Personas practice and theory

《Personas practice and theory》由会员上传分享，免费在线阅读，更多相关内容在学术论文-天天文库。

Personas:PracticeandTheoryJohnPruittAbstractMicrosoftCorporationPersonasisaninteractiondesigntechniquewithOneMicrosoftWayconsiderablepotentialforsoftwareproductRedmond,WA98052USAdevelopment.Inthreeyearsofuse,ourcolleaguesand+14257034938wehaveextendedAlanCooperstechniquetomakejpruitt@microsoft.comPersonasapowerfulcomplementtootherusabilitymethods.Afterdescribingandillustratingourapproach,JonathanGrudinweoutlinethepsychologicaltheorythatexplainswhyMicrosoftResearchPersonasaremoreengagingthandesignbasedOneMicrosoftWayprimarilyonscenarios.AsCooperandothershaveRedmond,WA98052USAobserved,Personascanengageteammembersvery+14257060784effectively.Theyalsoprovideaconduitforconveyingajgrudin@microsoft.combroadrangeofqualitativeandquantitativedata,andfocusattentiononaspectsofdesignandusethatothermethodsdonot.KeywordsPersonas,UserArchetypes,UserProfiles,UserResearch,DesignMethod,Scenarios,User-CenteredDesign.Industry/categoryComputersoftware,hardware,andtechnologyProjectstatementWehaveusedPersonasonprojectsthatrangefromPermissiontomakedigitalorhardcopiesofallorpartofthisworksmalltolarge.Thispaperwilldiscusstwosuchforpersonalorclassroomuseisgrantedwithoutfeeprovidedthatprojects,onesmallandonelarge.ThesmallerprojectcopiesarenotmadeordistributedforprofitorcommercialadvantageandthatcopiesbearthisnoticeandthefullcitationoninvolvedthefirstversionofanewWebbrowser,MSNthefirstpage.Tocopyotherwise,orrepublish,topostonserversExplorer.Thelarger,ourmostrecentPersonaseffort,ortoredistributetolists,requirespriorspecificpermissionand/orafee.Copyright2003,ACM.©2003ACM1-58113-728-103/00065.00 2wasinsupportoftheMicrosoftWindowsproductProcessdevelopmentteam.ThegoalofeacheffortwastohelpAnIntroductiontoPersonasateamidentifyandunderstanditstargetaudienceasTheuseofPersonas,fictionalpeople,inproductdesignwellasaidindesignanddevelopmentdecisionsforaiswidelyheralded:inAlanCooper’sbookTheInmatesspecificproductrelease.areRunningtheAsylum[8],intutorialsbyKimGoodwinofCooperDesign[11],andinworkshops[23]Projectparticipants[24],newsletters[5][14][20],on-lineresources[10]TheMSNExplorerproductdevelopmentteamconsisted[15]andresearchpapers[4][12][19].ofseveralhundredmembers.TheWindowsproductdevelopmentteamchangedgreatlyovertime,startingTheuseofabstractrepresentationsofusersoriginatedwithseveralhundredmembersandgrowingtoseveralinmarketing[e.g.,18],butCooper’suseofPersonas,thousandatthepeakoftheeffort.Eachteamtheirgoals,andactivityscenariosisfocusedondesign.comprisedprogrammers,qualityassurancetesters,Henotesthatdesignersoftenhaveavagueorprogrammanagers,designers,technicalwriters,contradictorysenseoftheirintendedusersandmayproductplanners,userresearchers,andmarketingbasescenariosonpeoplesimilartothemselves.Hisprofessionals,amongothers.“goal-directeddesign”providesfocusthroughthecreationoffictionalPersonaswhosegoalsformtheTheMSNExplorerPersonateamincludedonefull-timebasisforscenariocreation.Cooper’searlyPersonasusabilityengineerandthepart-timeeffortsofaproductwereroughsketches,butovertimehismethodevolveddesignerandtwoadditionalusabilityengineers.ThetoincludeinterviewsorethnographytocreatemoreWindowsPersonacreationteamconsistedof22people:detailedcharacters[16].severaltechnicalwriters,severalusabilityengineers,fourproductplanners,andtwomarketresearchers.PriortoCooper,otherspromotedtheuseofabstractAftertheWindowsPersonaswerecreated,theensuingrepresentationsofuserstoguidedesign:userprofilesPersonacampaigninvolvedthepart-timeeffortsofandscenariosderivedfromcontextualinquiry[13][25]severalusabilityengineers,ethnographers,graphicanduserclassesfleshedoutinto“userarchetypes”designers,andproductplanners.[17].Theserepresentationswerealsousedasabasisforscenarioconstruction.ProjectdatesanddurationTheMSNExplorerPersonaeffortstartedinJanuaryofCooper’sapproachcanbeeffective,butouruseof2000andlastedabout10months.TheactualcreationPersonasdivergesinseveralways.HeemphasizesanofthePersonastookabout2months.OurWindows“initialinvestigationphase”anddownplaysongoingPersonaeffortstartedaroundMarchof2001andisdatacollectionandusabilityengineering:“Seemslikeongoingatthetimeofwritingthispaper(roughlytwosandpaper….Veryexpensiveandtime-consuming,ityears),thoughtheinitialcreationandvalidationofthewasn’tsolvingthefundamentalproblem.”[8]WindowsPersonastookaboutthreemonths.Statementssuchas“Wealwaysdesignbeforeputting 3upbuildings”andclaimsthatdesignershaveaninnateprofiling,andfictionalcharacterdefinitionscreatedforabilitytomakeintuitiveleapsthatnomethodologycanuseinscenario-baseddesign.Onespecifictechnique,replace[11]understatethevalueofuserinvolvement.underthename“userarchetypes,”startedaround1995withasingleproductteamandfocusedprimarilyPersonasusedalonecanaiddesign,buttheycanbeonproductplanning,marketing,andproductmorepowerfulifusedtocomplement,notreplace,amessaging[17].TheirapproachresembledGeoffreyfullrangeofquantitativeandqualitativemethods.TheyMoore’s“targetingcustomercharacterizations”[18].canamplifytheeffectivenessofothermethods.Overtime,otherproductteamsadoptedthatmethod,andadaptedittobettersuitproductdevelopment.Personasmighthelpadesignerfocus.However,theirAlthoughmuchoftheadoptionandadaptationofgreatestvalueisinprovidingasharedbasisforPersona-likemethodsbyvariousteamshappenedcommunication.Cooperemphasizescommunicatingtheindependently,commonissuesaroseandsimilardesignanditsrationaleamongdesignersandtheirsolutionsweredeveloped.Fromothersaroundtheclients:“It’seasytoexplainandjustifydesigndecisionscompanywhohadbeendirectlyinvolvedwithcreatingwhenthey’rebasedonPersonagoals...”[16].Wehavetheseuserabstractionsorwhowereexpectedtouseextendedthis,usingPersonastocommunicateatheminproductdefinitionanddesign,wefoundthatbroaderrangeofinformationtomorepeople:theearlyPersona-likeeffortssufferedfromfourmajordesigners,developers,testers,writers,managers,problems:marketers,andothers.Informationfrommarketresearch,ethnographicstudies,instrumented1.Thecharacterswerenotbelievable;eithertheywereprototypes,usabilitytests,oranyotherrelevantsourceobviouslydesignedbycommittee(notbasedondata)Thefourproblemslistedontherightarealsonotedinarecentcanbeconveyedrapidlytoallprojectparticipants.ortherelationshiptodatawasnotclear.paperbyBlomquistandArvola2.Thecharacterswerenotcommunicatedwell.Oftenthe[4],describingaPersonaeffortOurexperiencewithPersonasthatwasnotconsideredfullymaincommunicationmethodwasaresume-likeWehaveactivelyusedPersonas,andrefinedoursuccessful.documentblownuptopostersizeandpostedaroundtechniquesforusingthem,foroverthreeyears.Whenthehallways.theMSNExplorereffortbegan,wedidnotsetouttocreatePersonas.Infact,wewereonlyvaguelyfamiliar3.Therewasnorealunderstandingabouthowtousethewiththeconcept.Ourgoalwastohelpadevelopmentcharacters.Inparticular,therewastypicallynothingteamunderstandandfocusonasetoftargetusers.WethatspoketoalldisciplinesorallstagesofthereadCooper’s1999bookandlookedaroundthedevelopmentcycle.industryandourcompanytoseehowotherteamshad4.Theprojectswereoftengrass-rootseffortswithlittleordefinedtheiraudiencesandcommunicatedthatnohigh-levelsupport(suchaspeopleresourcesforinformationtotheirbroaderteam.ManyproductteamscreatingandpromotingPersonas,budgetforpostersorwithinourcompanyhaddonesignificantworkwithothermaterialstomakethePersonasvisible,ormarketsegmentation,userroledefinition,user 4encouragementfromteamleaders:“thoushaltuseofthecycle.Wehadneithertimenorresourcestodothesecharacters”).originalresearch,butwerefortunatethatothershadcompletedseveralfieldstudiesandmarketresearchTheapproachoutlinedherewasdevelopedspecificallypertinenttoourproduct.Finally,thewholeideaoftoaddressthesefourproblems.Ithasbeenrefinedtousingfictionalcharacterstoaiddesignwasnewtomostaddressadditionalissuesencounteredalongtheway:peopleonourdevelopmentteam,sotherewasmuch!Howbesttocreateuserabstractions?resistancetoovercomeandeducationrequired.!Howmuchcanbefictional,andwhatshouldbebasedBythetimewebegantheWindowsPersonaseffort,ourondata?understandingofthemethodhadgrowntremendously!Whatdataismostappropriate?throughourexperiencesandthroughsharing!Howcandifferenttypesofdatabecombined?experienceswithotherPersonapractitioners[23].!Howcanyouvalidateyourcreations?BecauseofthesuccessofpreviousPersonaeffortsandthePersonabuzzaroundtheindustry,themethodhad!Canmultiplerelatedproductteamsshareacommonbecomemorefamiliarandfairlywellacceptedbythesetofabstractions?developmentteam.Weweregivenpeopleresources!Howcanyoudeterminewhethertheeffortwasworthandadecentbudgetforposters,events,andotherit?promotionalexploits.Mostimportant,Personaswere!Didtheproductgetbetterasaresult?andsoon.beingrequestedbyexecsandteamleadersaswellasmembersofthedesignanddevelopmentteam.WhatOurmethodandprocessbynecessitycombinedwehadsetouttodoinourfirsteffortwasmorelikelytechniquesgleanedfromthepreviousPersona-liketobeachievedinthislargereffort.effortswithwhatwecouldlearnfromCooper’sbook,whichwasnotwrittenasa“howto”manual.PracticedetailsCreatingandusingPersonas:OurapproachOurMSNExplorerPersonaseffortsufferedfromseveralThefollowingisabulletedsketchofourcurrentproblems.First,becausethiswasnewtous,webeganprocess.Whereappropriate,wecalloutdifferencesinwithlittleideaofhowmuchworkwasinvolvedandtheresource-lackingExplorerPersonaseffortandthewhatwouldbegained.Thus,obtainingresourcesandresource-intensiveWindowsPersonaseffort.creatingreasonabletimelineswasdifficult.Westartedwithnobudgetandtwopeoplewhohadplentyofother!Weattempttostartaneffortusingpreviouslyworktodo.WebeganthePersonaseffortastheexecuted,large-samplemarketsegmentationstudies,productvisionandinitialplanningwerebeingmuchlikethosediscussedbyWeinstein[27].Highestcompleted.BythetimewefinishedcreatingPersonas,prioritysegmentsarefleshedoutwithuserresearchwhichtookmuchlongerthanexpected,ourteamhadthatincludesfieldstudies,focusgroups,interviews,fullycompletedthebasicdesignandspecificationphaseandfurthermarketresearch.Weusemetricsaround 5marketsize,historicalrevenue,andstrategicor!AswewrotethePersonas’stories,weemployedcompetitiveplacementtodeterminewhichsegmentsqualitativedataandobservationalanecdoteswhereareenrichedintoPersonas.Wetrytokeepthesetofpossible.Anotquiteyetachievedgoalistohaveeverycharactersdowntoamanageablenumber:3to6statementinourPersonasgeneratedfromorrelatedtoPersonas,dependingonthebreadthofproductuse.userdataorobservation.!Generally,wecollectasmuchexistingrelatedmarket!InallPersonaefforts,weuseacentral“foundation”anduserresearchaspossible(frominternalanddocumentforeachPersonaasastorehouseforexternalsources)tohelpinformand“fillout”theinformationaboutthatPersona(data,keyattributes,Personas.WehaveyettostartaPersonaeffortinanphotos,referencematerials,andsoon).Figure1showsareathatdoesnothavesomeexistingquantitativeandthetableofcontentsforatypicalfoundationdocument.qualitativedata.Thus,ourownresearchendeavorsNotethatthefoundationdocumentisnottheprimarytypicallystartafterwecreateourPersonas.meansofcommunicatinginformationaboutthePersona!Althoughwehavenotyetcreatedfull-oninternationaltogeneralteammembers(moreonthisfollowing).ordisabledPersonas,weincludedinternationalmarketLikewise,thefoundationdocumentsdonotcontainallandaccessibilityinformationinourPersonas.Severaloforevenmostofthefeaturescenarios(forexample,ourpartnerteamshavealsocreated“anti-Personas”:“walk-through”scenariosarelocateddirectlyinthePersonasintendedtoidentifypeoplethatarefeaturespecs).Instead,thefoundationdocumentspecificallynotbeingdesignedfor.containsgoals,fears,andtypicalactivitiesthatmotivateandjustifyscenariosthatappearinfeature!InourlargerPersonaeffort,involving22people,wespecs,visiondocuments,storyboards,andsoforth.dividedtheteamsothateachPersona(6inall)had2ormorededicatedpeople.Attheotherextreme,two!LinksbetweenPersonacharacteristicsandthepeoplecreatedallfourMSNExplorerPersonas,thoughsupportingdataaremadeexplicitandsalientintheafewotherpeoplecontributedtoorreviewedvariousfoundationdocuments.Thesedocumentscontainaspectsoftheworkfromtimetotime.Asmentioned,copiousfootnotes,commentsonspecificdata,andlinksthislightereffortreliedsolelyonexistinguserresearch,toresearchreportsthatsupportandexplaintheandasaresult,generatedfarlessdetailedPersonas.Personas’characteristics.AllPersonaillustrationsandmaterialspointtothefoundationdocuments(whichare!TheWindowsPersonaseffortdrewonmanyresearchonanintranetsite)toenableteammemberstoaccessstudies.Wedivvieduptheresearchdocuments,withthesupportingdocumentation.eachteammemberbecomingwellacquaintedwithonlyafewstudies.Wethenheld“affinity”sessionswhere!OnceabasicPersonadescriptioniswritten,wefindwephysicallycutdatapointsandinteresting/relevantlocalpeopletoserveasmodelsandholdone-totwo-factsoutofthestudiesandpinnedthemtoawalltohourphotoshootstogetvisualmaterialtohelpformgroupsofrelatedfindingsacrossstudies.TheillustrateandcommunicateeachPersona.Wehaveresultinggroupsoffindingswereusedinwritingavoidedstockphotogalleriesbecausetheytypicallynarrativesthattoldthestoryofthedata. 6characteristicstoseehowwelltheymatchonlow-levelFigure1:ThetableofcontentsOverview–AlanWaters(BusinessOwner)characteristics.Wedothisbecauseourcreationmethodforafoundationdocument.GettoknowAlan,hisbusiness,andfamily.utilizesmultipledatasources,manyofwhicharenotADayintheLifedirectlycomparableorinherentlycompatible.FollowAlanthroughatypicalday.!OncethePersonas’documentsandmaterialsareinWorkActivitiesplace,weholdakick-offmeetingtointroducetheLookatAlan’sjobdescriptionandroleatwork.Personastotheteamatlarge.HouseholdandLeisureActivitiesGetinformationaboutwhatAlandoeswhenhe’snotatwork.!CommunicatingourPersonashasbeenmultifaceted,Goals,Fears,andAspirationsmultimodal,ongoing,andprogressivelydisclosesmoreUnderstandtheconcernsAlanhasabouthislife,career,andandmoreinformationaboutthem.Ourfoundationbusiness.documentsareavailabletoanyoneontheteamwhoComputerSkills,Knowledge,andAbilitieswishestoreviewthem,buttheyarenottheprimaryLearnaboutAlan’scomputerexperience.meansfordeliveringinformation.Instead,wecreateMarketSizeandInfluencemanyvariationsofposters,flyers,andhandoutsoverUnderstandtheimpactpeoplelikeAlanhaveonourbusiness.thecourseofthedevelopmentcycle.FortheWindowsDemographicAttributesPersonasweevencreatedafewgimmicky(andReadkeydemographicinformationaboutAlanandhisfamily.popular)promotionalitems(e.g.,squeezetoys,beerTechnologyAttributesglasses,andmousepads—sprinkledwithPersonaGetasenseofwhatAlandoeswithtechnology.imagesandinformation).WecreatedWebsitesthatTechnologyAttitudeshostfoundationdocuments,linkstosupportingReviewAlan’sperspectiveontechnology,pastandfuture.research,relatedcustomerdataandscenarios,andaCommunicatinghostoftoolsforusingthePersonas(screeningmaterialLearnhowAlankeepsintouchwithpeople.forrecruitingusabilitytestparticipants,spreadsheetInternationalConsiderationstools,comparisoncharts,postersandphotos,etc.).InFindoutwhatAlanislikeoutsidetheU.S.anongoing“Personafactoftheweek”e-mailQuotescampaign,eachPersonagetsareale-mailaddressHearwhatAlanhastosay.usedoccasionallytosendinformationtotheReferencesdevelopmentteam.Figure2showstwogeneralpostersSeesourcematerialsforthisdocument.designedtofurtherateam’sunderstandingofthePersonas.OnecomparesimportantcharacteristicsoffourPersonas.TheothercommunicatesthefactthatofferonlyoneortwoshotsofagivenmodelandtheourPersonasarebasedonrealpeopleandtriestoimagesaretoo“slick.”provideasenseoftheessenceofaPersonaby!ForourWindowsPersonaseffort,afterourPersonasprovidingquotationsfromrealuserswhoaresimilartowerecreated,wesetup“sanitycheck”sitevisitswiththatPersona.Figure3showstwopostersfromaseriesuserswhomatchthePersonasonhigh-level 7thatprovidesinformationspecificallyabouthowcustomersthinkaboutsecurityandprivacy.ThefirstagainprovidesrealquotesfromuserswhofitourvariousPersonaprofiles.ThesecondpostershowshowarealhackertargetedpeoplewhoresembleoneofourPersonas.!WeinstructourteaminPersonauseandprovidetoolstohelp.CooperdescribesPersonausemostlyasadiscussiontool.“WouldAlanusethisfeature?”Thisisvaluable,butwehavegeneratedadditionalactivitiesandincorporatedthemintospecificdevelopmentprocesses.WecreatedspreadsheettoolsanddocumenttemplatesforclearerandconsistentPersonautilization.AsanexampleofhowPersonascanbecomeexplicitlyFigure2:Twogeneralposters:onecomparingcharacteristicsinvolvedinthedesignanddevelopmentprocess,acrossPersonas;theotherpresentingrealquotesfromusersthatfittheprofileofoneofourPersonas.Figure4showsanabstractversionofafeature-Personaweightedprioritymatrixthatcanhelpprioritizefeaturesforaproductdevelopmentcycle.Intheexample,thescoringinthefeaturerowsisasfollows:-1(thePersonaisconfused,annoyed,orinsomewayharmedbythefeature),0(thePersonadoesn’tcareaboutthefeatureonewayortheother),+1(thefeatureprovidessomevaluetothePersona),+2(thePersonalovesthisfeatureorthefeaturedoessomethingwonderfulforthePersonaeveniftheydon’trealizeit).Thesumsareweightedaccordingtotheproportionofthemarketeachrepresents.Oncecompleted,therowscanbesortedaccordingtotheweightedsumandcriteriacanbecreatedtoestablishwhatfeaturesshouldbepursuedandwhatfeaturesshouldbereconsidered.Shownbelow,features2and4(Note:Figures2and3areshouldbemadeahighpriorityforthedevelopmentFigure:3:Twomoretargetedposters:onecommunicatingintentionallysmallandhidedetailteam;feature3shouldprobablybedropped.aspectsofsecurityandprivacyacrossallofourPersonas;thetopreserveproprietaryothershowinghowcertaintypesofhackerscantargetoneofinformationinthem.)ourPersonas 8Persona1Persona2Persona3Weight:503515WeightedSumFigure4:Afeature!AsacommunicationmechanismusefultothePersonaFeature101265byPersona-teamitself,wecreatePersonascreenersandrecruitweightedpriorityFeature2211150participantsforusabilityandmarketresearch.Wethenmatrix.Feature3-110-15categorize,analyze,andreportourfindingsbyPersonaFeature4111100type.FortheWindowsPersonas,wehavegonetotheextremeofcreatingaPersonauserpanel.ThroughanEtc.----outsidefirm,weestablisheda5,000-personpanelofusersthatmatchourPersonaprofiles.Wepollthe!Wemakeastrongefforttoensurethatallproductandpanelonaregularbasistobetterunderstandreportedfeaturespecificationdocumentscontainwalk-throughactivities,preferences,andopinions,aswellasscenariosthatutilizeourPersonas.Wedothesamereactionstoourfeatureplans,vision,andwithvisiondocuments,storyboards,demos,andsoimplementations.WehavenotagedourPersonasoverforth.UnfortunatelyfortheMSNExplorereffort,wetime,butwedorevisethemasnewdatabecomescompletedourPersonastoolateintheprocesstouseavailable.UnlikeCooper,wesupportastrong,ongoingthisapproach.efforttoobtainasmuchquantitativeandqualitativeinformationaboutusersaspossible,therebyimproving!DuringtheWindowsPersonaseffort,wecollectedtheselection,enrichment,andevolutionofsetsofPersonascenariosfromacrosstheproductteaminaPersonas.spreadsheetthatenablesustotrackandpolicetheuseofthePersonas(thisalsoenablesustoroughlygauge!Oneofourtechnicalwritinggroups,apartnertothethedirectionofaproductasitisdeveloped—forWindowsteam,usedtheWindowsPersonastoplanandexample,howmanyscenariosarewrittenforTobyvs.write“Howto”andreferencebooksforthepopularAbbywhenweknowAbbyisahigher-prioritytarget).press.Indoingso,theyexpandedthePersonastoincludenotionsoflearningstyle,bookusagepatterns,!DesignteamshavemadecreativevisualexplorationsandsoforthtoenrichhowtheyauthoredforspecificbasedonthePersonas.Morespecifically,theycreatedaudiences.brandingandstylecollagesbycuttingandpastingimagesthat“feellike”ourPersonasfromavarietyof!Althoughthishasn’thappenedforthePersonaeffortsmagazinesontoposterboards[Fig.5].Theythenuseddescribedhere,inothereffortsthequalityassurancetheseboardstodoavarietyofvisualtreatmentsacrosstestteamhasusedPersonastoorganizebugbashesseveralareasofourproduct.InanotherPersonaeffort,andselect/refinescenariosfortheirQAtesting.wetookthesetypesofexplorationstofocusgroupsto!FortheWindowsPersonas,weundertookalargeeffortunderstandindetailwhataspectsofthedesignsweretoreconciletwosetsoftargetaudiences(oneintheappealingandhowtheyworkedtogethertoformaformofPersonasandoneintheformofcustomerholisticstyle.AlthoughthePersonaswerenotcriticaltosegments)whenateamworkingonarelatedproductthisprocess,theyservedasspringboardthatinspiredwasdirectedtobe“bettertogether”withourproduct.creation. 9Figure5:APersonafocusedstylecollage.Figure6:AdesignexplorationbasedontheFigure5stylecollage.Resultsbashes—evenusedbyVPsarguingforuserconcernsinBenefitsofPersonasproductstrategymeetings).NotonlyhaveourItiscleartousthatPersonascancreateastrongfocusdevelopmentteamsengagedwithPersonas,butalsoonusersandworkcontextsthroughthefictionalizedcorrespondinglytheyhaveengagedwithourothersettings.Thoughwehavenottriedtoformallymeasureuser-centeredactivities.OurPersonacampaignstheimpactofourPersonas,thesubjectiveviewofthegeneratedamomentumthatincreasedgeneraluserPersonasandthesurroundingeffortbythefocusandawareness.WithourmostrecentPersonadevelopmentteamhasbeenfavorable.Awiderangeofeffort,we’vehadpartnerteamsbuildingrelatedbutteammembers(fromexecutivestodesignersanddifferentproductsadoptandadaptourPersonasinandevelopers)knowsaboutanddiscussourproductinefforttoenhancecross-teamcollaboration,synergy,termsofthePersonas.We’veseenPersonasgofromandcommunication.scattereduse(inearlyPersonaprojects)towidespreadadoptionandunderstanding(inrecentproductcycles).TheactofcreatingPersonashashelpedusmakeourOurPersonasareseeneverywhereandusedbroadlyassumptionsaboutthetargetaudiencemoreexplicit.(forexample,featurespecs,visiondocuments,Oncecreated,thePersonashavehelpedmakestoryboards,demo-ware,designdiscussions,bugassumptionsanddecision-makingcriteriaequally 10explicit.Whyarewebuildingthisfeature?WhyarewetestaproductfocusingonAlanscenarios,anotherdaybuildingitlikethis?WithoutPersonas,developmentfocusingonLaurascenarios.Asstatedinthepreviousteamsroutinelymakedecisionsaboutfeaturesandsection,thisworksfortestersandotherproductteamimplementationwithoutrecognizingorcommunicatingmembers,in“bugbashes,”forexample.Antheirunderlyingassumptionsaboutwhowillusetheexperiencedtesterreportedfeelingthathewasproductandhowitwillbeused.The“featurebyidentifying“therightkind”ofproblemsindrawingonPersona-weightedprioritymatrix”describedintheknowledgeofaPersonainguidinghistestscriptsandprevioussectionisagoodexampleofthis.Usingthatactivities.Comparethistoanobservationfromastudytoolinevitablyresultsinfavoredfeaturesorseeminglyofinterfacedevelopment:importantfeaturesbeingpusheddowninthelist.Whenthishappens,teamsmustbeveryexplicitwiththeirSomepeoplerealizedthattestsconductedbyQualityreasoningtogetafeaturebackintheplan.WestressControltoensurethattheproductmatchesspecificationtotheteamthatthistoolisnot“golden,”itisaguide;werenotsufficient.Onemanagernoted,“Iwouldsaythatexceptionscanandshouldbemade,whenappropriate.testingshouldbedonebyagroupoutsideDevelopment,‘causeDevelopmentknowshowthecodeworks,andevenPersonasareamediumforcommunication;aconduitthoughyoudon’twantitto,yoursubconsciousmakesyouforinformationaboutusersandworksettingsderivedtestthewayyouknowitworks.See,thosepeopleinthefromethnographies,marketresearch,usabilitystudies,QualityControlgrouphavenothingtodowithcustomers.interviews,observations,andsoon.OnceasetofThey’renotusers.”Personasisfamiliartoateam,anewfindingcanbeInfact,twomembersofFieldSupportwerereportedtoinstantlycommunicated:“AlancannotusethesearchhavefoundmorebugsthantheQualityControlgroupintoolonyourWebpage”hasanimmediacythat“athelatestrelease,andtheyhadaccomplishedthisbysubsetofparticipantsintheusabilitystudyhadworkingwiththeproductastheyimaginedthatusersproblemswiththesearchtool”doesn’t,especiallyforwould.TestingbyFieldSupportwasaninnovativeteammemberswhonow,forallintentsandpurposes,experiment,however,andnotpartoftheacceptedseeAlan(aPersona)asarealperson.Wehavefounddevelopmentprocess.thistobeextremelypowerfulforcommunicatingresults“TheQualityControlgrouphasalotofsystematicandfurtheringourteammates’understandingofthetesting,andyouneedsomeofthat,butatthesametime,Personas.youneedsomebodywhoisessentiallyacustomer.ItisasifyouhadacustomerinhousewhousesitthewayaFinally,Personasfocusattentiononaspecifictargetcustomerwouldeveryday,andisparticularlytoughonitaudience.Themethodhelpsestablishwhoisandandshakesallthesethingsout.That’swhatthesetwoconsequentlywhoisnotbeingdesignedfor.Personasguysdid,anditwasjustinvaluable.”[21,p.64]explicitlydonotcovereveryconceivableuser.Theyalsohelpfocussequentiallyondifferentkindsofusers.Forexample,aqualityassuranceengineercanoneday 11TheFieldSupportengineerscould“testasauser”users.We’vehadsomesuccessincollaboratinghere,becauseoftheirextensiveexperiencewithcustomers.butthereareroughedges.ThatPersonauseresultsinsimilarpositivereportsisencouraging.Finally,wehaveseenacertainlevelof“Personamania”withinourorganizationandothers.PersonascanbeRisksofPersonasoverused.Atworst,theycouldreplaceotheruser-GettingtherightPersonaorsetofPersonasisacenteredmethods,ongoingdatacollection,orproductchallenge.Cooperarguesthatdesigningforanyoneevaluation.Personasarenotapanacea.Theyshouldexternalpersonisbetterthantryingtodesignvaguelyaugmentandenhance:augmentexistingdesignforeveryoneorspecificallyforoneself.Thismaybeprocessesandenhanceuserfocus.We’vefoundthattrue,anditdoesfeelasthoughsettlingonasmallsetPersonasenhanceusertestingandotherevaluationofPersonasprovidessomeinsurance,butitalsoseemsmethods,fieldresearch,scenariogeneration,designclearthatPersonasshouldbedevelopedforaparticularexploration,andsolutionbrainstorming.effort.Inmakingchoicesitbecomesclearthatthechoiceshaveconsequences.Forexample,theywillDiscussionguideparticipantselectionforfuturestudiesandcouldTheoryofmind:howpersonasworkbeusedtofilteroutdatafromsourcesnotmatchingAtfirstencounter,Personasmayseemtoo“arty”foraoneofthePersonaprofiles.scienceandengineering-basedenterprise.Itmayseemmorelogicaltofocusdirectlyonscenarios,whichafterRelatedtothisisthetemptationofPersonareuse.AfteralldescribetheactualworkprocessesoneaimstotheinvestmentindevelopingPersonasandacquaintingsupport.Cooperofferednoexplanationforwhyitispeoplewiththem,itmaybedifficulttoavoidover-bettertodevelopPersonasbeforescenarios.extendingtheirusewhenitwouldbebettertodisbandonecastofcharactersandrecruitanotherone.ItcanFor25years,psychologistshavebeenexploringourbegoodorbadwhenourpartnerteamsadoptoradaptabilitytopredictanotherperson’sbehaviorbyourPersonas.Differentteamsandproductshaveunderstandingtheirmentalstate.TheoryofMindfirstdifferentgoals,sothePersonasarestretchedabit.Soaskedwhetherprimatessharethisability[22]andthenfar,thestretchinghasbeenmodestandcloselytiedtoexploreditsdevelopmentinchildren[1].Everydayofdata(becauseourtargetcustomersdoindeedoverlap),ourlives,startingveryyoung,weusepartialbutitisaconcern.knowledgetodrawinferences,makepredictions,andformexpectationsaboutthepeoplearoundus.WeareInaddition,marketingandproductdevelopmenthavenotalwaysright,butwelearnfromexperience.differentneedsthatrequiredifferentPersonaWheneverwesayordosomething,weanticipatetheattributes,andsometimesdifferenttargetaudiences.reactionsofotherpeople.MisjudgmentsstandoutinMarketingisgenerallyinterestedinbuyerbehaviorandmemory,butweusuallygetitright.customers;productdevelopmentisinterestedinend 12PersonasinvokethispowerfulhumancapabilityandMethodactingusesagreatdealofdetailtoenablebringittothedesignprocess.Well-craftedPersonaspeopletogeneraterealisticbehavior.Detailedhistoriesaregenerative:Oncefullyengagedwiththem,youcanarecreatedforpeopleandevenobjects,detailthatisalmosteffortlesslyprojectthemintonewsituations.Innotexplicitlyreferredtobutwhichisdrawnoncontrast,ascenariocoversjustwhatitcovers.implicitlybytheactor.Ifteammembersaretold,“MarketresearchshowsthatAfictionbasedonresearchcanbeusedto20%ofourtargetusershaveboughtcellphones,”itcommunicate.Forexample,watchingacharactermaynothelpthemmuch.Iftold“AlanhasboughtasuccumbslowlytoadementiaonER,onecancellphone”andAlanisafamiliarPersona,theycanunderstandthediseaseandperhapsevendesignimmediatelybeginextrapolatinghowthiscouldaffecttechnologytosupportsufferers,iftheportrayalisbehavior.Theycancreatescenarios.Wedothiskindofbasedonrealobservationanddata.extrapolationallthetime,weareskilledatit—notperfect,butveryskilled.MergingpersonaswithotherapproachesAsnotedabove,weseePersonascomplementingotherTHEPOWEROFFICTIONTOENGAGEapproaches,orusedwhereanotherapproachisPeopleroutinelyengagewithfictionalcharactersinimpractical.novels,movies,andtelevisionprograms,oftenfiercely.TheyshoutadvicetofictionalcharactersandargueSCENARIOSANDTASKANALYSISoverwhattheyhavedoneoff-screenorafterthenovelScenariosareanaturalelementofPersona-basedends.Particularlyinongoingtelevisiondramasordesignanddevelopment.InCarroll’swords[7],asituationcomedies,characterscometoresemblescenarioisastorywithasetting,agents,oractorswhonormalpeopletosomeextent.Perhapsbetterlookinghavegoalsorobjectives,andaplotorsequenceoforwittieronaverage,butmoderatelycomplex—actionsandevents.Giventhatscenarioshave“actors”stereotypeswouldbecomeboringovertime.andPersonascomewithscenarios,thedistinctionisinwhichcomesfirst,whichtakesprecedence.ActorsorMETHODACTINGANDFOCUSINGONDETAILagentsinscenario-baseddesignaretypicallynotManyactorspreparebyobservingandtalkingwithdefinedfullyenoughtopromotegenerativepeoplewhoresemblethefictionalcharactertheywillengagement.ConsiderCarroll’sexample:portray.AswithPersonas,thefictionalcharacterisbasedonrealdata.Anactorintuitsdetailsofthe“Anaccountantwishestoopenafolderonasystemcharacter’sbehaviorinnewsituations.Adesigner,desktopinordertoaccessamemoonbudgets.However,developer,ortesterissupportedindoingthesameforthefolderiscoveredupbyabudgetspreadsheetthatthethepeopleonwhomaPersonaisbased.accountantwishestorefertowhilereadingthememo.Thespreadsheetissolargethatitnearlyfillsthedisplay.Theaccountantpausesforseveralseconds,resizesthe 13spreadsheet,movesitpartiallyoutofthedisplay,opensnovel,predictablestereotypesbecomeboring,somorethefolder,opensthememo,resizesandrepositionsthecomplex,realisticcharactersaremoreeffective.memoandcontinuesworking.”CONTEXTUALDESIGNANDETHNOGRAPHYThelifelessnessofcharactersinsuchscenarioshasContextualDesign[3],apowerfulapproachtobeencritiquedfromawriter’sperspective[19]andbyobtainingandanalyzingbehavioraldata,isastrongscenario-baseddesignresearcherswhosuggestusingcandidateforinformingPersonas.Asitevolvedovercaricatures,perhapsshockingorextremeones[6][9].twodecades,ContextualDesignincreasinglystressedBødkerwritesin[6],“Itgivesabettereffecttocreatecommunicationwithteammembers,waystosharescenariosthatarecaricatures…itismucheasier…toknowledgeacquiredinthefield.Personasareprimarilyrelateto.…Notthatthey‘believe’inthecaricatures,atooltoachievethisandthusanaturalpartnertoindeedtheydonot,butitismucheasiertouseone’sContextualDesign[4][14].commonsensejudgmentwhenconfrontedwithanumberofextremesthanwhenjudgingbasedonsomeEthnographicdatamayhelpthemostindevelopingkindof‘middleground.’”ShealsorecommendsrealisticPersonas,whenavailableinsufficientdepth.constructingbothutopianandnightmarishscenariosQuantitativedatamaybenecessaryinselectingaroundaproposeddesigntostimulatereflection.appropriatePersonas,butdoesnotreplaceobservation.Again,theparalleltomethodactingarises.TaskanalysisisgenerallydirectedtowardformalrepresentationsthatareparticularlydifficulttoengageWhynotjustuserealpeople?Designingforarealwithgeneratively.Anexceptionis[2],whichpersonisbetterthandesigningblind,butjustaboutrecommendsdetailedcharactersketches.everyonehassomebehaviorsonewouldnotwanttofocusdesignon.UsingarealindividualwouldexcludeThesethoughtfulanalysespointtoweaknessesinorcomplicatetheuseofdatafrommarketresearch,scenariostakenalone.Unlessbasedstronglyondata,ausabilitytesting,andsoon.Itcouldunderminethescenariocanbecreatedtopromoteanyfeature,anyconfidenceofteammembersinthegeneralityofposition(utopianordystopian),andcanbedifficulttoparticularbehaviors—teammembersdostepbackandengagewith.recognizethataPersonarepresentsagroupofpeople,aswhentheydescribe“testingsixAlans.”Personasneednotbeextremeorstereotypedcharacters;theteamengageswiththemoveralongPARTICIPATORYDESIGNANDVALUE-SENSITIVEDESIGNenoughtimetoabsorbnuances,aswedowithrealParticipatoryorcooperativedesignfocusesonthepeople.Thisdurationofengagementiscritical.Inaeventualusersofasystemorapplication.Ithasthemovie,heroesandvillainsmaybestereotypedbecausesamegoalofengagingdeveloperswithuserbehaviorofaneedtodescribethemquickly,aswithstand-aloneandalsoenlistsourabilitytoanticipatebehaviorsofscenarios.Butinanongoingtelevisionseriesorafamiliarpeople.Whendesigningforarelativelysmall, 14accessiblegroupofpeople,thisapproachmakestheAcknowledgmentsmostsense.ProductdevelopmentismorechallengingWethankGaynaWilliams,ShariSchneider,Markforparticipatorydesign.WediscusstherelationshipofPatterson,ChrisNodder,HollyJamesen,TamaraAdlin,Personasandparticipatorydesignindepthin[12].LarryParsons,StevePoltrock,JeanetteBlomberg,andmembersoftheMicrosoftPersonasandQualgroups.Earlyparticipatorydesigneffortsincludedastrongfocusonsociopoliticaland“qualityoflife”issues.TheseReferencesissuesaremoresignificanttodayasthereachof[1]Astington,J.W.,&Jenkins,J.M.(1995).Theoryofmindcomputingextends[26].Althoughtheindustryanddevelopmentandsocialunderstanding.CognitionandEmotion,9,151-65.manycompanieshaveengagedtheseissuesatahighlevel,mostusabilityandinteractiondesigntechniques[2]Benyon,D.&Macauley,C.(2002).ScenariosandtheHCI-SEdesignproblem.Interactingwithcomputers,avoidaddressingtheseissues.Personausebrings14,4,397-405.sociopoliticalissuestothesurface.EachPersonahasa[3]Beyer,H.&Holtzblatt,K.(1998).Contextualdesign.gender,age,race,ethnic,familyorcohabitationMorganKaufmann.arrangement,socio-economicbackground,work,orhomeenvironment.Thisprovidesaneffectiveavenue[4]Blomquist,Å.&Arvola,M.(2002).Personasinaction:Ethnographyinaninteractiondesignteam.ToappearforrecognizingandperhapschangingassumptionsinProc.NordiCHI2002.aboutusers.IfonepopulatedaPersonasetwith[5]Brechin,E.(2002).Reconcilingmarketsegmentsandmiddle-agedwhitemales,itwouldbeobviousthatthisPersonas.isamistake.http://www.cooper.com/newsletters/2002_02/reconciling_market_segments_and_Personas.htmCooperwrites“allthingsbeingequal,Iwillusepeople[6]Bødker,S.(2000).Scenariosinuser-centreddesign—ofdifferentraces,genders,nationalities,andcolors.”Settingthestageforreflectionandaction.InteractingRealism,not“politicalcorrectness,”ishisstatedgoal.withcomputers,13,1,61-75.Hestereotypesifhefeelsitwillprovidemorecredence[7]Carroll,J.(2000).Makinguse:Scenario-baseddesignandavoidscastingstronglyagainstexpectationsifheofhuman-computerinteractions.MITPress.feelsitwillunderminecredibility.[8]Cooper,A.(1999).Theinmatesarerunningtheasylum.Macmillan.Personausedoesrequiredecision-making.Itisn’ta[9]Djajadiningrat,J.P.,Gaver,W.W.&Frens,J.W.(2000).science.Ifnotusedappropriately,anypowerfultoolInteractionrelabellingandextremecharacters:cantakeonedownthewrongpath,asinlyingwithMethodsforexploringaestheticinteractions.Proc.DISstatisticsorusingnon-representativevideoexamples.2000,66-71.Personasareonesuchpowerfultool.Itisuptoallofus[10]Freydenson,E.(2002).BringingyourPersonastolifeintogethertodevelopeffectivewaystousethem.reallife.http://boxesandarrows.com/archives/002343.php 15[11]Goodwin,K.(2002).Goal-directedmethodsforgreat[22]Poltrock,S.E.andGrudin,J.(1994).Organizationaldesign.CHI2002tutorial.obstaclestointerfacedesignanddevelopment:Twohttp://www.acm.org/sigchi/chi2002/tut-sun.html#9participantobserverstudies.ACMTransactionsonComputer-HumanInteraction,1,1,52-80.[12]Grudin,J.&Pruitt,J.(2002).Personas,participatorydesign,andproductdevelopment:Aninfrastructurefor[23]Premack,D.&Woodruff,G.(1978).Doestheengagement.Proc.PDC2002,144-161.chimpanzeehaveatheoryofmind?Behavioral&BrainSciences,4,515-526.[13]Hackos,J.&Redish,J.(1998).Userandtaskanalysisforinterfacedesign.JohnWileyandSons,NewYork.[24]Pruitt,J.S.,Jamesen,H.&Adlin,T.(2001).Personas,userarchetypes,andotheruserrepresentationsin[14]Holtzblatt,K.(2002).Personasandcontextualdesign.softwaredesign.UPA2001workshop.http://www.incent.com/community/design_corner/02_http://www.upassoc.org/conf2001/reg/program/worksh0913.htmlops/w2.html[15]Hourihan,M.(2002).Takingthe“you”outofuser:My[25]Pruitt,J.S.,Jamesen,H.&Adlin,T.(2002).CreatingexperienceusingPersonas.Boxesandarrows.andusingpersonas:Apractitioner’sworkshop.UPAhttp://boxesandarrows.com/archives/002330.php2002workshop.[16]InterviewwithKimGoodwin(2002).UserInterface7http://www.usabilityprofessionals.org/conferences/200East.2/program/workshops/wkshop_Personas.php[17]http://www.uiconf.com/uie-7/goodwin_interview.htm[26]Tahir,M.F.(1997).Who'sontheothersideofyour[18]Mikkelson,N.&Lee,W.O.(2000).Incorporatingusersoftware:Creatinguserprofilesthroughcontextualarchetypesintoscenario-baseddesign.Proc.UPA2000.inquiry.Proc.UPA’97.[19]Moore,G.A.(1991).Crossingthechasm.Harper[27]Value-sensitivedesignsite:http://www.vsdesign.org/CollinsPublishers,NewYork.[28]Weinstein,Art,(1998).Definingyourmarket:winning[20]Nielsen,L.(2002).Fromusertocharacter—anstrategiesforhigh-tech,industrial,andservicefirms.investigationintouser-descriptionsinscenarios.Proc.NewYork:HaworthPress.DIS2002.[21]Perfetti,C.(2002).Personas:Matchingadesigntotheusers'goals.UserInterface7East.http://www.uiconf.com/uie-7/goodwin_article.htm