleforpowerfromwhichonemanwouldemergevictorious.Theboundaryconditions,Sloanrealized,demandedasolutiontoaconstitutionalproblem—tobesolvedthroughanewstructure:decentralizationwhichbalancedlocalautonomyofoperationswithcentralcontrolofdirectionandpolicy. Adecisionthatdoesnotsatisfytheboundaryconditionsisworsethanonewhichwronglydefinestheproblem.Itisallbutimpossibletosalvagethedecisionthatstartswiththerightpremisesbutstopsshortoftherightconclusions.Furthermore,clearthinkingabouttheboundaryconditionsisneededtoknowwhenadecisionhastobeabandoned.Themostcommoncauseoffailureinadecisionliesnotinitsbeingwronginitially.Rather,itisasubsequentshiftinthegoals—thespecifications—whichmakesthepriorrightdecisionsuddenlyinappropriate.Andunlessthedecisionmakerhaskepttheboundaryconditionsclear,soastomakepossibletheimmediatereplacementoftheoutflankeddecisionwithanewandappropriatepolicy,hemaynotevennoticethatthingshavechanged.Forexample: FranklinD.Rooseveltwasbitterlyattackedforhisswitchfromconservativecandidatein1932toradicalpresidentin1933.Butitwasn’tRooseveltwhochanged.Thesuddeneconomiccollapsewhichoccurredbetweenthesummerof1932andthespringof1933changedthespecifications.Apolicyappropriatetothegoalofnationaleconomicrecovery—whichaconservativeeconomicpolicymighthavebeen—wasnolongerappropriatewhen,withtheBankHoliday,thegoalhadtobecomepoliticalandsocialcohesion.Whentheboundaryconditionschanged,Rooseveltimmediatelysubstitutedapoliticalobjective(reform)forhisformereconomicone(recovery). Aboveall,clearthinkingabouttheboundaryconditionsisneededtoidentifythemostdangerousofallpossibledecisions:theoneinwhichthespecificationsthathavetobesatisfiedareessentiallyincompatible.Inotherwords,thisisthedecisionthatmight—justmight—workifnothingwhatevergoeswrong.AclassiccaseisPresidentKennedy’sBayofPigsdecision: OnespecificationwasclearlyCastro’soverthrow.Theotherwastomakeitappearthattheinvasionwasa“spontaneous”uprisingoftheCubans.Butthesetwospecificationswouldhavebeencompatiblewitheachotheronlyifanimmediateisland-wideuprisingagainstCastrowouldhavecompletelyparalyzedtheCubanarmy.Andwhilethiswasnotimpossible,itclearlywasnotprobableinsuchatightlycontrolledpolicestate. Decisionsofthissortareusuallycalled“gambles.”Butactuallytheyarisefromsomethingmuchlessrationalthanagamble—namely,ahopeagainsthopethattwo(ormore)clearlyincompatiblespecificationscanbefulfilledsimultaneously.Thisishopingforamiracle;andthetroublewithmiraclesisnotthattheyhappensorarely,butthattheyare,alas,singularlyunreliable. Everyonecanmakethewrongdecision.Infact,everyonewillsometimesmakeawrongdecision.Butnoexecutiveneedstomakeadecisionwhich,onthefaceofit,seemstomakesensebut,inreality,fallsshortofsatisfyingtheboundaryconditions. TheDecision Theeffectiveexecutivehastostartoutwithwhatis“right”ratherthanwhatisacceptablepreciselybecauseacompromiseisalwaysnecessaryintheend.Butifwhatwillsatisfytheboundaryconditionsisnotknown,thedecisionmakercannotdistinguishbetweentherightcompromiseandthewrongcompromise—andmayendupbymakingthewrongcompromise.Consider: Iwastaughtthislessonin1944whenIstartedonmyfirstbigconsultingassignment.ItwasastudyofthemanagementstructureandpoliciesofGeneralMotorsCorporation.AlfredP.Sloan,Jr.,whowasthenchairmanandchiefexecutiveofficerofthecompany,calledmetohisofficeatthestartofmyassignmentandsaid:“Ishallnottellyouwhattostudy,whattowrite,orwhatconclusionstocometo.Thisisyourtask.Myonlyinstructiontoyouistoputdownwhatyouthinkisrightasyouseeit.Don’tyouworryaboutourreaction.Don’tyouworryaboutwhetherwewilllikethisordislikethat.Anddon’tyou,aboveall,concernyourselfwiththecompromisesthatmightbeneededtomakeyourconclusionsacceptable.Thereisnotoneexecutiveinthiscompanywhodoesnotknowhowtomakeeverysingleconceivablecompromisewithoutanyhelpfromyou.Buthecan’tmaketherightcompromiseunlessyoufirsttellhimwhatrightis.” Theeffectiveexecutiveknowsthattherearetwodifferentkindsofcompromise.Oneisexpressedintheoldproverb,“Halfaloafisbetterthannobread.”Theother,inthestoryofthejudgmentofSolomon,isclearlybasedontherealizationthat“halfababyisworsethannobabyatall.”Inthefirstinstance,theboundaryconditionsarestillbeingsatisfied.Thepurposeofbreadistoprovidefood,andhalfaloafisstillfood.Halfababy,however,doesnotsatisfytheboundaryconditions.Forhalfababyisnothalfofalivingandgrowingchild. Itisawasteoftimetoworryaboutwhatwillbeacceptableandwhatthedecisionmakershouldorshouldnotsaysoasnottoevokeresistance.(Thethingsoneworriesaboutseldomhappen,whileobjectionsanddifficultiesnoonethoughtaboutmaysuddenlyturnouttobealmostinsurmountableobstacles.)Inotherwords,thedecisionmakergainsnothingbystartingoutwiththequestion,“Whatisacceptable?”Forintheprocessofansweringit,heorsheusuallygivesawaytheimportantthingsandlosesanychancetocomeupwithaneffective—letalonetheright—answer. TheAction Convertingthedecisionintoactionisthefifthmajorelementinthedecisionprocess.Whilethinkingthroughtheboundaryconditionsisthemostdifficultstepindecisionmaking,convertingthedecisionintoeffectiveactionisusuallythemosttime-consumingone.Yetadecisionwillnotbecomeeffectiveunlesstheactioncommitmentshavebeenbuiltintoitfromthestart.Infact,nodecisionhasbeenmadeunlesscarryingitoutinspecificstepshasbecomesomeone’sworkassignmentandresponsibility.Untilthen,itisonlyagoodintention. Theflawinsomanypolicystatements,especiallythoseofbusiness,isthattheycontainnoactioncommitment—tocarrythemoutisnoone’sspecificworkandresponsibility.Smallwonderthenthatthepeopleintheorganizationtendtoviewsuchstatementscynically,ifnotasdeclarationsofwhattopmanagementisreallynotgoingtodo. Convertingadecisionintoactionrequiresansweringseveraldistinctquestions:Whohastoknowofthisdecision?Whatactionhastobetaken?Whoistotakeit?Whatdoestheactionhavetobesothatthepeoplewhohavetodoitcandoit?Thefirstandthelastofthesequestionsaretoooftenoverlooked—withdireresults.Astorythathasbecomealegendamongoperationsresearchersillustratestheimportanceofthequestion,“Whohastoknow?”: Amajormanufacturerofindustrialequipmentdecidedseveralyearsagotodiscontinueoneofitsmodelsthathadforyearsbeenstandardequipmentonalineofmachinetools,manyofwhichwerestillinuse.Itwas,therefore,decidedtosellthemodeltopresentownersoftheoldequipmentforanotherthreeyearsasareplacement,andthentostopmakingandsellingit.Ordersforthisparticularmodelhadbeengoingdownforagoodmanyyears.Buttheyshotupimmediatelyascustomersreorderedagainstthedaywhenthemodelwouldnolongerbeavailable.Noonehad,however,asked,“Whoneedstoknowofthisdecision?” Consequently,nobodyinformedthepurchasingclerkwhowasinchargeofbuyingthepartsfromwhichthemodelitselfwasbeingassembled.Hisinstructionsweretobuypartsinagivenratiotocurrentsales—andtheinstructionsremainedunchanged. Thus,whenthetimecametodiscontinuefurtherproductionofthemodel,thecompanyhadinitswarehouseenoughpartsforanother8to10yearsofproduction,partsthathadtobewrittenoffataconsiderableloss. Theactionmustalsobeappropriatetothecapacitiesofthepeoplewhohavetocarryitout.Thus: AlargeU.S.chemicalcompanyfounditself,inrecentyears,withfairlylargeamountsofblockedcurrencyintwoWestAfricancountries.Toprotectthismoney,topmanagementdecidedtoinvestitlocallyinbusinesseswhichwould:(1)contributetothelocaleconomy,(2)notrequireimportsfromabroad,and(3)ifsuccessful,bethekindthatcouldbesoldtolocalinvestorsifandwhencurrencyremittancesbecamepossibleagain.Toestablishthesebusinesses,thecompanydevelopedasimplechemicalprocesstopreserveatropicalfruit—astaplecropinbothcountries—which,upuntilthen,hadsufferedseriousspoilageintransittoitsWesternmarkets. Thebusinesswasasuccessinbothcountries.ButinonecountrythelocalmanagersetthebusinessupinsuchamannerthatitrequiredhighlyskilledandtechnicallytrainedmanagementofakindnoteasilyavailableinWestAfrica.Intheothercountry,thelocalmanagerthoughtthroughthecapacitiesofthepeoplewhowouldeventuallyhavetorunthebusiness.Consequently,heworkedhardatmakingboththeprocessandthebusinesssimple,andatstaffinghisoperationfromthestartwithlocalnationalsrightuptothetopmanagementlevel. Thisarticlealsoappearsin: ThePeterF.DruckerReader Leadership&ManagingPeople Book PeterF.Drucker 24.95 AddtoCart Save Share Afewyearslateritbecamepossibleagaintotransfercurrencyfromthesetwocountries.But,thoughthebusinessflourished,nobuyercouldbefoundforitinthefirstcountry.Nooneavailablelocallyhadthenecessarymanagerialandtechnicalskillstorunit,andsothebusinesshadtobeliquidatedataloss.Intheothercountry,somanylocalentrepreneurswereeagertobuythebusinessthatthecompanyrepatriateditsoriginalinvestmentwithasubstantialprofit. Thechemicalprocessandthebusinessbuiltonitwereessentiallythesameinbothplaces.Butinthefirstcountrynoonehadasked,“Whatkindofpeopledowehaveavailabletomakethisdecisioneffective?Andwhatcantheydo?”Asaresult,thedecisionitselfbecamefrustrated. Thisactioncommitmentbecomesdoublyimportantwhenpeoplehavetochangetheirbehavior,habits,orattitudesifadecisionistobecomeeffective.Here,theexecutivemustmakesurenotonlythattheresponsibilityfortheactionisclearlyassigned,butthatthepeopleassignedarecapableofcarryingitout.Thusthedecisionmakerhastomakesurethatthemeasurements,thestandardsforaccomplishment,andtheincentivesofthosechargedwiththeactionresponsibilityarechangedsimultaneously.Otherwise,theorganizationpeoplewillgetcaughtinaparalyzinginternalemotionalconflict.Considerthesetwoexamples: WhenTheodoreVailwaspresidentoftheBellTelephoneSystem60yearsago,hedecidedthatitsbusinesswasservice.ThisdecisionexplainsinlargepartwhytheUnitedStates(andCanada)hastodayaninvestor-owned,ratherthananationalized,telephonesystem.YetthispolicystatementmighthaveremainedadeadletterifVailhadnotatthesametimedesignedyardsticksofserviceperformanceandintroducedtheseasameanstomeasure,andultimatelytoreward,managerialperformance.TheBellmanagersofthattimewereusedtobeingmeasuredbytheprofitability(oratleastbythecost)oftheirunits.Thenewyardsticksresultedintherapidacceptanceofthenewobjectives. Insharpcontrastistherecentfailureofabrilliantchairmanandchiefexecutivetomakeeffectiveaneworganizationstructureandnewobjectivesinanold,large,andproudU.S.company.Everyoneagreedthatthechangeswereneeded.Thecompany,aftermanyyearsasleaderofitsindustry,showeddefinitesignsofaging.Inmanymarketsnewer,smaller,andmoreaggressivecompetitorswereoutflankingit.Butcontrarytotheactionrequiredtogainacceptanceforthenewideas,thechairman—inordertoplacatetheopposition—promotedprominentspokesmenoftheoldschoolintothemostvisibleandhighestsalariedpositions—inparticularintothreenewexecutivevicepresidencies.Thismeantonlyonethingtothepeopleinthecompany:“Theydon’treallymeanit.”Ifthegreatestrewardsaregivenforbehaviorcontrarytothatwhichthenewcourseofactionrequires,theneveryonewillconcludethatthisiswhatthepeopleatthetopreallywantandaregoingtoreward. OnlythemosteffectiveexecutivecandowhatVaildid—buildtheexecutionofhisdecisionintothedecisionitself.Buteveryexecutivecanthinkthroughwhatactioncommitmentsaspecificdecisionrequires,whatworkassignmentsfollowfromit,andwhatpeopleareavailabletocarryitout. TheFeedback Finally,informationmonitoringandreportinghavetobebuiltintothedecisiontoprovidecontinuoustesting,againstactualevents,oftheexpectationsthatunderliethedecisions.Decisionsaremadebypeople.Peoplearefallible;atbest,theirworksdonotlastlong.Eventhebestdecisionhasahighprobabilityofbeingwrong.Eventhemosteffectiveoneeventuallybecomesobsolete. Thissurelyneedsnodocumentation.Andeveryexecutivealwaysbuildsorganizedfeedback—reports,figures,studies—intohisorherdecisiontomonitorandreportonit.Yetfartoomanydecisionsfailtoachievetheiranticipatedresults,orindeedevertobecomeeffective,despiteallthesefeedbackreports.JustastheviewfromtheMatterhorncannotbevisualizedbystudyingamapofSwitzerland(oneabstraction),adecisioncannotbefullyandaccuratelyevaluatedbystudyingareport.Thatisbecausereportsare,ofnecessity,abstractions. Effectivedecisionmakersknowthisandfollowarulewhichthemilitarydevelopedlongago.Thecommanderwhomakesadecisiondoesnotdependonreportstoseehowitisbeingcarriedout.Thecommanderoranaidegoesandlooks.Thereasonisnotthateffectivedecisionmakers(oreffectivecommanders)distrusttheirsubordinates.Rather,theylearnedthehardwaytodistrustabstract“communications.” Withthecomingofthecomputerthisfeedbackelementwillbecomeevenmoreimportant,forthedecisionmakerwillinalllikelihoodbeevenfurtherremovedfromthesceneofaction.Unlessheorsheaccepts,asamatterofcourse,thatheorshehadbettergooutandlookatthesceneofaction,heorshewillbeincreasinglydivorcedfromreality.Allacomputercanhandleisabstractions.Andabstractionscanbereliedononlyiftheyareconstantlycheckedagainstconcreteresults.Otherwise,theyarecertaintomislead. Togoandlookisalsothebest,ifnottheonlyway,foranexecutivetotestwhethertheassumptionsonwhichthedecisionhasbeenmadearestillvalidorwhethertheyarebecomingobsoleteandneedtobethoughtthroughagain.Andtheexecutivealwayshastoexpecttheassumptionstobecomeobsoletesoonerorlater.Realityneverstandsstillverylong. Failuretogooutandlookisthetypicalreasonforpersistinginacourseofactionlongafterithasceasedtobeappropriateorevenrational.Thisistrueforbusinessdecisionsaswellasforgovernmentalpolicies.ItexplainsinlargemeasurethefailureofStalin’scoldwarpolicyinEurope,butalsotheinabilityoftheUnitedStatestoadjustitspoliciestotherealitiesofaEuroperestoredtoprosperityandeconomicgrowth,andthefailureoftheBritishtoaccept,untiltoolate,therealityoftheEuropeanCommonMarket.Moreover,inanybusinessIknow,failuretogooutandlookatcustomersandmarkets,atcompetitorsandtheirproducts,isalsoamajorreasonforpoor,ineffectual,andwrongdecisions. Decisionmakersneedorganizedinformationforfeedback.Theyneedreportsandfigures.Butunlesstheybuildtheirfeedbackarounddirectexposuretoreality—unlesstheydisciplinethemselvestogooutandlook—theycondemnthemselvestoasteriledogmatism. ConcludingNote Decisionmakingisonlyoneofthetasksofanexecutive.Itusuallytakesbutasmallfractionofhisorhertime.Buttomaketheimportantdecisionsisthespecificexecutivetask.Onlyanexecutivemakessuchdecisions. Aneffectiveexecutivemakesthesedecisionsasasystematicprocesswithclearlydefinedelementsandinadistinctsequenceofsteps.Indeed,tobeexpected(byvirtueofpositionorknowledge)tomakedecisionsthathavesignificantandpositiveimpactontheentireorganization,itsperformance,anditsresultscharacterizestheeffectiveexecutive. AversionofthisarticleappearedintheJanuary1967issueofHarvardBusinessReview. ReadmoreonDecisionmakingandproblemsolving orrelatedtopics LeadershipandFeedback PD PeterF.Drucker(November19,1909–November11,2005)wasanAustrian-bornAmericanmanagementconsultant,educator,andauthorwhosewritingscontributedtothephilosophicalandpracticalfoundationsofthemodernbusinesscorporation.Hewasalsoaleaderinthedevelopmentofmanagementeducation,heinventedtheconceptknownasmanagementbyobjectives,andhehasbeendescribedas“thefounderofmodernmanagement.” Tweet Post Share Save GetPDF BuyCopies Print ReadmoreonDecisionmakingandproblemsolving orrelatedtopics LeadershipandFeedback PartnerCenter Diversity Latest Magazine Ascend Topics Podcasts Video Store TheBigIdea Data&Visuals CaseSelections