<span class='text_page_counter'>(1)</span><div class='page_container' data-page=1></div>
<span class='text_page_counter'>(2)</span><div class='page_container' data-page=2></div>
<span class='text_page_counter'>(3)</span><div class='page_container' data-page=3></div>
<span class='text_page_counter'>(4)</span><div class='page_container' data-page=4></div>
<span class='text_page_counter'>(5)</span><div class='page_container' data-page=5></div>
<span class='text_page_counter'>(6)</span><div class='page_container' data-page=6>

Chapter14:OrganizationalRamifications
ChiefDataMonetizationOfficer

Privacy,Trust,andDecisionGovernance
UnleashingOrganizationalCreativity
Summary

HomeworkAssignment
Notes

Chapter15:Stories

CustomerandEmployeeAnalytics
ProductandDeviceAnalytics

NetworkandOperationalAnalytics
CharacteristicsofaGoodBusinessStory
Summary

HomeworkAssignment
Notes

</div>
<span class='text_page_counter'>(7)</span><div class='page_container' data-page=7>

<b>ListofIllustrations</b>

Chapter1:TheBigDataBusinessMandate

Figure1.1BigDataBusinessModelMaturityIndex
Figure1.2Moderndata/analyticsenvironment
Chapter2:BigDataBusinessModelMaturityIndex

Figure2.1BigDataBusinessModelMaturityIndex

Figure2.2Crossingtheanalyticschasm

Figure2.3Packagingandsellingaudienceinsights
Figure2.4Optimizeinternalprocesses

Figure2.5Createnewmonetizationopportunities
Chapter3:TheBigDataStrategyDocument

Figure3.1Bigdatastrategydecompositionprocess
Figure3.2Bigdatastrategydocument

Figure3.3Chipotle's2012lettertotheshareholders

Figure3.4Chipotle's“increasesamestoresales”businessinitiative
Figure3.5Chipotlekeybusinessentitiesanddecisions

Figure3.6CompletedChipotlebigdatastrategydocument
Figure3.7BusinessvalueofpotentialChipotledatasources

Figure3.8ImplementationfeasibilityofpotentialChipotledatasources
Figure3.9Chipotleprioritizationofusecases

Figure3.10SanFranciscoGiantsbigdatastrategydocument
Figure3.11Chipotle'ssamestoresalesresults

Chapter4:TheImportanceoftheUserExperience
Figure4.1Originalsubscribere-mail

Figure4.2Improvedsubscribere-mail
Figure4.3Actionablesubscribere-mail

Figure4.4Apprecommendations

Figure4.5TraditionalBusinessIntelligencedashboard
Figure4.6Actionablestoremanagerdashboard

</div>
<span class='text_page_counter'>(8)</span><div class='page_container' data-page=8>

Figure4.9Localeventsusecase
Figure4.10Localweatherusecase
Figure4.11Financialadvisordashboard
Figure4.12Clientpersonalinformation
Figure4.13Clientfinancialinformation
Figure4.14Clientfinancialgoals

Figure4.15Financialcontributionsrecommendations
Figure4.16Spendanalysisandrecommendations
Figure4.17Assetallocationrecommendations
Figure4.18Otherinvestmentrecommendations

Chapter5:DifferencesBetweenBusinessIntelligenceandDataScience
Figure5.1SchmarzoTDWIkeynote,August2008

Figure5.2OaklandA'sversusNewYorkYankeescostperwin
Figure5.3BusinessIntelligenceversusdatascience

Figure5.4CRISP:CrossIndustryStandardProcessforDataMining
Figure5.5BusinessIntelligenceengagementprocess

Figure5.6TypicalBItoolgraphicoptions
Figure5.7Datascientistengagementprocess
Figure5.8Measuringgoodnessoffit

Figure5.9Dimensionalmodel(starschema)

Figure5.10UsingflatfilestoeliminateorreducejoinsonHadoop
Figure5.11Samplecustomeranalyticprofile

Figure5.12Improvecustomerretentionexample
Chapter6:DataScience101

Figure6.1Basictrendanalysis

Figure6.2Compoundtrendanalysis
Figure6.3Trendlineanalysis

Figure6.4Boxplotanalysis

Figure6.5Geographical(spatial)trendanalysis
Figure6.6Pairsplotanalysis

</div>
<span class='text_page_counter'>(9)</span><div class='page_container' data-page=9>

Figure6.8Clusteranalysis

Figure6.9Normalcurveequivalentanalysis

Figure6.10Normalcurveequivalentsellerpricinganalysisexample
Figure6.11Associationanalysis

Figure6.12Convertingassociationrulesintosegments
Figure6.13Graphanalysis

Figure6.14Textmininganalysis
Figure6.15Sentimentanalysis

Figure6.16Traversepatternanalysis

Figure6.17Decisiontreeclassifieranalysis
Figure6.18Cohortsanalysis

Chapter7:TheDataLake

Figure7.1Characteristicsofadatalake
Figure7.2Theanalyticsdilemma

Figure7.3Thedatalakelineofdemarcation
Figure7.4CreateaHadoop-baseddatalake
Figure7.5Createananalyticsandbox

Figure7.6MoveETLtothedatalake

Figure7.7HubandSpokeanalyticsarchitecture
Figure7.8Datascienceengagementprocess
Figure7.9Whatdoesthefuturehold?

Figure7.10EMCFederationBusinessDataLake
Chapter8:ThinkingLikeaDataScientist

Figure8.1FootLocker'skeybusinessinitiatives

Figure8.2ExamplesofFootLocker'sin-storemerchandising
Figure8.3FootLocker'sstoremanagerpersona

Figure8.4FootLocker'sstrategicnounsorkeybusinessentities

Figure8.5Thinkinglikeadatascientistdecompositionprocess
Figure8.6Recommendationsworksheettemplate

Figure8.7FootLocker'srecommendationsworksheet

</div>
<span class='text_page_counter'>(10)</span><div class='page_container' data-page=10>

Figure8.9Thinkinglikeadatascientistdecompositionprocess
Chapter9:“By”AnalysisTechnique

Figure9.1Identifyingmetricsthatmaybebetterpredictorsofperformance
Figure9.2NBAshootingeffectiveness

Figure9.3LeBronJames'sshootingeffectiveness
Chapter10:ScoreDevelopmentTechnique

Figure10.1FICOscoreconsiderations
Figure10.2FICOscoredecisionrange
Figure10.3Recommendationsworksheet

Figure10.4Updatedrecommendationsworksheet
Figure10.5Completedrecommendationsworksheet
Figure10.6PotentialFootLockercustomerscores
Figure10.7FootLockerrecommendationsworksheet
Figure10.8CLTVbasedonsales

Figure10.9MorepredictiveCLTVscore
Chapter11:MonetizationExercise

Figure11.1“Adayinthelife”customerpersona
Figure11.2Fitnesstrackerprioritization

Figure11.3Monetizationroadmap
Chapter12:MetamorphosisExercise

Figure12.1BigDataBusinessModelMaturityIndex
Figure12.2Patientactionableanalyticprofile

Chapter13:PowerofEnvisioning

Figure13.1BigDataVisionWorkshopprocessandtimeline
Figure13.2BigDataVisionWorkshopillustrativeanalytics
Figure13.3BigDataVisionWorkshopuserexperiencemock-up
Figure13.4PrioritizeHealthcareSystems'susecases

Figure13.5Prioritizationmatrixtemplate
Figure13.6Prioritizationmatrixprocess
Chapter14:OrganizationalRamifications

</div>
<span class='text_page_counter'>(11)</span><div class='page_container' data-page=11></div>
<span class='text_page_counter'>(12)</span><div class='page_container' data-page=12>

<b>ListofTables</b>

Chapter1:TheBigDataBusinessMandate

Table1.1ExploitingTechnologyInnovationtoCreateEconomic-Driven
BusinessOpportunities

Table1.2EvolutionoftheBusinessQuestions
Chapter2:BigDataBusinessModelMaturityIndex

Table2.1BigDataBusinessModelMaturityIndexSummary
Chapter3:TheBigDataStrategyDocument

Table3.1MappingChipotleUseCasestoAnalyticModels

Chapter5:DifferencesBetweenBusinessIntelligenceandDataScience
Table5.1BIAnalystVersusDataScientistCharacteristics

Chapter6:DataScience101

Table6.12014–2015TopNBARPMRankings
Table6.2CaseStudySummary

Chapter7:TheDataLake

Table7.1DataLakeDataTypes

Chapter8:ThinkingLikeaDataScientist

Table8.1EvolutionofFootLocker'sBusinessQuestions
Chapter9:“By”AnalysisTechnique

Table9.1LeBronJames'sShootingPercentages
Chapter10:ScoreDevelopmentTechnique

Table10.1PotentialScoresforOtherIndustries
Chapter11:MonetizationExercise

Table11.1PotentialFitnessTrackerRecommendations
Table11.2RecommendationDataRequirements

Table11.3RecommendationsValueVersusFeasibilityAssessment
Chapter12:MetamorphosisExercise

</div>
<span class='text_page_counter'>(13)</span><div class='page_container' data-page=13></div>
<span class='text_page_counter'>(14)</span><div class='page_container' data-page=14></div>
<span class='text_page_counter'>(15)</span><div class='page_container' data-page=15>

<b>OverviewoftheBookandTechnology</b>

Thedayswhenbusinessstakeholderscouldrelinquishcontrolofdataand
analyticstoITareover.Thebusinessstakeholdersmustbefrontandcenterin
championingandmonetizingtheorganization'sdatacollectionandanalysis
efforts.Businessleadersneedtounderstandwhereandhowtoleveragebigdata,
exploitingthecollisionofnewsourcesofcustomer,product,andoperationaldata
coupledwithdatasciencetooptimizekeybusinessprocesses,uncovernew

monetizationopportunities,andcreatenewsourcesofcompetitivedifferentiation.
Andwhileit'snotrealistictoconvertyourbusinessusersintodatascientists,it's
<i>criticalthatweteachthebusinessuserstothinklikedatascientistssotheycan</i>
collaboratewithITandthedatascientistsonusecaseidentification,requirements
definition,businessvaluation,andultimatelyanalyticsoperationalization.

Thisbookprovidesabusiness-hardenedframeworkwithsupportingmethodology
andhands-onexercisesthatnotonlywillhelpbusinessuserstoidentifywhere
andhowtoleveragebigdataforbusinessadvantagebutwillalsoprovide

</div>
<span class='text_page_counter'>(16)</span><div class='page_container' data-page=16>

<b>HowThisBookIsOrganized</b>

Thebookisorganizedintofoursections:

<b>PartI:BusinessPotentialofBigData.</b>PartIincludesChapters1through
4andsetsthebusiness-centricfoundationforthebook.HereiswhereI

introducetheBigDataBusinessModelMaturityIndexandframethebigdata
discussionaroundtheperspectivethat“organizationsdonotneedabigdata
strategyasmuchastheyneedabusinessstrategythatincorporatesbigdata.”

<b>PartII:DataScience.</b>PartIIincludesChapters5through7andcoversthe
principlebehinddatascience.Thesechaptersintroducesomedatascience
basicsandexplorethecomplementarynatureofBusinessIntelligenceanddata
scienceandhowthesetwodisciplinesarebothcomplementaryanddifferentin
theproblemsthattheyaddress.

<b>PartIII:DataScienceforBusinessStakeholders.</b>PartIIIincludes
Chapters8through12andseekstoteachthebusinessusersandbusiness
leadersto“thinklikeadatascientist.”Thispartintroducesamethodologyand
severalexercisestoreinforcethedatasciencethinkingandapproach.Ithasa
lotofhands-onwork.

</div>
<span class='text_page_counter'>(17)</span><div class='page_container' data-page=17></div>
<span class='text_page_counter'>(18)</span><div class='page_container' data-page=18></div>
<span class='text_page_counter'>(19)</span><div class='page_container' data-page=19>

<b>WhoShouldReadThisBook</b>

Thisbookistargetedtowardbusinessusersandbusinessmanagement.Iwrote
thisbooksothatIcoulduseitinteachingmyBigDataMBAclass,soincludedall
ofthehands-onexercisesandtemplatesthatmystudentswouldneedto

successfullyearntheirBigDataMBAgraduationcertificate.

<i>Ithinkfolkswouldbenefitbyalsoreadingmyfirstbook,BigData:</i>

</div>
<span class='text_page_counter'>(20)</span><div class='page_container' data-page=20>

<b>ToolsYouWillNeed</b>

</div>
<span class='text_page_counter'>(21)</span><div class='page_container' data-page=21>

<b>What'sontheWebsite</b>

Youcandownloadthe“ThinkingLikeaDataScientist”workbookfromthebook's
websiteatwww.wiley.com/go/bigdatamba.Andoh,theremightbeanothersurprise

</div>
<span class='text_page_counter'>(22)</span><div class='page_container' data-page=22>

<b>WhatThisMeansforYou</b>

AsstudentsfrommyclassatUSFhavetoldme,thismaterialallowsthemtotakea

problemorchallengeanduseawell-thought-outprocesstodrivecross-organizationalcollaborationtocomeupwithideastheycanturnintoactions

</div>
<span class='text_page_counter'>(23)</span><div class='page_container' data-page=23></div>
<span class='text_page_counter'>(24)</span><div class='page_container' data-page=24>

<b>PartI</b>

<b>BusinessPotentialofBigData</b>

Chapters1through4setthefoundationfordrivingbusinessstrategieswithdata
science.Inparticular,theBigDataBusinessModelMaturityIndexhighlightsthe
realmofwhat'spossiblefromabusinesspotentialperspectivebyprovidingaroad
mapthatmeasurestheeffectivenessofyourorganizationtoleveragedataand
analyticstopoweryourbusinessmodels.

<b>InThisPart</b>

Chapter1:TheBigDataBusinessMandate

Chapter2:BigDataBusinessModelMaturityIndex
Chapter3:TheBigDataStrategyDocument

</div>
<span class='text_page_counter'>(25)</span><div class='page_container' data-page=25></div>
<span class='text_page_counter'>(26)</span><div class='page_container' data-page=26>

<b>Chapter1</b>

<b>TheBigDataBusinessMandate</b>

<i>Havingtroublegettingyourseniormanagementteamtounderstandthe</i>
<i>businesspotentialofbigdata?Can'tgetyourmanagementleadershipto</i>
<i>considerbigdatatobesomethingotherthananITscienceexperiment?Are</i>

<i>yourline-of-businessleadersunwillingtocommitthemselvesto</i>

<i>understandinghowdataandanalyticscanpowertheirtopinitiatives?</i>
<i>Ifso,thenthis“BigDataSeniorExecutiveCarePackage”isforyou!</i>
<i>Andforalimitedtime,yougetanunlimitedlicensetosharethiscare</i>
<i>packagewithasmanyseniorexecutivesasyoudesire.Butyoumustact</i>
<i>NOW!Becomethelifeofthecompanypartieswithyourextensive</i>

<i>knowledgeofhownewcustomer,product,andoperationalinsightscan</i>
<i>guideyourorganization'svaluecreationprocesses.Andmaybe,justmaybe,</i>
<i>getapromotionintheprocess!!</i>

<b>NOTE</b>

</div>
<span class='text_page_counter'>(27)</span><div class='page_container' data-page=27>

<b>BigDataMBAIntroduction</b>

Thedayswhenbusinessusersandbusinessmanagementcanrelinquishcontrolof
dataandanalyticstoITareover,oratleastfororganizationsthatwanttosurvive
beyondtheimmediateterm.Thebigdatadiscussionnowneedstofocusonhow
organizationscancouplenewsourcesofcustomer,product,andoperationaldata
withadvancedanalytics(datascience)topowertheirkeybusinessprocessesand
<i>elevatetheirbusinessmodels.Organizationsneedtounderstandthattheydonot</i>
<i>needabigdatastrategyasmuchastheyneedabusinessstrategythat</i>
<i>incorporatesbigdata.</i>
<i>TheBigDataMBAchallengesthethinkingthatdataandanalyticsareancillaryor</i>
<i>a“bolton”tothebusiness;thatdataandanalyticsaresomeoneelse'sproblem.In</i>
agrowingnumberofleadingorganizations,dataandanalyticsarecriticalto
businesssuccessandlong-termsurvival.Businessleadersandbusinessusers
readingthisbookwilllearnwhytheymusttakeresponsibilityforidentifying
whereandhowtheycanapplydataandanalyticstotheirbusinesses—otherwise


theyputtheirbusinessesatriskofbeingmadeobsoletebymorenimble,data-drivencompetitors.
<i>TheBigDataMBAintroducesanddescribesconcepts,techniques,methodologies,</i>
<i><b>andhand-onexercisestoguideyouasyouseektoaddressthebigdatabusiness</b></i>
<i><b>mandate.Thebookprovideshands-onexercisesandhomeworkassignmentsto</b></i>
maketheseconceptsandtechniquescometolifeforyourorganization.Itprovides
recommendationsandactionsthatenableyourorganizationtostarttoday.Andin
<i>theprocess,BigDataMBAteachesyouto“thinklikeadatascientist.”</i>

</div>
<span class='text_page_counter'>(28)</span><div class='page_container' data-page=28>

<b>Figure1.1</b>BigDataBusinessModelMaturityIndex

TheBigDataBusinessModelMaturityIndexprovidesaroadmapforhow

organizationscanintegratedataandanalyticsintotheirbusinessmodels.TheBig
DataBusinessModelMaturityIndexiscomposedofthefollowingfivephases:

<b>Phase1:BusinessMonitoring.IntheBusinessMonitoringphase,</b>

organizationsareleveragingdatawarehousingandBusinessIntelligenceto
monitortheorganization'sperformance.

<b>Phase2:BusinessInsights.TheBusinessInsightsphaseisabout</b>

leveragingpredictiveanalyticstouncovercustomer,product,andoperational
insightsburiedinthegrowingwealthofinternalandexternaldatasources.In
thisphase,organizationsaggressivelyexpandtheirdataacquisitioneffortsby
couplingalloftheirdetailedtransactionalandoperationaldatawithinternal
datasuchasconsumercomments,e-mailconversations,andtechniciannotes,
aswellasexternalandpubliclyavailabledatasuchassocialmedia,weather,
traffic,economic,demographics,homevalues,andlocaleventsdata.

<b>Phase3:BusinessOptimization.IntheBusinessOptimizationphase,</b>

organizationsapplyprescriptiveanalyticstothecustomer,product,and
operationalinsightsuncoveredintheBusinessInsightsphasetodeliver
actionableinsightsorrecommendationstofrontlineemployees,business

managers,andchannelpartners,aswellascustomers.ThegoaloftheBusiness
Optimizationphaseistoenableemployees,partners,andcustomersto

optimizetheirkeydecisions.

<b>Phase4:DataMonetization.IntheDataMonetizationphase,</b>

organizationsleveragethecustomer,product,andoperationalinsightsto
createnewsourcesofrevenue.Thiscouldincludesellingdata—orinsights—
intonewmarkets(acellularphoneprovidersellingcustomerbehavioraldata
toadvertisers),integratinganalyticsintoproductsandservicestocreate

</div>
<span class='text_page_counter'>(29)</span><div class='page_container' data-page=29>

<b>Phase5:BusinessMetamorphosis.TheholygrailoftheBigData</b>

BusinessModelMaturityIndexiswhenanorganizationtransitionsits

businessmodelfromsellingproductstoselling“business-as-a-service.”Think
GEselling“thrust”insteadofjetengines.ThinkJohnDeereselling“farming
optimization”insteadoffarmingequipment.ThinkBoeingselling“airmiles”
insteadofairplanes.Andintheprocess,theseorganizationswillcreatea

platformenablingthird-partydeveloperstobuildandmarketsolutionsontop
oftheorganization'sbusiness-as-a-servicebusinessmodel.

Ultimately,bigdataonlymattersifithelpsorganizationsmakemoremoneyand
improveoperationaleffectiveness.Examplesincludeincreasingcustomer

acquisition,reducingcustomerchurn,reducingoperationalandmaintenance
costs,optimizingpricesandyield,reducingrisksanderrors,improving

compliance,improvingthecustomerexperience,andmore.

</div>
<span class='text_page_counter'>(30)</span><div class='page_container' data-page=30></div>
<span class='text_page_counter'>(31)</span><div class='page_container' data-page=31></div>
<span class='text_page_counter'>(32)</span><div class='page_container' data-page=32>

innovationslikeHadoopandSpark,therealdiscussionshouldbeaboutthe

economicimpactofbigdata.Newtechnologiesdon'tdisruptbusinessmodels;it's
whatorganizationsdowiththesenewtechnologiesthatdisruptsbusinessmodels
andenablesnewones.Let'sreviewanexampleofonesucheconomic-driven
businesstransformation:thesteamengine.

Thesteamengineenabledurbanization,industrialization,andtheconqueringof
newterritories.Itliterallyshrankdistanceandtimebyreducingthetimerequired
tomovepeopleandgoodsfromonesideofacontinenttotheother.Thesteam
engineenabledpeopletoleavelow-payingagriculturaljobsandmoveintocities
forhigher-payingmanufacturingandclericaljobsthatledtoahigherstandardof
living.

Forexample,citiessuchasLondonshotupintermsofpopulation.In1801,before
theadventofGeorgeStephenson'sRocketsteamengine,Londonhad1.1million
residents.Aftertheinvention,thepopulationofLondonmorethandoubledto2.7
millionresidentsby1851.Londontransformedthenucleusofsocietyfromsmall
tight-knitcommunitieswheretextileproductionandagriculturewereprevalent
intobigcitieswithavarietyofjobs.Thesteamlocomotiveprovidedquicker
transportationandmorejobs,whichinturnbroughtmorepeopleintothecities
anddrasticallychangedthejobmarket.By1861,only2.4percentofLondon's

populationwasemployedinagriculture,while49.4percentwereinthe

manufacturingortransportationbusiness.Thesteamlocomotivewasamajor
turningpointinhistoryasittransformedsocietyfromlargelyruraland

agriculturalintourbanandindustrial.2

</div>
<span class='text_page_counter'>(33)</span><div class='page_container' data-page=33></div>
<span class='text_page_counter'>(34)</span><div class='page_container' data-page=34></div>
<span class='text_page_counter'>(35)</span><div class='page_container' data-page=35></div>
<span class='text_page_counter'>(36)</span><div class='page_container' data-page=36>

warehousing)andanalytics(datascience)environmentsdifferently.Thesetwo
environmentshaveverydifferentcharacteristicsandservedifferentpurposes.The
datalakecanmakebothoftheBIanddatascienceenvironmentsmoreagileand
moreproductive(Figure1.2).

<b>Figure1.2</b>Moderndata/analyticsenvironment

<b>CROSS-REFERENCE</b>

Chapter7(”TheDataLake“)introducestheconceptofadatalakeandtherole
thedatalakeplaysinsupportingyourexistingdatawarehouseandBusiness
Intelligenceinvestmentswhileprovidingthefoundationforyourdatascience
environment.Chapter7discusseshowthedatalakecanun-cuffyourdata
scientistsfromthedatawarehousetouncoverthosevariablesandmetricsthat
mightbebetterpredictorsofbusinessperformance.Italsodiscusseshowthe
datalakecanfreeupexpensivedatawarehouseresources,especiallythose
resourcesassociatedwithExtract,Transform,andLoad(ETL)dataprocesses.

<b>Don'tThink“WhatHappened,”Think“WhatWillHappen”</b>

Businessusershavebeentrainedtocontemplatebusinessquestionsthatmonitor
thecurrentstateofthebusinessandtofocusonretrospectivereportingonwhat
happened.BusinessusershavebecomeconditionedbytheirBIanddata

warehouseenvironmentstoonlyconsiderquestionsthatreportoncurrent
businessperformance,suchas“HowmanywidgetsdidIselllastmonth?”and
“Whatweremygrosssaleslastquarter?”

</div>
<span class='text_page_counter'>(37)</span><div class='page_container' data-page=37></div>
<span class='text_page_counter'>(38)</span><div class='page_container' data-page=38>

Unfortunately,thattypeofthinkinghasledtosiloeddatafiefdoms,siloed
decisions,andanun-empoweredandfrustratedbusinessteam.Organizations
needtothinkdifferentlyabouthowtheyempoweralloftheiremployees.

Organizationsneedtofindawaytopromoteandnurturecreativethinkingand
groundbreakingideasacrossalllevelsoftheorganization.Thereisnoedictthat
statesthatthebestideasonlycomefromseniormanagement.

Thekeytobigdatasuccessisempoweringcross-functionalcollaborationand
exploratorythinkingtochallengelong-heldorganizationalrulesofthumb,
heuristics,and“gut”decisionmaking.Thebusinessneedsanapproachthatis
inclusiveofallthekeystakeholders—IT,businessusers,businessmanagement,
channelpartners,andultimatelycustomers.Thebusinesspotentialofbigdatais
onlylimitedbythecreativethinkingoftheorganization.

<b>CROSS-REFERENCE</b>

Chapter13(“PowerofEnvisioning”)discusseshowtheBIanddatascience
teamscancollaboratetobrainstorm,test,andrefinenewvariablesthatmight
bebetterpredictorsofbusinessperformance.Wewillintroduceseveral

techniquesandconceptsthatcanbeusedtodrivecollaborationbetweenthe
businessandITstakeholdersandultimatelyhelpyourdatascienceteam
uncovernewcustomer,product,andoperationalinsightsthatleadtobetter
businessperformance.Chapter14(“OrganizationalRamifications”)

</div>
<span class='text_page_counter'>(39)</span><div class='page_container' data-page=39></div>
<span class='text_page_counter'>(40)</span><div class='page_container' data-page=40></div>
<span class='text_page_counter'>(41)</span><div class='page_container' data-page=41>

<b>HomeworkAssignment</b>

Usethefollowingexercisestoapplywhatyoulearnedinthischapter.

<b>Exercise#1:Identifyakeybusinessinitiativeforyourorganization,</b>

somethingthebusinessistryingtoaccomplishoverthenext9to12months.It
mightbesomethinglikeimprovecustomerretention,optimizecustomer

acquisition,reducecustomerchurn,optimizepredictivemaintenance,reduce
revenuetheft,andsoon.

<b>Exercise#2:Brainstormandwritedownwhat(1)customer,(2)product,and</b>

(3)operationalinsightsyourorganizationwouldliketouncoverinorderto
supportthetargetedbusinessinitiative.Startbycapturingthedifferenttypes
ofdescriptive,predictive,andprescriptivequestionsyou'dliketoanswerabout
thetargetedbusinessinitiative.Tip:Don'tworryaboutwhetherornotyou
havethedatasourcesyouneedtoderivetheinsightsyouwant(yet).

<b>Exercise#3:Brainstormandwritedowndatasourcesthatmightbeusefulin</b>

uncoveringthosekeyinsights.Lookbothinternallyandexternallyfor

</div>
<span class='text_page_counter'>(42)</span><div class='page_container' data-page=42>

<b>Notes</b>

1_{Hopkins,Brian,FatemehKhatibloowithKyleMcNabb,JamesStaten,Andras}

Cser,HolgerKisker,Ph.D.,LeslieOwens,JenniferBelissent,Ph.D.,Abigail

Komlenic,“ResetOnBigData:EmbraceBigDatatoEngageCustomersat
Scale,”ForresterResearch,2014.

</div>
<span class='text_page_counter'>(43)</span><div class='page_container' data-page=43></div>
<span class='text_page_counter'>(44)</span><div class='page_container' data-page=44>

<b>Chapter2</b>

<b>BigDataBusinessModelMaturityIndex</b>

Organizationsdonotunderstandhowfarbigdatacantakethemfromabusiness
transformationperspective.Organizationsdon'thaveawayofunderstanding
whattheultimatebigdataendstatewouldorcouldlooklikeoranswering
questionssuchas:

WhereandhowshouldIstartmybigdatajourney?

HowcanIcreatenewrevenueormonetizationopportunities?

HowdoIcomparetootherswithrespecttomyorganization'sadoptionofbig
dataasabusinessenabler?

HowfarcanIpushbigdatatopower—eventransform—mybusinessmodels?
<i><b>Tohelpaddressthesetypesofquestions,I'vecreatedtheBigDataBusiness</b></i>
<i><b>ModelMaturityIndex.Notonlycanorganizationscanusethisindexto</b></i>
understandwheretheysitwithrespecttootherorganizationsinexploitingbig
dataandadvancedanalyticstopowertheirbusinessmodels,buttheindex

providesaroadmaptohelporganizationsacceleratetheintegrationofdataand
analyticsintotheirbusinessmodels.

</div>
<span class='text_page_counter'>(45)</span><div class='page_container' data-page=45>

<b>Chapter2Objectives</b>

IntroducetheBigDataBusinessModelMaturityIndexasaframeworkfor
organizationstomeasurehoweffectivetheyareatleveragingdataand
analyticstopowertheirbusinessmodels

Discusstheobjectivesandcharacteristicsofeachofthefivephasesofthe
BigDataBusinessModelMaturityIndex:BusinessMonitoring,Business
Insights,BusinessOptimization,DataMonetization,andBusiness

Metamorphosis

<i><b>Discusshowtheeconomicsofbigdataandthefourbigdatavalue</b></i>
<i><b>driverscanenableorganizationstocrosstheanalyticschasmand</b></i>
advancepasttheBusinessMonitoringphaseintotheBusinessInsights
andBusinessOptimizationphases

</div>
<span class='text_page_counter'>(46)</span><div class='page_container' data-page=46>

<b>IntroducingtheBigDataBusinessModelMaturity</b>

<b>Index</b>

Organizationsaremovingatdifferentpaceswithrespecttowhereandhowthey
areadoptingbigdataandadvancedanalyticstocreatebusinessvalue.Some

organizationsaremovingverycautiously,astheyareunclearastowhereandhow
tostartandwhichofthebevyofnewtechnologyinnovationstheyneedtodeploy
inordertostarttheirbigdatajourneys.Othersaremovingatamoreaggressive
pacebyacquiringandassemblingabigdatatechnologyfoundationbuiltonmany
newbigdatatechnologiessuchasHadoop,Spark,MapReduce,YARN,Mahout,
Hive,HBase,andmore.

However,aselectfewarelookingbeyondjustthetechnologytoidentifywhereand
howtheyshouldbeintegratingbigdataintotheirexistingbusinessprocesses.

Theseorganizationsareaggressivelylookingtoidentifyandexploitopportunities
tooptimizekeybusinessprocesses.Andtheseorganizationsareseekingnew

monetizationopportunities;thatis,seekingoutbusinessopportunitieswherethey
can

Packageandselltheiranalyticinsightstoothers

Integrateadvancedanalyticsintotheirproductsandservicestocreate
“intelligent”products

Createentirelynewproductsandservicesthathelpthementernewmarkets
andtargetnewcustomers

Thesearethefolkswhorealizethattheydon'tneedabigdatastrategyasmuchas
theyneedabusinessstrategythatincorporatesbigdata.Andwhenorganizations
“flipthatbyte”onthefocusoftheirbigdatainitiatives,thebusinesspotentialis
almostboundless.

OrganizationscanusetheBigDataBusinessModelMaturityIndexasa

</div>
<span class='text_page_counter'>(47)</span><div class='page_container' data-page=47>

<b>Figure2.1</b>BigDataBusinessModelMaturityIndex

OrganizationstendtofindthemselvesinoneoffivephasesontheBigData
BusinessModelMaturityIndex:

<b>Phase1:BusinessMonitoring.IntheBusinessMonitoringphase,</b>

organizationsareapplyingdatawarehousingandBusinessIntelligence

techniquesandtoolstomonitortheorganization'sbusinessperformance(also
calledBusinessPerformanceManagement).

<b>Phase2:BusinessInsights.IntheBusinessInsightsphase,organizations</b>

<i>aggressivelyexpandtheirdataassetsbyamassingalloftheirdetailed</i>
transactionalandoperationaldataandcouplingthattransactionaland

operationaldatawithnewsourcesofinternaldata(e.g.,consumercomments,
e-mailconversations,techniciannotes)andexternaldata(e.g.,socialmedia,
weather,traffic,economic,data.gov)sources.OrganizationsintheBusiness
Insightsphasethenusepredictiveanalyticstouncovercustomer,product,and
operationalinsightsburiedinandacrossthesedatasources.

<b>Phase3:BusinessOptimization.IntheBusinessOptimizationphase,</b>

organizationsbuildonthecustomer,product,andoperationalinsights

uncoveredintheBusinessInsightsphasebyapplyingprescriptiveanalyticsto
optimizekeybusinessprocesses.OrganizationsintheBusinessOptimization
phasepushtheanalyticresults(e.g.,recommendations,scores,rules)to

frontlineemployeesandbusinessmanagerstohelpthemoptimizethetargeted
businessprocessthroughimproveddecisionmaking.TheBusiness

Optimizationphasealsoprovidesopportunitiesfororganizationstopush
analyticinsightstotheircustomersinordertoinfluencecustomerbehaviors.
AnexampleoftheBusinessOptimizationphaseisaretailerthatdelivers
analytic-basedmerchandisingrecommendationstothestoremanagersto
optimizemerchandisemarkdownsbasedonpurchasepatterns,inventory,

weatherconditions,holidays,consumercomments,andsocialmediapostings.

<b>Phase4:DataMonetization.TheDataMonetizationphaseiswhere</b>

</div>
<span class='text_page_counter'>(48)</span><div class='page_container' data-page=48></div>
<span class='text_page_counter'>(49)</span><div class='page_container' data-page=49></div>
<span class='text_page_counter'>(50)</span><div class='page_container' data-page=50></div>
<span class='text_page_counter'>(51)</span><div class='page_container' data-page=51></div>
<span class='text_page_counter'>(52)</span><div class='page_container' data-page=52>

theaveragecampaignperformanceincertainmarketsoncertaindaysof
theweek

Customersthatarereactingtwotothreestandarddeviationsoutsidethe
normintheirpurchasepatternsforcertainproductcategoriesincertain
weatherconditions

Supplierswhosecomponentsareoperatingoutsidetheupperorlower
limitsofacontrolchartinextremecoldweathersituations

<b>CROSS-REFERENCE</b>

Forthepredictiveanalyticstobeeffective,organizationsneedtobuild
detailedanalyticprofilesforeachindividualbusinessentity—customers,
patients,students,windturbines,jetengines,ATMs,etc.Thecreationand
roleofanalyticprofilesisatopiccoveredinChapter5,“DifferencesBetween
BusinessIntelligenceandDataScience.”

<i><b>BusinessInsightsPhaseChallenge</b></i>

TheBusinessInsightsphaseisthemostdifficultstageoftheBigDataBusiness
ModelMaturityIndexbecauseitrequiresorganizationsto“thinkdifferently”
abouthowtheyapproachdataandanalytics.Therules,techniques,and

approachesthatworkedintheBusinessIntelligenceanddatawarehouseworlds
donotnecessaryapplytotheworldofbigdata.Thisistrulythe“crossingthe

analyticschasm”moment(seeFigure2.2).

<b>Figure2.2</b>Crossingtheanalyticschasm

</div>
<span class='text_page_counter'>(53)</span><div class='page_container' data-page=53></div>
<span class='text_page_counter'>(54)</span><div class='page_container' data-page=54></div>
<span class='text_page_counter'>(55)</span><div class='page_container' data-page=55>

<b>Figure2.3</b>Packagingandsellingaudienceinsights

Integratinganalyticinsightsdirectlyintoanorganization'sproductsand
servicestocreate“intelligent”productsorservices,suchas:

Carsthatlearnacustomer'sdrivingpatternsandbehaviorsandadjust
drivercontrols,seats,mirrors,brakepedals,suspension,steering,
dashboarddisplays,etc.tomatchthecustomer'sdrivingstyle
TelevisionsandDVRsthatlearnwhattypesofshowsandmoviesa

customerlikesandsearchacrossthedifferentcableandInternetchannels
tofindandautomaticallyrecordsimilarshowsforthatcustomer

Ovensthatlearnhowacustomerlikescertainfoodspreparedandcooks
theminthatmannerautomaticallyandalsoincluderecommendationsfor
otherfoodsandrecipesthat“otherslikeyou”enjoy

Jetenginesthatcaningestweather,elevation,windspeed,andother
environmentaldatatomakeadjustmentstobladeangles,tilt,yaw,and
rotationspeedstominimizefuelconsumptionduringflight

Repackaginginsightstocreateentirelynewproductsandservicesthathelp
organizationstoenternewmarketsandtargetnewcustomersoraudiences.
Forexample,organizationscancapture,analyze,andpackagecustomer,
product,andoperationalinsightsacrosstheoverallmarketinordertohelp
channelpartnerstomoreeffectivelymarketandselltotheircustomers,such

as:

Onlinedigitalmarketplaces(Yahoo,Google,eBay,Facebook)could
leveragegeneralmarkettrendsandothermerchantperformancedatato
providerecommendationstosmallmerchantsoninventory,ordering,
merchandising,marketing,andpricing.

</div>
<span class='text_page_counter'>(56)</span><div class='page_container' data-page=56></div>
<span class='text_page_counter'>(57)</span><div class='page_container' data-page=57>

patterns,localweather,localwaterquality,andlocalenvironmentalconditions
suchaslocalwaterconservationeffortsandenergycosts

Retailersmovingintothe“ShoppingOptimization”businessbyrecommending
specificproductsgivencustomers'currentbuyingpatternsascomparedwith
otherslikethem,includingrecommendationsforproductsthattheymaynot
evensell(think“Miracleon43rdStreet”)

Airlinesmovingintothe“TravelDelight”businessofnotonlyoffering

</div>
<span class='text_page_counter'>(58)</span><div class='page_container' data-page=58>

<b>BigDataBusinessModelMaturityIndexLessons</b>

<b>Learned</b>

Therearesomeinterestinglessonsthatorganizationswilldiscoverasthey
progressthroughthephasesoftheBigDataBusinessModelMaturityIndex.
Understandingtheselessonsaheadoftimeshouldhelpprepareorganizationsfor
theirbigdatajourney.

<b>Lesson1:FocusInitialBigDataEffortsInternally</b>

ThefirstthreephasesoftheBigDataBusinessModelMaturityIndexseekto
extractmorefinancialorbusinessvalueoutoftheorganization'sinternal

processesorbusinessinitiatives.Thefirstthreephasesdrivebusinessvalueanda
ReturnonInvestment(ROI)byseekingtointegratenewsourcesofcustomer,
product,operational,andmarketdatawithadvancedanalyticstoimprovethe
decisionsthataremadeaspartoftheorganization'skeyinternalprocessand
businessinitiatives(seeFigure2.4).

<b>Figure2.4</b>Optimizeinternalprocesses

<i><b>Theinternalprocessoptimizationeffortsstartbyseekingtoleveragethe</b></i>
organization'sBusinessIntelligenceanddatawarehouseassets.Thisincludes
buildingonthedatawarehouse'sdatasources,dataextractionandenrichment
algorithms,dimensions,metrics,keyperformanceindicators,reports,and

</div>
<span class='text_page_counter'>(59)</span><div class='page_container' data-page=59>

<b>TheFourBigDataValueDrivers</b>

1. Accesstoalltheorganization'sdetailedtransactionalandoperationaldata
atthelowestlevelofgranularity(attheindividualcustomer,machine,or
devicelevel).

2. Integrationofunstructureddatafrombothinternal(consumercomments,
e-mailthreads,techniciannotes)andexternalsources(socialmedia,

mobile,publiclyavailable)withthedetailedtransactionalandoperational
datatoprovidenewmetricsandnewdimensionsagainstwhichto

optimizekeybusinessprocesses.

3. Leveragereal-time(orright-time)dataanalysistoacceleratethe

organization'sabilitytoidentifyandactoncustomer,product,andmarket

opportunitiesinatimeliermanner.

4. Applypredictiveanalyticsanddataminingtouncovercustomer,product,
andoperationalinsightsorareasof“unusualness”buriedinthemassive
volumesofdetailedstructuredandunstructureddatathatareworthyof
furtherbusinessinvestigation.

Organizationsmustleveragethesefourbigdatavaluedriverstocrosstheanalytics
chasmbyuncoveringnewcustomer,product,andoperationalinsightsthatcanbe
usedtooptimizekeybusinessprocesses—whetherdeliveringactionable

recommendationstofrontlineemployeesandbusinessmanagersordelivering
“NextBestOffer”orrecommendationstodelightcustomersandbusiness
partners.

<b>Lesson2:LeverageInsightstoCreateNewMonetization</b>

<b>Opportunities</b>

ThelasttwophasesoftheBigDataBusinessModelMaturityIndexarefocusedon
externalmarketopportunities;opportunitiestocreatenewmonetizationor

</div>
<span class='text_page_counter'>(60)</span><div class='page_container' data-page=60>

<b>Figure2.5</b>Createnewmonetizationopportunities

Thisisthepartofthebigdatajourneythatcatchesmostorganizations'attention:
theopportunitytoleveragetheinsightsgatheredthroughtheoptimizationoftheir
keybusinessprocessestocreatenewrevenueormonetizationopportunities.

Organizationsareeagertoleveragenewcorporateassets—data,analytics,and
businessinsights—inordertocreatenewsourcesofrevenue.Thisisthe“4Ms”
phaseoftheBigDataBusinessModelMaturityIndexwhereorganizationsfocus

<i>onleveragingdataandanalyticstocreatenewopportunitiesto“MakeMeMore</i>
<i>Money!”</i>

<b>Lesson3:PreparingforOrganizationalTransformation</b>

Tofullyexploitthebigdataopportunity,subtleorganizationalandcultural
changeswillbenecessaryfortheorganizationtoadvancealongthematurity
index.Iforganizationsareseriousaboutintegratingdataandanalyticsintotheir
businessmodels,thenthreeorganizationalorculturaltransformationswillneed
totakeplace:

<b>1.TreatDataasanAsset.Organizationsmuststarttotreatdataasanasset</b>

tobenurturedandgrown,notacosttobeminimized.Organizationsmust
developaninsatiableappetiteformoreandmoredata—eveniftheyare
unclearastohowtheywillusethatdata.Thisisasignificantculturalchange
fromthedatawarehousedayswherewetreateddataasacosttobeminimized.

<b>2.LegallyProtectYourAnalyticsIntellectualProperty.Organizations</b>

mustputintoplaceformalprocessesandprocedurestocapture,track,refine,
andevenlegallyprotecttheiranalyticassets(e.g.,analyticmodels,data

enrichmentalgorithms,andanalyticresultssuchasscores,recommendations,
andassociationrules)askeyorganizationalintellectualproperty.Whilethe
underlyingtechnologiesmaychangeovertime,theresultingdataandanalytic

</div>
<span class='text_page_counter'>(61)</span><div class='page_container' data-page=61>

<b>3.GetComfortableUsingDatatoGuideDecisions.Business</b>

managementandbusinessusersmustgainconfidenceinusingdataand
analyticstoguidetheirdecisionmaking.Organizationsmustgetcomfortable
withmakingbusinessdecisionsbasedonwhatthedataandtheanalyticstell
themversusdefaultingtothe“HighestPaidPerson'sOpinion”(HIPPO).The
organization'sinvestmentsindata,analytics,people,processes,and

technologywillbefornaughtiftheorganizationisn'tpreparedtomake
decisionsbasedonwhatthedataandtheanalyticstellthem.Withthatsaid,
it'simportantthattheanalyticinsightsarepositionedas“recommendations”
thatbusinessusersandbusinessmanagementcanaccept,reject,ormodify.In
thatway,organizationscanleverageanalyticstoestablishorganizational

</div>
<span class='text_page_counter'>(62)</span><div class='page_container' data-page=62></div>
<span class='text_page_counter'>(63)</span><div class='page_container' data-page=63>

throughimproveddecisionmaking(orimprovedoperationaleffectivenessfor
non-profitorganizations).Bigdataholdsthepotentialtobothoptimizekey
businessprocessesandcreatenewmonetizationorrevenueopportunities.
Insummary:

TheBigDataBusinessModelMaturityIndexprovidesaframeworkfor
organizationstomeasurehoweffectivetheyareatleveragingdataand
analyticstopowertheirbusinessmodels.

ThefivephasesoftheBigDataBusinessModelMaturityIndexareBusiness
Monitoring,BusinessInsights,BusinessOptimization,DataMonetization,and
BusinessMetamorphosis.

Theeconomicsofbigdataandthefourbigdatavaluedriverscanenable
organizationstocrosstheanalyticschasm.

</div>
<span class='text_page_counter'>(64)</span><div class='page_container' data-page=64>

<b>HomeworkAssignment</b>

Usethefollowingexercisestoapplyandreinforcetheinformationpresentedin
thischapter:

<b>Exercise#1:Listtwoorthreeofyourorganization'skeybusinessprocesses.</b>

Thatis,writedowntwoorthreebusinessprocessesthatuniquelydifferentiate
yourorganizationfromyourcompetition.

<b>Exercise#2:Listthefourbigdatavaluedriversthatareenabledbythe</b>

economicsofbigdataanddescribehoweachmightimpactoneofyour
organization'skeybusinessprocessesidentifiedinExercise#1.

<b>Exercise#3:FortheselectedkeybusinessprocessesidentifiedinExercise</b>

#1,describehoweachkeybusinessprocessmightbeimprovedasittransitions
alongthefivephasesoftheBigDataBusinessModelMaturityIndex.Identify
thecustomer,product,andoperationalramificationsthateachofthefive
phasesmighthaveontheselectedkeybusinessprocess.

<b>Exercise#4:Listtheculturalchangesthatyourorganizationmustaddressif</b>

</div>
<span class='text_page_counter'>(65)</span><div class='page_container' data-page=65></div>
<span class='text_page_counter'>(66)</span><div class='page_container' data-page=66>

<b>Chapter3</b>

<b>TheBigDataStrategyDocument</b>

Oneofthebiggestchallengesorganizationsfacewithrespecttobigdatais

<i><b>identifyingwhereandhowtostart.Thebigdatastrategydocument,detailed</b></i>
inthischapter,providesaframeworkforlinkinganorganization'sbusiness

strategyandsupportingbusinessinitiativestotheorganization'sbigdataefforts.
Thebigdatastrategydocumentguidestheorganizationthroughtheprocessof
breakingdownitsbusinessstrategyandbusinessinitiativesintopotentialbigdata
businessusecasesandthesupportingdataandanalyticrequirements.

<b>NOTE</b>

<i>ThebigdatastrategydocumentfirstappearedinmybookBigData:</i>

<i>UnderstandingHowDataPowersBigBusiness.Sincethenandcourtesyof</i>
severalclientengagements,significantimprovementshavebeenmadetohelp
userstouncoverbigdatausecases.Inparticular,theprocesshasbeen

enhancedtoclarifythebusinessvalueandimplementationfeasibility
assessmentsofthedifferentdatasourcesandusecaseprioritization(see
Figure3.1).

</div>
<span class='text_page_counter'>(67)</span><div class='page_container' data-page=67>

<b>Chapter3Objectives</b>

Establishcommonterminologyforbigdata.

Examinetheconceptofabusinessinitiativeandprovidesomeexamplesof
wheretofindthesebusinessinitiatives.

Introducethebigdatastrategydocumentasaframeworkforhelping
organizationstoidentifytheusecasesthatguidewhereandhowtheycan
starttheirbigdatajourneys.

Provideahands-onexampleofthebigdatastrategydocumentinaction

usingChipotle,achainoforganicMexicanfoodrestaurants(andoneof
myfavoriteplacestoeat!).

Introduceworksheetstohelporganizationstodeterminethebusiness
valueandimplementationfeasibilityofthedatasourcesthatcomeoutof
thebigdatastrategydocumentprocess.

<i><b>Introducetheprioritizationmatrixasatoolthatcandrivebusiness</b></i>
andITalignmentaroundprioritizingtheusecasesbasedonbusinessvalue
andimplementationfeasibilityovera9-to12-monthwindow.

</div>
<span class='text_page_counter'>(68)</span><div class='page_container' data-page=68>

<b>EstablishingCommonBusinessTerminology</b>

Beforewelaunchintothebigdatastrategydocumentdiscussion,weneedto
defineafewcriticaltermstoensurethatweareusingconsistentterminology
throughoutthechapterandthebook:

<i><b>CorporateMission.Whytheorganizationexists;defineswhatan</b></i>

organizationisandtheorganization'sreasonforbeing.Forexample,TheWalt
DisneyCompany'scorporatemissionis“tobeoneoftheworld'sleading

producersandprovidersofentertainmentandinformation.”1

<i><b>BusinessStrategy.Howtheorganizationisgoingtoachieveitsmissionover</b></i>

thenexttwotothreeyears.

<i><b>StrategicBusinessInitiatives.Whattheorganizationplanstodoto</b></i>

achieveitsbusinessstrategyoverthenext9to12months;usuallyincludes
businessobjectives,financialtargets,metrics,andtimeframes.

<b>BusinessEntities.Thephysicalobjectsorentities(e.g.,customers,patients,</b>

students,doctors,windturbines,trucks)aroundwhichthebusinessinitiative
willtrytounderstand,predict,andinfluencebehaviorsandperformance
<i><b>(sometimesreferredtoasthestrategicnounsofthebusiness).</b></i>

<b>BusinessStakeholders.Thosebusinessfunctions(sales,marketing,</b>

finance,storeoperations,logistics,andsoon)thatimpactorareimpactedby
thestrategicbusinessinitiative.

<b>BusinessDecisions.Thedecisionsthatthebusinessstakeholdersneedto</b>

makeinsupportofthestrategicbusinessinitiative.

<i><b>BigDataUseCases.Theanalyticusecases(decisionsandcorresponding</b></i>

actions)thatsupportthestrategicbusinessinitiative.

<b>Data.Thestructuredandunstructureddatasources,bothinternaland</b>

</div>
<span class='text_page_counter'>(69)</span><div class='page_container' data-page=69>

<b>IntroducingtheBigDataStrategyDocument</b>

Thebigdatastrategydocumenthelpsorganizationsaddressthechallengeof
identifyingwhereandhowtostarttheirbigdatajourneys.Thebigdatastrategy

documentusesasingle-pageformatthatanyorganizationcanuse(profitornon-profit)thatlinksanorganization'sbigdataeffortstoitsbusinessstrategyandkey

businessinitiatives.Thebigdatastrategydocumentiseffectiveforthefollowing
reasons:

It'sconcise.Itfitsonasinglepagesothatanyoneintheorganizationcan
quicklyreviewittoensureheorsheisworkingonthetoppriorityitems.
It'sclear.Itclearlydefineswhattheorganizationneedstodoinorderto
achievetheorganization'skeybusinessinitiatives.

It'sbusinessrelevant.Itstartsbyfocusingonthebusinessstrategyand
supportinginitiativesbeforeitdivesintothedataandtechnology

requirements.

Thebigdatastrategydocumentiscomposedofthefollowingsections(seeFigure
3.2):

Businessstrategy

Keybusinessinitiatives
Keybusinessentities
Keydecisions

</div>
<span class='text_page_counter'>(70)</span><div class='page_container' data-page=70>

<b>Figure3.2</b>Bigdatastrategydocument

Therestofthechapterwilldetaileachofthesesectionsandprovideguidelinesfor
howtheorganizationcantriagetheorganization'sbusinessstrategyintothe

financialdrivers(orusecases)onwhichtheorganizationcanfocusitsbigdata
efforts.WewilluseacasestudyaroundChipotleMexicanGrillstoreinforcethe
triageandanalysisprocess.

<b>IdentifyingtheOrganization'sKeyBusinessInitiatives</b>

Thestartingpointforthebigdatastrategydocumentprocessistoidentifythe
organization'sbusinessinitiativesoverthenext9to12months.Thatis,whatis

thebusinesstryingtoaccomplishoverthenext9to12months?This9-to12-monthtimeframeiscritical,asit

Focusestheorganization'sbigdataeffortsonsomethingthatisofimmediate
valueandrelevancetothebusiness

Createsasenseofurgencyfortheorganizationtomovequicklyanddiligently
Givesthebigdataprojectamorerealisticchanceofdeliveringapositive
ReturnonInvestment(ROI)andafinancialpaybackin12monthsorless
Abusinessinitiativesupportsthebusinessstrategyandhasthefollowing
characteristics:

Criticaltoimmediate-termbusinessand/orfinancialperformance(usually9-to12-monthtimeframe)

</div>
<span class='text_page_counter'>(71)</span><div class='page_container' data-page=71></div>
<span class='text_page_counter'>(72)</span><div class='page_container' data-page=72>

<b>Figure3.3</b>Chipotle's2012lettertotheshareholders

FromthePresident'sLetter,wecanidentifyatleastfourkeybusinessinitiatives
forthecomingyear:

Improveemployee(talent)acquisition,maturation,andretention(whichis
especiallyimportantforanorganizationwhere90percentofitsmanagement
hascomeupthroughtheranksofthestore).

Continuedouble-digitrevenuegrowth(up20.3percentin2012)byopening

newstores(opened183over100in2012).

Increasesamestoresalesgrowth(7.1percentgrowthin2012).

ImprovemarketingeffectivenessonbuildingtheChipotlebrandandengaging
withcustomersinwaysthatcreatestronger,deeperbonds.

Whileanyfourofthesebusinessinitiativesareripeforthebigdatastrategy
document,fortheremainderofthisexercise,we'llfocusonthe“increasesame
storesales”businessinitiativebecauseincreasingsalesofabusinessentityor
outletisrelevantacrossanumberofdifferentindustries(i.e.,hospitality,gaming,
banking,insurance,retail,highereducation,healthcareproviders).

<b>IdentifyKeyBusinessEntitiesandKeyDecisions</b>

</div>
<span class='text_page_counter'>(73)</span><div class='page_container' data-page=73>

<b>NOTE</b>

Itisaroundthesebusinessentitiesthatwearegoingtowanttocapturethe
behaviors,tendencies,patterns,trends,preferences,etc.attheindividual
entitylevel.Forexample,acreditcardcompanywouldwanttocaptureBill
Schmarzo'sspecifictravelandbuyingpatternsandtendenciesinorderto
betterdetectfraudandimprovemerchantmarketingoffers.

Figure3.4showsthetemplatethatwearegoingtousetosupportthebigdata
strategydocumentprocess.Wehavealreadycapturedourtargeted“increasesame
storesales”businessinitiative.

<b>Figure3.4</b>Chipotle's“increasesamestoresales”businessinitiative

Takeamomenttowritedownwhatyouthinkmightbethekeybusinessentitiesor

strategicnounsforthe“increasesamestoresales”businessinitiative:

HerearethreebusinessentitiesthatIcameupwith:
Stores

</div>
<span class='text_page_counter'>(74)</span><div class='page_container' data-page=74></div>
<span class='text_page_counter'>(75)</span><div class='page_container' data-page=75></div>
<span class='text_page_counter'>(76)</span><div class='page_container' data-page=76>

Businessentity:Localevents
Decision:

Decision:

Businessentity:Localcompetitors
Decision:

Decision:

<b>NOTE</b>

Someofthedecisionswillbeverysimilar.That'sgoodbecauseitallowsthe
organizationtoapproachthedecisionsfrommultipleperspectives.

Figure3.5showsiswhattheChipotlebigdatastrategydocumentlooksatthis
pointintheexercisewiththeadditionofsomeofthebusinessdecisions.

<b>Figure3.5</b>Chipotlekeybusinessentitiesanddecisions

<b>IdentifyFinancialDrivers(UseCases)</b>

</div>
<span class='text_page_counter'>(77)</span><div class='page_container' data-page=77>

<b>CROSS-REFERENCE</b>

Whileitishardtoactuallydothisgroupingprocessinabook,theuseof

facilitationtechniquestohelpbrainstormandgroupthesedecisionswillbe
<i>coveredintheFacilitationTechniquessectionofChapter13,“Powerof</i>
Envisioning.”

ForChipotle's“increasesamestoresales”businessinitiative,thefollowingare
likelyfinancialdriversorusecases:

Increasestoretraffic(acquirenewcustomers,increasefrequencyofrepeat
customers)

Increaseshoppingbagrevenueandmargins(cross-sellcomplementary
products,up-sell)

Increasenumberofcorporateevents(catering,repeatcateringevents)

Improvepromotionaleffectiveness(HalloweenBoo-ritto,Christmasgiftcards,
graduation,holiday,andspecialeventgiftcards)

Improvenewproductintroductioneffectiveness(seasonal,holiday)

Theentirebigdatastrategydocumentprocesshasbeendesignedtouncoverthese
usecases—toidentifythosefinancialdriversthatsupportourtargetedbusiness
initiative.Theusecasesandfinancialdriversarethepointofthebigdatastrategy
<i>documentwherewefocustheorganizationonthe“MakeMeMoreMoney”big</i>
dataopportunities.

</div>
<span class='text_page_counter'>(78)</span><div class='page_container' data-page=78></div>
<span class='text_page_counter'>(79)</span><div class='page_container' data-page=79>

<b>Figure3.6</b>CompletedChipotlebigdatastrategydocument

<b>IdentifyandPrioritizeDataSources</b>

Withtheusecasesandfinancialdriversidentified,wearenowreadytomoveinto
thedataandmetricsenvisioningprocess.Wewanttobrainstormdatasources
(regardlessofwhetherornotyoucurrentlyhaveaccesstothesedatasources)that
<i><b>mightyieldnewinsightsthatsupportthetargetedbusinessinitiative.Wewantto</b></i>
unleashthebusinessandITteams'creativethinkingtobrainstormdatasources
thatmightyieldnewcustomer,product,store,campaign,andoperationalinsights
thatcouldimprovetheeffectivenessofthedifferentusecases.

Forexample,Chipotledatasourcesthatwereidentifiedaspartoftheenvisioning
exercisescouldinclude:

PointofSalesTransactions
MarketBaskets

</div>
<span class='text_page_counter'>(80)</span><div class='page_container' data-page=80>

Weather

TrafficPatterns
Yelp

Zillow/Realtor.com

Twitter/Facebook/Instagram
Twellow/Twellowhood

ZipCodeDemographics
EventBrite

MaxPreps
MobileApp

Butnotalldatasourcesareofequalbusinessvalueorhaveequalimplementation
feasibility.Thedatasourcesneedtobeevaluatedinlightof

Thebusinessvaluethatdatasourcecouldprovideinsupportoftheindividual
usecase

Thefeasibility(orease)ofacquiring,cleaning,aligning,normalizing,
enriching,andanalyzingthosedatasources

Sowewanttoaddtwoprocesses(worksheets)tothebigdatastrategydocument
processthatwillevaluatethebusinessvalueandimplementationfeasibilityof
eachofthepotentialdatasources.

</div>
<span class='text_page_counter'>(81)</span><div class='page_container' data-page=81>

<b>Figure3.7</b>BusinessvalueofpotentialChipotledatasources

You'dwanttogothroughagroupbrainstormingprocesswiththebusiness

stakeholderstoassesstherelativevalueofeachdatasourcewithrespecttoeach
<i>usecase.Thebusinessusersownthebusinessvaluedeterminationbecausethey</i>
arebestpositionedtobeabletounderstandandquantifythebusinessvaluethat
eachdatasourcecouldprovidetotheusecases.

<b>NOTE</b>

IlikeusingHarveyBalls(in

boththedatavalueandthefeasibilityassessmentcharts.TheHarveyBalls
quicklyandeasilycommunicatetherelativevalueofeachdatasourcewith
respecttoeachusecase.

ReviewingthedatavalueassessmentchartinFigure3.7,youcanquicklyuncover
somekeyobservations,suchasthefollowing:

Detailedpoint-of-saledataisimportanttoalloftheusecases.

InsightsfromtheStoreDemographicsdataareimportanttofourofthefive
usecases.

MiningConsumerCommentshasasurprisingstrongimpactacrossfourofthe
fiveusecases.

</div>
<span class='text_page_counter'>(82)</span><div class='page_container' data-page=82>

promotionaleffectiveness”usecasesbuthaslittleimpactonthe“increase
shoppingbagrevenue,”“increasenumberofcorporateevents,”or“improve
newproductintroductioneffectiveness”usecases.

Next,youwanttounderstandtheimplementationfeasibilityforeachofthe
potentialdatasources.ThispartoftheexerciseisprimarilydrivenbytheIT
organizationsinceitisbestpositionedtounderstandtheimplementation

challengesandrisksassociatedwitheachofthedatasources,suchaseaseofdata
acquisition,cleanlinessofthedata,dataaccuracy,datagranularity,costof

acquiringthedata,organizationalskillsets,toolproficiencies,andotherrisk
<i><b>factors.TheimplementationfeasibilityassessmentchartforChipotle's</b></i>
“increasesamestoresales”businessinitiativelookslikeFigure3.8.

<b>Figure3.8</b>ImplementationfeasibilityofpotentialChipotledatasources

FromtheChipotleimplementationfeasibilityassessmentchartinFigure3.8,we
canquicklymakethefollowingobservations:

PointofSales,MarketBaskets,andStoreManagerDemographicsdatais
readilyavailableandeasytointegrate(likelyduetothemasterdata

managementanddatagovernanceeffortsnecessarytoloadthisdataintoa
datawarehouse).

ConsumerCommentsdata,whichwasveryvaluableinthebusinessvalue
assessment,hasseveralimplementationrisks.Lackoforganizational

</div>
<span class='text_page_counter'>(83)</span><div class='page_container' data-page=83>

SocialMediadata,whichwasratedaboutmid-valueinthevalueassessment
exercise,alsolookstobearealchallenge.Manyofthesamecleanliness,
accuracy,andgranularityissuesexist,withtheaddedissuethatthisisdata
thatwillneedtobe“acquired”throughsomemeans.Probablynotthefirstdata
sourceyouwanttodealwithinthisusecase.

</div>
<span class='text_page_counter'>(84)</span><div class='page_container' data-page=84>

<b>IntroducingthePrioritizationMatrix</b>

Thefinalstepinthebigdatastrategydocumentprocessistotakethebusinessand
ITstakeholdersthroughausecaseprioritizationprocess.Whilewewillcoverthe
<i><b>prioritizationmatrixindetailinChapter13,Iwanttointroducetheconcept</b></i>
hereasthenaturalpointofconcludingthebigdatastrategydocumentprocess.
Aspartofthebigdatastrategydocument,wehavenowdonetheworktoidentify
theusecasesthatsupporttheorganization'skeybusinessinitiative,brainstormed
additionaldatasources,anddeterminedtheapplicabilityofthosedatasources
fromabusinessvalueandimplementationfeasibilityassessment.Wearenow
readytoprioritizetheusecasesbasedontheirrelativebusinessvalueand
implementationfeasibilityoverthenext9to12months(seeFigure3.9).

<b>Figure3.9</b>Chipotleprioritizationofusecases

<b>WARNING</b>

</div>
<span class='text_page_counter'>(85)</span><div class='page_container' data-page=85></div>
<span class='text_page_counter'>(86)</span><div class='page_container' data-page=86></div>
<span class='text_page_counter'>(87)</span><div class='page_container' data-page=87></div>
<span class='text_page_counter'>(88)</span><div class='page_container' data-page=88>

Preservestartingpitchingeffectivenessthroughouttheregularseasonand
playoffsbyoptimizingpitchcounts,pitcherrotations,pitcherrests,etc.
Improvebattingandsluggingproficiencybyoptimizingtrades,freeagent
signings,minorleaguepromotions,andcontractextensions

Increasein-game“smallball”runsscoredeffectivenessthroughtheoptimal
combinationofbatters,hitting,stealing,baserunning,andsacrificehitting
strategies

Accelerateminorleagueplayerdevelopmentthroughplayerstrengthand
conditioningtraining,gamesituations,andminorleagueassignments

Optimizein-gamepitchselectiondecisionsthroughimprovedunderstanding
ofbatterandpitchermatchups

Figure3.10showstheresultingbigdatastrategydocument.

<b>Figure3.10</b>SanFranciscoGiantsbigdatastrategydocument

Next,wewouldbrainstormthepotentialdatasourcestosupporttheusecases,
including:

<b>PersonnelPlayerHealth.Thisshouldincludepersonalhealthhistory</b>

(weight,health,BMI,injuries,therapy,medications),physicalperformance
metrics(60-footdashtime,longtossdistances,fastballvelocity),andworkout
history(benchpress,deadlift,crunchesandpushupsin60seconds,frequency

andrecencyofworkouts).

<b>StartingPitcherPerformance.Thisshouldincludeadetailedpitching</b>

</div>
<span class='text_page_counter'>(89)</span><div class='page_container' data-page=89></div>
<span class='text_page_counter'>(90)</span><div class='page_container' data-page=90>

<b>Summary</b>

Thischapterfocusedonthebigdatastrategydocumentandkeyrelatedtopics
including:

Introducedtheconceptofabusinessinitiativeandprovidedsomeexamplesof
wheretofindthesebusinessinitiatives

Introducedthebigdatastrategydocumentasaframeworkforhelping

organizationstoidentifytheusecasesthatguidewhereandhowtheycanstart
theirbigdatajourneys

Providedahands-onexampleofthebigdatastrategydocumentinactionusing
Chipotle,achainoforganicMexicanfoodrestaurants

Introducedworksheetstohelporganizationstodeterminethebusinessvalue
andimplementationfeasibilityofthedatasourcesthatcomeoutofthebig
datastrategydocumentprocess

IntroducedtheprioritizationmatrixasatooltohelpdrivebusinessandIT
alignmentaroundthetoppriorityusecasesovera9-to12-monthwindow
Hadsomefunbyapplyingthebigdatastrategydocumenttotheworldof
professionalbaseball

Thischapteroutlinedthebigdatastrategydocumentasaframeworktohelpan

organizationidentifywhereandhowtostartitsbigdatajourneyinsupportofthe
organization's9-to12-monthkeybusinessinitiatives.Thebigdatastrategy

documentisatooltoensurethatyourbigdatajourneyisvaluableandrelevant
fromabusinessperspective.

ToswingbackaroundtotheChipotlecasestudy,Figure3.11showssomeinitial
resultsofthecompany'ssuccesswithits“increasesamestorerevenues”business
initiative.(Formoreinformation,seethearticleat

</div>
<span class='text_page_counter'>(91)</span><div class='page_container' data-page=91>

<b>Figure3.11</b>Chipotle'ssamestoresalesresults

It'snicetoseethatourChipotleusecaseactuallyhasarealbusinessstorybehind
it.Butthenagain,everybigdatainitiativeshouldhavearealbusinessstory

</div>
<span class='text_page_counter'>(92)</span><div class='page_container' data-page=92>

<b>HomeworkAssignment</b>

Usethefollowingexercisestoapplythebigdatastrategydocumenttoyour
organization(oroneofyourfavoriteorganizations).

<b>Exercise#1:Startbyidentifyingyourorganization'skeybusinessinitiatives</b>

overthenext9to12months.

<b>Exercise#2:Selectoneofyourbusinessinitiatives,andthenbrainstormthe</b>

keybusinessentitiesorstrategicnounsthatimpactthatselectedbusiness
initiative.Asareminder,itisaroundtheindividualbusinessentitiesthatwe
wanttocapturethebehaviors,tendencies,patterns,trends,preferences,etc.at
theindividualbusinessentitylevel.

<b>Exercise#3:Next,brainstormthekeydecisionsthatneedtobemadeabout</b>

eachkeybusinessentitywithrespecttothetargetedbusinessinitiative.

<b>Exercise#4:Nextwewanttogroupthedecisionsintocommonusecases;</b>

thatis,clusterthosedecisionsthatseemsimilarintheirbusinessorfinancial
objectives.

<b>Exercise#5:Thenbrainstormthedifferentdatasourcesthatyoumightneed</b>

tosupportthoseusecases:

Identifypotentialinternalstructured(transactionaldatasources,

operationaldatasources)andunstructured(consumercomments,notes,
workorders,purchaserequests)datasources

Identifypotentialexternaldatasources(socialmedia,blogs,publicly
available,data.gov,websites,mobileapps)thatyoualsomightwantto
consider

<b>Exercise#6:Usethedataassessmentworksheetstodeterminetherelative</b>

businessvalueandimplementationfeasibilityofeachoftheidentifieddata
sourceswithrespecttothedifferentusecases.

<b>Exercise#7:Finally,usetheprioritizationmatrixtorankeachoftheuse</b>

</div>
<span class='text_page_counter'>(93)</span><div class='page_container' data-page=93>

<b>Notes</b>

</div>
<span class='text_page_counter'>(94)</span><div class='page_container' data-page=94></div>
<span class='text_page_counter'>(95)</span><div class='page_container' data-page=95>

<b>Chapter4</b>

<b>TheImportanceoftheUserExperience</b>

Theuserexperienceisoneofthesecretstobigdatasuccess,andoneofmy
favoritetopics.Iforganizationscannotdeliverinsightstoitsemployees,

managers,partners,andcustomersinawaythatisactionable,thenwhyeven
<i>bother.OneofthekeystosuccessintheBigDataMBAisto“beginwithanendin</i>
mind”withrespecttounderstandinghowtheanalyticresultsaregoingtobe

deliveredtofrontlineemployees,businessmanagers,channelpartners,and
<i>customersinawaythatisactionable.TheBigDataMBAseeksto“closethe</i>

</div>
<span class='text_page_counter'>(96)</span><div class='page_container' data-page=96></div>
<span class='text_page_counter'>(97)</span><div class='page_container' data-page=97>

<b>Chapter4Objectives</b>

Reviewanexampleofan“unintelligent”userexperience.

Highlighttheimportanceof“thinkingdifferently”withrespecttocreating
anactionabledashboardversusbuildingatraditionalBusinessIntelligence
dashboard.

Reviewasampleactionabledashboardtargetingfrontlinestoremanagers.
Reviewanothersampleactionabledashboard(financialadvisor

dashboard)​targetingbusiness-to-businesschannelpartners.

ThischapterwillchallengethetraditionalBusinessIntelligenceapproachesto

buildingdashboardsbyseekingtoleverageanalyticinsights(e.g.,

</div>
<span class='text_page_counter'>(98)</span><div class='page_container' data-page=98>

<b>TheUnintelligentUserExperience</b>

OneofmyfavoritesubjectsagainstwhichIlovetorailisthe“unintelligent”user
experience.Thisisaproblemcausedby,inmyhumbleopinion,thelackofeffort
byorganizationstounderstandtheirkeybusinessstakeholderswellenoughtobe
abletodeliveractionableinsightsinsupportoftheorganizations'keybusiness
initiatives.Andthisuserexperienceproblemisoftenonlyexacerbatedbybigdata.
Hereisareal-worldexampleofhowNOTtoleverageactionableanalyticsinyour
organization'sengagementswithyourcustomers.Thenameshavebeenchanged
toprotecttheguilty.

MydaughterAmeliagotthee-mail(seeFigure4.1)fromourcellphoneprovider
warningherthatshewasabouttoexceedhermonthlydatausagelimitof2GB.
Shewasveryupsetthatshewasabouttogooverherlimit,anditwouldstart
costingher(actually,me)anadditional$10.00perGBoverthelimit.(Note:The
“Monday,August13,2012”dateinthefigurewillplayanimportantroleinthis
story.)

<b>Figure4.1</b>Originalsubscribere-mail

IaskedAmeliawhatinformationshethoughtsheneededinordertomakea
<i><b>decisionaboutalteringherFacebook,Pandora,Vine,Snapchat,andInstagram</b></i>
usage(sincethosearethemaindatahogculpritsinhercase)sothatshewould
notexceedherdataplanlimits.Shethoughtforawhileandthensaidthatshe
thoughtsheneededthefollowinginformation:

Howmuchofherdataplandoesshehaveleftinthecurrentmonth?
Whendoeshernewmonthorbillingperiodstart?

</div>
<span class='text_page_counter'>(99)</span><div class='page_container' data-page=99></div>
<span class='text_page_counter'>(100)</span><div class='page_container' data-page=100>

experiencethatorganizationsshouldbetargeting.

Ourcellularprovidercouldhaveprovidedauserexperiencethathighlightedthe
informationandinsightsnecessarytohelpAmeliamakeadecisionaboutdata
usage.Theuserexperiencecouldhavelookedsomethinglikethee-mailmessage
showninFigure4.2.

<b>Figure4.2</b>Improvedsubscribere-mail

Thissamplee-mailhasalltheinformationthatAmelianeedstomakeadecision
aboutusagebehaviorsincluding:

Actualusagetodate(65percent)

Aforecastofusagebytheendoftheperiod(67percent)

Thedatewhenthedataplanwillreset(in1dayonAugust14)

</div>
<span class='text_page_counter'>(101)</span><div class='page_container' data-page=101>

<b>ConsumerCaseStudy:ImproveCustomer</b>

<b>Engagement</b>

Butlet'stakethiscasestudyonestepfurther.Let'ssaythatthereactuallywas
goingtobeaproblemwithAmelia'susageandherdataplan.Whatif82percentof
datausagehadbeenconsumedwith50percentofusageperiodremaining?How
dowemaketheuserexperienceandthecustomerengagementuseful,relevant,
andactionable?

Themock-upshowninFigure4.3offersonepotentialapproachbasedonthe
sameprinciplesdiscussedearlier:provideenoughinformationtohelpAmelia

changeherusagebehaviors.However,FutureTelcocouldalsotaketheuser
experienceandcustomerengagementonestepfurtherandofferhersome
recommendationstoavoidthedataplanoverage.

Forexample,FutureTelcocouldofferprescriptiveadviceabouthowtoreduce
dataconsumptionsuchas:

Transitioningtoappsthataremoredatausageefficient(i.e.,transitioning
fromPandoratoRdiooriHeartRadioforstreamingradio,assumingthatRdio
andiHeartRadioaremoreefficientintheirusageofthedatabandwidth)

Turningoffappsinthebackgroundthatareunnecessarilyconsumingdata
suchasmappingapps(likeAppleMaporWaze)orappsthatareusingGPS
tracking

FutureTelcocouldevenofferAmeliaoptionstoavoidpayinganoveragepenalty
(seeFigure4.3)suchas:

Purchasea1-monthdatausageupgradefor$2.00(whichischeaperthanthe
$10overagepenalty)

</div>
<span class='text_page_counter'>(102)</span><div class='page_container' data-page=102>

<b>Figure4.3</b>Actionablesubscribere-mail

Butwait,thereisevenmorethatFutureTelcocoulddotoimprovethecustomer
experience.FutureTelcocouldanalyzeAmelia'sappusagetendenciesand

recommendnewappsbasedonotherappsthatuserslikeAmeliause,similarto
whatAmazonandNetflixdo(seeFigure4.4).

<b>Figure4.4</b>Apprecommendations

Thislevelofcustomerintimacycanopenupallsortsofnewmonetization
opportunitiessuchas:

</div>
<span class='text_page_counter'>(103)</span><div class='page_container' data-page=103>

Helpappdeveloperstobemoresuccessfulwhilecollectingreferralfees,co-marketingfees,andothermonetizationideasthatalignwiththeapp
developers'businessobjectives

Cellularprovidersarenotaloneinmissingopportunitiestoleveragecustomer
insightsinordertoprovideamorerelevant,moremeaningfulcustomer

experience.Manyorganizationsaresittingongoldminesofinsightsabouttheir
customers'buyingandusagepatterns,tendencies,propensities,andareasof

</div>
<span class='text_page_counter'>(104)</span><div class='page_container' data-page=104>

<b>BusinessCaseStudy:EnableFrontlineEmployees</b>

Ihadtheopportunitytorunavisionworkshopforagroceryretailer.Thegoalof
thesessionwastoidentifyhowthegrocerychaincouldleveragebigdataand
advancedanalyticstodeliveractionableinsights(orrecommendations)tostore
managersinordertohelpthemimprovestoreperformance.

Bigdatacantransformthebusinessbyenablingacompletelynewuserexperience
(UEX)builtaroundinsightandrecommendationsversusjusttraditionalBusiness
Intelligencechartsandtables.Retailers,likemostorganizations,canleverage
detailed,historicaltransactionaldata—coupledwithnewsourcesof“right-time”
datalikelocalcompetitors'promotions(e.g.,“bestfooddays,”whichistheday
whengrocerystoresposttheirweeklypromotions),weather,andevents—to

uncovernewinsightabouttheircustomers,products,merchandising,competitors
andoperations.Bigdataprovidesorganizationstheabilityto(1)rapidlyingest
thesenewsourcesofcustomer,product,andoperationaldataandthen(2)
leveragedatasciencetoyieldreal-time,actionableinsights.

Let'swalkthroughanexampleofintegratingbigdatawithatraditionalBI
dashboardtocreateamoreactionableuserexperiencethatempowersfrontline
employeesandmanagers.

<b>StoreManagerDashboard</b>

</div>
<span class='text_page_counter'>(105)</span><div class='page_container' data-page=105>

<b>Figure4.5</b>TraditionalBusinessIntelligencedashboard

ThechallengewiththesetraditionalBIdashboardsisthatunlessyouarean
analyst,it'snotclearwhatactiontheuserissupposedtotake.Arrowsup,
sideways,anddown…Icanseemyperformance,butthedashboarddoesn't
provideanyinsightstotellthestoremanagerwhatactionstotake.

Theotherchallengeisthatthestoremanager(likemostfrontlineemployeesand
managers)likelydoesnothaveaBIorananalyticsbackground(likelyworkedhis
wayuptheranksinthegrocerystore).Asaresult,UEXandtheactionable

insightsandrecommendationsarecriticalbecausethestoremanagerdoesnot
knowhowtodrillintotheBIreportsanddashboardstouncoverinsightsbasedon
therawdata.

</div>
<span class='text_page_counter'>(106)</span><div class='page_container' data-page=106>

<b>Figure4.6</b>Actionablestoremanagerdashboard

InFigure4.6,SectionAshowsspecificproduct,promotion,placement,and
pricingrecommendationsbasedonthelayoutofaspecificstore.SectionB

providesspecificrecommendationsconcerningpricing,merchandising,inventory,
staffing,promotions,etc.forthestoremanager.

EachrecommendationinSectionBispresentedwithAccept[+]orReject[-]
options.Ifthestoremanageracceptstherecommendationbyselecting[+],that
recommendationisexecuted(e.g.,raiseprices,addpromotion,addinventory,
etc.).However,ifthestoremanagerrejectstherecommendation,thenthe

actionabledashboardcapturesthereasonfortherejectionsothatthesupporting
analyticmodelscanbeconstantlyfine-tuned(seeFigure4.7).

Finally,thestoremanagercanselecttheMoreoptioninSectionBandmodifythe
recommendationbasedonhisownexperience.Allowingthestoremanagerto
modifytherecommendationsbasedonhispersonalexperiencesallowsthe

underlyinganalyticmodelstoconstantlylearnwhatworksandwhatdoesn'twork
andbuildonthebestpracticesandlearningsfromtheorganization'smost

</div>
<span class='text_page_counter'>(107)</span><div class='page_container' data-page=107>

<b>Figure4.7</b>Storemanageraccept/rejectrecommendations

<b>SampleUseCase:CompetitiveAnalysis</b>

Oneusecaseforthestoremanagerdashboardenablesthestoremanagerto
monitorlocalcompetitiveactivityandpromotions.Thegroceryindustryisvery
locallycompetitive.Competitors,forthemostpart,arewithinjustafewmilesor
evenblocksofeachother.Inthiscompetitiveanalysisusecase,thedashboard
providesamapofthelocalgroceryandbeveragecompetitors(seeSectionCof
Figure4.7).Hoveringoveranyparticularcompetitoronthemapimmediately
bringsupitscurrentmarketingflyer.Thestoremanager(orhisbusinessanalyst)
canbrowsethrougheachofthecompetitors'flyersandmakecustomstore

</div>
<span class='text_page_counter'>(108)</span><div class='page_container' data-page=108>

<b>Figure4.8</b>Competitiveanalysisusecase

Liketheotherrecommendations,thestoremanager'scustomrecommendations
willbemonitoredforeffectivenesssothattheanalyticmodelscanbeconstantly
updatedandrefined.

<b>AdditionalUseCases</b>

</div>
<span class='text_page_counter'>(109)</span><div class='page_container' data-page=109>

<b>Figure4.9</b>Localeventsusecase

Anotherusecaseistointegratethelocalweatherforecastintothestoremanager
dashboard.Thestoremanagercananalyzethelocalweatherforecastsandmake
adjustmentsforinventory,merchandising,andpromotionsbasedonwhetherthe
weatherwillbewarmerorcolderthanexpected(seeFigure4.10).Thedashboard
canautomaticallyanalyzesimilarweatherconditionsandpredicttheimpacton
storetrafficandproductcategorysalesanddeliverrelevantrecommendationsto
thestoremanager.

<b>Figure4.10</b>Localweatherusecase

</div>
<span class='text_page_counter'>(110)</span><div class='page_container' data-page=110></div>
<span class='text_page_counter'>(111)</span><div class='page_container' data-page=111></div>
<span class='text_page_counter'>(112)</span><div class='page_container' data-page=112>

Adjustinvestmentstrategies(short-term,long-term)
Reallocatefinancialportfolio

Changeinvestmentvehicles(stocks,bonds,mutualfunds,etc.)

<b>Figure4.11</b>Financialadvisordashboard

Thegoalofthefinancialadvisordashboardistouncoverinsightsabouttheclient's
investmentperformanceandprovideclient-specificrecommendationsthathelp
theseclientsreachtheirfinancialgoals.Togenerateactionable,accurate

recommendations,we'regoingtoneedtoknowasmuchabouttheclientas

possible,including:

Currentandhistoricalpersonalbackgroundinformation(e.g.,maritalstatus,
spouse'sfinancialandemploymentsituation,numberandageofchildren,
outstandingmortgageonhome(s)andanysecondaryrealestateinvestments)
Currentfinancialinvestmentsandotherassets(e.g.,stocks,bonds,mutual
funds,IRAs,401-Ks,REITs)

Currentandhistoricalincome(andexpenditures,ifpossible)
Financialgoalswithspecifictimelines

Weneedtoensurethatthefinancialadvisordashboardprovidesenoughvalueto
boththefinancialadvisorandtheadvisor'sclientsinordertoincenttheclientsto
shareasmuchofthisdataaspossible.

<b>InformationalSectionsofFinancialAdvisorDashboard</b>

</div>
<span class='text_page_counter'>(113)</span><div class='page_container' data-page=113>

<b>ClientPersonalInformation:Thefirstpartofthedashboardpresents</b>

relevantclientpersonalandfinancialinformation.FSIwantstogatherasmuch
personalinformationasisrelevantwhentheclientfirstopenshisaccounts.But
aftertheclientopenshisaccount,thereneedstobeaconcertedefforttokeepthe
dataupdatedandcapturenewlifestyle,lifestage,employment,andfamily

information.Muchofthatclientdatacanbecapturedviadiscussionsand

interactionsthatthefinancialadvisorishavingwiththeclient(e.g.,informational
calls,e-maildialogues,officevisits,annualreviews).Whilethisinformationis
goldtoFSI,muchofthisdatanevergetspastthefinancialadvisors'personal
contactmanagementande-mailsystems.FSImustprovidecompellingreasonsto

persuadethefinancialadvisorsandclientstosharemoreofthisdatawithFSI(see
Figure4.12).

Someleading-edgeorganizationsareprovidingincentives(e.g.,discounts,

promotions,contests,rewards)forclientstosharetheirsocialmediainteractions.
Obviously,accesstotheclient'scurrentsituationandplansaspostedonsocial
mediasitesisgoldwhenitcanbeminedtouncoveractivitiesthatmightaffecthis
financialneeds(e.g.,vacations,buyinganewcar,upcomingweddingplans,

promotions,jobchanges,childrenchangingschools).

<b>ClientFinancialStatus:Thenextsectionofthedashboardprovidesan</b>

overviewoftheclient'scurrentfinancialstatus.Again,themoredatathatcanbe
gatheredabouttheclient'sfinancialsituation(e.g.,investments,home,spending,
debt),themoreaccurateandprescriptivetheanalyticmodelswillbe(seeFigure
4.13).

</div>
<span class='text_page_counter'>(114)</span><div class='page_container' data-page=114>

<b>Figure4.13</b>Clientfinancialinformation

Inthisexample,wehavedetailsonalltheclient'sfinancialinvestmentswithFSI.
However,theclientmight(andlikelydoes)havefinancialinvestmentswithother
firmscourtesyofhisemployer's401kprograms,wholelifeinsurancepolicies,and
otherstocks,bonds,andfunds.Andthatdoesn'tevenconsidersubstantial

investmentsinnonfinancialinstrumentslikehisprimaryresidence,vacation
home,antiques,andcollectibles.

Incentingclientstosharetheirentirefinancialportfolioiscomplicatedbyhow

harditisforaclienttopullallthatinformationtogetherinoneplace.However,
Mint.comhasfiguredouthowtoaggregatefinancialspendingfromcreditcards
andbankchecks.Theinclusionoftheclient'sexpendituredatacouldbeinvaluable
inbuildingaclientprofileanddevelopingspecific,actionablefinancial

recommendations.

<b>ClientFinancialGoals:Thefinalinformationalsectionofthedashboard</b>

containstheclient'sfinancialgoals.Therearelikelyonlyasmallnumberofgoals,
andtheyprobablydon'tchangethatoften.However,itisdifficulttodevelop
meaningfulclientfinancialrecommendationswithoutup-to-dateclientfinancial
goals.Fromadatacollectionperspective,thisisprobablytheeasiestdatato

</div>
<span class='text_page_counter'>(115)</span><div class='page_container' data-page=115>

<b>Figure4.14</b>Clientfinancialgoals

However,let'ssaythattheclienteitherwon'tsharehisfinancialgoalsorhasn't
eventhoughtthroughwhathisfinancialgoalsneedtobe.Thisiscommonwhen
dealingwithretirementplanning,sincemanyclientsaren'tclearorrealisticabout
theirretirementgoals.Inthesesituations,FSIcouldleveragetheinformationthat
ithasabout“similar”clientstomakeretirementgoalrecommendations.IfFSIhas
theclient'scurrentfinancialinvestmentsandcurrentsalary,FSIcouldmakea
prettyintelligentguessastotheclient'sretirementgoals.

<b>RecommendationsSectionofFinancialAdvisorDashboard</b>

Nowlet'sgetintothemeatofthefinancialadvisordashboard.Theclient

informationsectionsofthedashboardweremeanttoprovideaneasyandefficient
waytocapturetheclient'skeylifestyle,demographic,andfinancialdata,aswellas

hisfinancialgoals.Nowwecancreatepredictivemodelstopredictthelikely

resultsofdifferentfinancialoptionsandactions,andthencreateprescriptive
modelsinordertodeliverclient-specificrecommendationsthathelptheclientto
reachhisfinancialgoals.Thisfinancialadvisordashboardcoversfourdifferent
areasfordeliveringclient-specificfinancialrecommendations:

Financialcontributions
Spendinganalysis

Assetallocation

Otherfinancialinvestments

<i><b>FinancialContributionsRecommendations</b></i>

Thefirstsetofrecommendationsisfocusedonhelpingtheclientoptimize

financialcontributions(seeFigure4.15).Thetypesofclientdecisionsthatcould
bemodeledinclude:

</div>
<span class='text_page_counter'>(116)</span><div class='page_container' data-page=116>

Onetimepaymentstojump-startlaggingfinancialgoals

Reallocatemonthlyorperiodicpaymentsagainstdifferentfinancialgoals
Changeretirement,newcar,andnewhometargetdates

<b>Figure4.15</b>Financialcontributionsrecommendations

Wecouldemploydatasciencetoanalyzetheclient'sdetailedfinancialdata,

comparethatdatawithbenchmarksacrosssimilarclientsanddevelopclient-specificanalyticprofiles.Thefinancialadvisordashboardcouldprovidea“what
if”capabilitythatallowsthefinancialadvisortoworkwiththeclienttotestout
differentscenarios(e.g.,changestoinvestmentamounts,changestofinancialgoal
targetdates).

<i><b>SpendingAnalysisRecommendations</b></i>

Thesecondsetofrecommendationsisfocusedonhelpingtheclientoptimize
spendinghabits.Thisiswhereaccesstotheclient'screditcardandbanking
statements(maybeviaMint.comand/orhischeckingaccounts)couldyield
valuableinsightstohelptheclientminimizecashoutflowandincreasefinancial
investments(seeFigure4.16).Thetypesofspendingdecisionsthatwouldneedto
bemodeledinclude:

Consolidatingexpendituresofsimilarproductsandservices

Flaggingexpendituresthatareabnormallyhighgiventheclient'sfamily
situation,homelocation,etc.

Integratingcustomerloyaltyprograminformationtofindretailerswhocan
providebestpricesonfoodandhouseholdstaples

Increasinginsurancedeductiblestolowerpremiums

</div>
<span class='text_page_counter'>(117)</span><div class='page_container' data-page=117>

<b>Figure4.16</b>Spendanalysisandrecommendations

Therearelotsofopportunitiestoleverageexternaldatasourcesandbestpractices
acrosstheFSIclientbasetofindbetterdealsinanattempttoreducetheclient's
discretionaryspending.Thereareseveralretail,insurance,travel,hospitality,
entertainment,cellphone,andotherwebsitesfromwhichdatacouldbegathered.

Thisdatacouldbeusedtocreaterecommendationstoreducetheclient'sspending
andoptimizetheclient'smonthlybudget,withthesavingsbeingusedtoincrease
financialcontributionsagainsttheclient'sfinancialgoals.

<i><b>AssetAllocationRecommendations</b></i>

Thethirdsetofrecommendationsisfocusedonhelpingclientsoptimizetheir
assetallocationinlightoftheirfinancialgoals.Byleveragingbestpracticesacross
otherclients,portfolios,andinvestmentinstruments,prescriptiveanalyticscanbe
developedtomakespecificassetallocationrecommendationsthatsupportasset
allocationdecisionssuchas(seeFigure4.17):

Whichstocksandbondstosellorbuyagainstspecificfinancialgoalportfolios
Portfolioallocationdecisionsthatproperlybalancetherisk-returnratioofthe
client'sportfolioinlightofrisktoleranceandfinancialgoals

Otherfinancialinstrumentsthatcanacceleratetheclient'sprogressagainst
financialgoalsorreduceriskforthoseshort-termfinancialgoals

<b>Figure4.17</b>Assetallocationrecommendations

</div>
<span class='text_page_counter'>(118)</span><div class='page_container' data-page=118>

client'sdesiredrisklevel.Tofurtherprotecttheclient'sinvestmentassets,an
aggregatedviewofthemarketplacecouldyieldmoretimelyinsightsintostocks
andbondsthataresuddenlyhotorcold.Thisisalsoanareawherereal-time

analyticscanbeleveragedtoensurethatnosuddenmarketmovementsexposethe
clienttounnecessaryassetallocationrisks.Thedashboardcouldalsosupportan
interactive“whatif”collaborationdirectlywiththeclienttogleanevenmoredata
andinsightsabouttheclient'sinvestmentpreferencesandtoleranceforrisk.

<i><b>OtherInvestmentRecommendations</b></i>

Thefourthsetofrecommendationsisfocusedonotherassetsthatclientsneedto
consideraspartoftheiroverallfinancialstrategy.Realestate(theclient'shome
andanyvacationhomes)isprobablythemostobvious.Thisisanareawhere
recommendationsaboutotherinvestmentoptionscanbedeliveredtohelp
supportclientdecisionsregarding(seeFigure4.18):

Identifyingtheidealamountofinsuranceneededgivenhomevaluation
changes

HomeimprovementprojectsthatyieldthebestROIforparticularhousetypes,
budgets,andlocationsovertime

Identifyingtherighttimetobuyorsellahome,andevenmaking
recommendationsastowhatpricetobidforhomesinselectareas
Bestareastolookforsecondaryand/orvacationhomeinvestments
Mostcost-effectivelocationstoliveinafterretirement

<b>Figure4.18</b>Otherinvestmentrecommendations

Thereisabevyofexternaldatasourcesthatcanbeleveragedtohelpfacilitate
analyticsinthisarea.Forexample,ZillowandRealtor.comproviderealestate
valuationsandmonthlychangesinrealestatevaluationsthatcouldbe

</div>
<span class='text_page_counter'>(119)</span><div class='page_container' data-page=119>

<b>Summary</b>

Bigdatacanpoweramorerelevantandmoreactionableuserexperience.Instead
ofoverwhelmingbusinessuserswithanendlessarrayofcharts,reports,and
dashboardsandforcingusersto“sliceanddice”theirwaytoinsights,wecan

insteadleveragethewealthofavailablestructuredandunstructureddatasources,
inreal-time,coupledwithdatasciencetouncovercustomer,product,and

operationalinsightsburiedinthedata.Wecanleveragethoseinsightstocreate
frontlineemployee,manager,andcustomerrecommendationsandthenmeasure
theeffectivenessofthoserecommendationssothatwearecontinuouslyrefining
ouranalyticmodels.

</div>
<span class='text_page_counter'>(120)</span><div class='page_container' data-page=120>

<b>HomeworkAssignment</b>

Usethefollowingexercisetoapplywhatyoulearnedinthischapter.

<b>Exercise#1:Selectoneofyourorganization'soutward-facingdashboards,</b>

websites,ormobileapps.Ifnotsomethingfromyourorganization,thenselect
awebsiteordashboardthatyouuseregularly.Thatmightincludesomething
fromyourbank,creditcardprovider,cellularprovider,orutilitycompany.
Grabafewscreencapturesofthedashboardorwebsite.

<b>Exercise#2:Thinkthroughhowyouastheuserusethisdashboard,website,</b>

ormobiletomakedecisions.Writedownthosedecisionsthatyoutrytomake
fromthewebsite.Forexample,fromyourutility,youmightwanttomake
decisionsaboutenergyandwaterconsumption,yourwaste/garbageplan,and
maybeevenwhichofthedifferentappliancerebatesyoumightwantto

consider.

<b>Exercise#3:Next,addarecommendationspanelthathassuggestionsfor</b>

eachofthedecisionsthatyoucapturedinStep2.Forourutilityexample,one
recommendationmightbe“Onlywater3daysaweekfrom6:00a.m.to7:00
a.m.tosaveapproximately$12.50permonthonyourmonthlywaterbill.”Or
anotherrecommendationmightbe“Replaceyourexistingdryerwithamore
efficientmodelliketheSamsungDV457tosave$21.75onyourmonthly
energybill.”

<b>Exercise#4:Finally,identifypotentialexternaldatasourcesthatmight</b>

</div>
<span class='text_page_counter'>(121)</span><div class='page_container' data-page=121></div>
<span class='text_page_counter'>(122)</span><div class='page_container' data-page=122>

<b>PartII</b>

<b>DataScience</b>

Thesethreechaptersintroducedatascienceasakeybusinessdisciplinethathelps
organizations“crosstheanalyticschasm”fromtheBusinessMonitoringto

BusinessInsightsandBusinessOptimizationphases.Thesechapterswill

introducetheconceptofdatascienceandthenbroadenthediscussiontocover
whatdatasciencetechniquestouseinwhichbusinessscenarios.

<b>InThisPart</b>

Chapter5:DifferencesBetweenBusinessIntelligenceandDataScience
Chapter6:DataScience101

</div>
<span class='text_page_counter'>(123)</span><div class='page_container' data-page=123></div>
<span class='text_page_counter'>(124)</span><div class='page_container' data-page=124>

<b>Chapter5</b>

<b>DifferencesBetweenBusinessIntelligenceandData</b>

<b>Science</b>

IwashiredbyalargeInternetportalcompanyin2007toheadupeffortsto
developitsadvertiseranalytics.Theobjectiveoftheadvertiseranalyticsproject
wastohelptheInternetportalcompany'sadvertisersandagenciesoptimizetheir
advertisingspendacrosstheInternetportal'sadnetwork.Theinternalcodename
fortheprojectwas“LookingGlass”becausewewantedtotaketheadvertisersand
agenciesthroughan“AliceinWonderland”typeofexperienceinhowwedelivered
actionableinsightstohelpourkeybusinessstakeholders—MediaPlanners&

BuyersandCampaignManagers—successfullyoptimizetheiradvertisingspendon
theInternetportal'sadnetwork.Butinmanyways,itwasmethatwentthrough
thelookingglass.

Severalmonthslater(August2008),IhadtheopportunitytokeynoteatTheData
WarehouseInstitute(TDWI)conferenceinSanDiego.ItaughtaclassatTDWIon
howtobuildanalyticapplications,soIwasbothfamiliarwithandabigfanofthe
TDWIconferences(andstillam).However,inmykeynote,Itoldtheaudiencethat
everythingthatIhadtaughtthemabouthowtobuildanalyticapplicationswas
wrong(seeFigure5.1).

<b>Figure5.1</b>SchmarzoTDWIkeynote,August2008

Likewithmyownpersonalexperience,manyorganizationsandindividualsare
confusedbythedifferencesintroducedbybigdata,especiallythedifferences
betweenBusinessIntelligence(BI)anddatascience.BigdataisnotbigBI.Big
dataisakeyenablerofanewdisciplinecalleddatasciencethatseekstoleverage
newsourcesofstructuredandunstructureddata,coupledwithpredictiveand
prescriptiveanalytics,touncovernewvariablesandmetricsthatarebetter

</div>
<span class='text_page_counter'>(125)</span><div class='page_container' data-page=125>

ThischapterdiscussesthedifferencesbetweenBIanddatascience:

Thequestionsaredifferent.

Theanalyticcharacteristicsaredifferent.

Theanalyticengagementprocessesaredifferent.
Thedatamodelsaredifferent.

Thebusinessviewisdifferent.

</div>
<span class='text_page_counter'>(126)</span><div class='page_container' data-page=126>

<b>WhatIsDataScience?</b>

Datascienceisacomplicatednewdisciplinethatrequiresadvancedskillsand
competenciesinareassuchasstatistics,computerscience,datamining,

mathematics,andcomputerprogramming.Ashadbeenstatedcountlesstimes,
datascientistsarethebusiness“rockstars”ofthe21stcentury.

Althoughwhatdatascientistsdocanbequitecomplex,whattheyaretryingto
achieveisnot.Infact,Ifindthattheverybestintroductorybooktodatascienceis
<i>Moneyball:TheArtofWinninganUnfairGamebyMichaelLewis(W.W.Norton</i>
&Company,2004).ThebookisabouttheOaklandA'sGeneralManagerBilly
Beane'suseofsabermetricstohelpthesmall-marketOaklandA'sprofessional
baseballteamoutperformcompetitorswithsignificantlylargerbankrolls.The
bookyieldsthemostaccuratedescriptionofdatascience:

<i>Datascienceisaboutfindingnewvariablesandmetricsthatarebetter</i>
<i>predictorsofperformance.</i>

That'sit—nothingmore—andyes,datascienceisthatsimple.Butthepowerof
thatsimplestatementisgamechanging,ascanbeseeninFigure5.2andthe

successthatBillyBeaneandtheOaklandA'shaveachievedbymakingplayer
acquisitionsandin-gamedecisionsbasedonadifferent,morepredictivesetof
metrics.

<b>Figure5.2</b>OaklandA'sversusNewYorkYankeescostperwin

Thebookalsohasanothervaluablelesson:goodideascanbecopied.So

</div>
<span class='text_page_counter'>(127)</span><div class='page_container' data-page=127>

predictive“on-basepercentage”metric.

<b>BIVersusDataScience:TheQuestionsAreDifferent</b>

WhenclientsaskmetoexplainthedifferencebetweenaBusinessIntelligence
analystandadatascientist,Istartbyexplainingthatthetwodisciplineshave
differentobjectivesandseektoanswerdifferenttypesofquestions(seeFigure
5.3).

<b>Figure5.3</b>BusinessIntelligenceversusdatascience

<b>BIQuestions</b>

BIfocusesondescriptiveanalytics:thatis,the“Whathappened?”typesof
questions.Examplesinclude:

HowmanywidgetsdidIselllastmonth?

WhatweresalesbyzipcodeforChristmaslastyear?
HowmanyunitsofProductXwerereturnedlastmonth?

Whatwerecompanyrevenuesandprofitsforthepastquarter?

HowmanyemployeesdidIhirelastyear?

BIfocusesonreportingonthecurrentstateofthebusiness,orasisnow
commonlycalledBusinessPerformanceManagement(BPM).BIprovides
retrospectivereportstohelpbusinessuserstomonitorthecurrentstateofthe
businessandanswerquestionsabouthistoricalbusinessperformance.These
reportsandquestionsarecriticaltothebusiness,sometimesrequiredfor
regulatoryandcompliancereasons.

</div>
<span class='text_page_counter'>(128)</span><div class='page_container' data-page=128>

ofunder-andover-performance.Buteventheseanalyticsarefocusedon
monitoringwhathappenedtothebusiness.

<b>DataScienceQuestions</b>

Ontheotherhand,datascientistsareinsearchofvariablesandmetricsthatare
betterpredictorsofbusinessperformance.Consequently,datascientistsfocuson
predictiveanalytics(“Whatislikelytohappen?”)andprescriptiveanalytics

(“WhatshouldIdo?”)typesofquestions.Forexample:
PredictiveQuestions(Whatislikelytohappen?)

HowmanywidgetswillIsellnextmonth?

WhatwillsalesbyzipcodebeoverthisChristmasseason?
HowmanyunitsofProductXwillbereturnednextmonth?

Whatareprojectedcompanyrevenuesandprofitsfornextquarter?
HowmanyemployeeswillIneedtohirenextyear?

PrescriptiveQuestions(WhatshouldIdo?)

Order[5,000]ComponentZtosupportwidgetsalesfornextmonth.
Hire[Y]newsalesrepsbythesezipcodestohandleprojectedChristmas
sales.

Setaside[$125K]infinancialreservetocoverProductXreturns.

Sellthefollowingproductmixtoachievequarterlyrevenueandmargin
goals.

Increasehiringpipelineby35percenttoachievehiringgoals.

Toanswerthesepredictiveandprescriptivequestions,datascientistsbuild

</div>
<span class='text_page_counter'>(129)</span><div class='page_container' data-page=129>

<b>TheAnalystCharacteristicsAreDifferent</b>

AnotherareaofdifferencebetweenBIanddatascienceisintheattitudinal
characteristicsandworkapproachofthepeoplewhofillthoseroles(seeTable
5.1).

<b>Table5.1</b>BIAnalystVersusDataScientistCharacteristics

<b>Area</b> <b>BIAnalyst</b> <b>DataScientist</b>

Focus Reports,KPIs,trends Patterns,correlations,models

Process Static,comparative Exploratory,experimentation,visual
Datasources Pre-planned,addedslowly Onthefly,asneeded

Transform Upfront,carefullyplanned In-database,ondemand,enrichment

Dataquality Singleversionoftruth “Goodenough,”probabilities

Datamodel Schemaonload Schemaonquery

Analysis Retrospective,descriptive Predictive,prescriptive

</div>
<span class='text_page_counter'>(130)</span><div class='page_container' data-page=130>

<b>Figure5.4</b>CRISP:CrossIndustryStandardProcessforDataMining

Datasciencetakesaverysimilarapproach:establishabusinesshypothesisor
question;exploredifferentcombinationsofdataandanalyticstobuild,test,and
refinetheanalyticmodel;andwash,rinse,andrepeatuntilthemodelprovesthat
itcanprovidetherequired“analyticlift”whilereachingasatisfactorygoodnessof
fit.Finallytheanalyticsaredeployedoroperationalizedincludingpossibly

</div>
<span class='text_page_counter'>(131)</span><div class='page_container' data-page=131>

<b>TheAnalyticApproachesAreDifferent</b>

Unfortunately,theseexplanationsareinsufficienttoanswersatisfactorilythe
questionofwhat'sdifferentbetweenBusinessIntelligenceanddatascience.So
let'sexaminecloselythedifferentengagementapproaches(includinggoals,tools,
andtechniques)thattheBIanalystandthedatascientistusetodotheirjobs.

<b>BusinessIntelligenceAnalystEngagementProcess</b>

TheBIanalystengagementprocessisadisciplinethathasbeendocumented,
taughtandrefinedoverthreedecadesofbuildingdatawarehousesandBI

environments.Figure5.5providesahigh-levelviewoftheprocessthatatypicalBI
analystuseswhenengagingwiththebusinessuserstobuildouttheBIand

supportingdatawarehouseenvironments.

<b>Figure5.5</b>BusinessIntelligenceengagementprocess

<b>Step1:Pre-buildDataModel.Theprocessstartsbybuildingthe</b>

foundationaldatamodel.Whetheryouuseadatawarehouseordatamartor
hub-and-spokeapproach,whetheryouuseastar,snowflake,normalizedor
dimensionalschema,theBIanalystmustgothroughaformalrequirements
gatheringprocesswiththebusinessuserstoidentifyall(oratleastthevast
majorityof)thequestionsthatthebusinessuserswanttoanswer.Inthis
requirementsgatheringprocess,theBIanalystmustidentifythefirst-and
second-levelquestionsthebusinessuserswanttoaddressinordertobuilda
robustandextensibledatamodel.Forexample:

</div>
<span class='text_page_counter'>(132)</span><div class='page_container' data-page=132></div>
<span class='text_page_counter'>(133)</span><div class='page_container' data-page=133>

SQLrequest.TheBIanalystscanalsospecifygraphicalrenderingoptions(bar
charts,linecharts,piecharts)untiltheygettheexactreportand/orgraphic
thattheywant(seeFigure5.6).

<b>Figure5.6</b>TypicalBItoolgraphicoptions

TheBItoolsareverypowerfulandrelativelyeasytouseifthedatamodelis

configuredproperly.Bytheway,thisisagoodexampleofthepowerofschemaon
load.Thistraditionalschemaonloadapproachremovesmuchoftheunderlying
datacomplexityfromthebusinessuserswhocanthenusetheBItoolsgraphical
userinterfacetomoreeasilyqueryandexplorethedata(thinkself-serviceBI).
Insummary,theBIapproachreliesonapre-builtdatamodel(schemaonload),
whichenablesuserstoquicklyandeasilyquerythedata—aslongasthedatathat
theywanttoqueryisalreadydefinedandloadedintothedatawarehouse.Ifthe
dataisnotinthedatawarehouse,thenaddingdatatoanexisting​warehousecan

takemonthstomakehappen.Notonlydoesmodifyingthedatawarehouseto
includeanewdatasourcerequireasignificantamountoftime,buttheprocess
canbeverycostly,asdataschemashavetobeupdatedtoincludethenewdata
source,newETLprocesseshavetobeconstructedtotransformandnormalizethe
datatofitintotheupdateddataschemas,andexistingreportsanddashboards
mayhavetobeupdatedtoincludethenewdata.

<b>TheDataScientistEngagementProcess</b>

Thedatascienceprocessissignificantlydifferent.Infact,thereisverylittlefrom
theBIanalystengagementprocessthatcanbereusedinthedatascience

</div>
<span class='text_page_counter'>(134)</span><div class='page_container' data-page=134>

<b>Figure5.7</b>Datascientistengagementprocess

<b>Step1:DefineHypothesistoTest.Step1ofthedatascienceengagement</b>

processstartswiththedatascientistsidentifyingthepredictiontheywantto
makeorhypothesisthattheywanttotest.Thisisaresultofcollaboratingwith
thebusinesssubjectmatterexperttounderstandthekeysourcesofbusiness
differentiation(e.g.,howtheorganizationdeliversvalue)andthenconstruct
theassociatedhypothesesorpredictions.

<b>Step2:GatherData…andMoreData.Instep2ofthedatascience</b>

engagementprocess,thedatascientistgathersrelevantorpotentially

interestingdatafromamultitudeofsources—bothinternalandexternaltothe
organization—andpushesthatdataintothedatalakeoranalyticsandbox.The
datalakeisagreatfoundationalcapabilityforthisprocess,asthedata

scientistscanacquireandingestanydatatheywant(as-is),testthedataforits
valuegiventhehypothesisorprediction,andthendecidewhethertoinclude
thatdataintheanalyticmodel.Thisiswhereanenvisioningexercisecanadd
considerablevalueinfacilitatingthecollaborationbetweenthebusinessusers
<i>andthedatascientiststoidentifydatasourcesthatmayhelpimprove</i>

predictiveresults.

<b>Step3:BuildDataModel.Step3iswherethedatascientistsdefineand</b>

</div>
<span class='text_page_counter'>(135)</span><div class='page_container' data-page=135></div>
<span class='text_page_counter'>(136)</span><div class='page_container' data-page=136></div>
<span class='text_page_counter'>(137)</span><div class='page_container' data-page=137>

<b>TheDataModelsAreDifferent</b>

Thedatamodelsthatareusedinthedatawarehousetosupportanorganization's
BIeffortsaresignificantlydifferentfromthedatamodelsthedatascientistsprefer
touse.

<b>DataModelingforBI</b>

TheworldofBI(akaquery,reporting,dashboards)requiresadatamodeling
techniquethatallowsbusinessuserstocreatetheirownreportingandqueries.To
supportthisneed,RalphKimballpioneereddimensionalmodeling—orstar

schemas—whileatMetaphorComputersbackinthe1980s(seeFigure5.9).

<b>Figure5.9</b>Dimensionalmodel(starschema)

Thedimensionalmodelwasdesignedtoaccommodatetheanalysisneedsofthe
businessusers,withtwoimportantdesignconcepts:

<b>Facttables(populatedwithmetricsormeasures)correspondtotransactional</b>

</div>
<span class='text_page_counter'>(138)</span><div class='page_container' data-page=138>

<b>Dimensiontables(populatedwithattributesaboutthatdimension)</b>

representthe“nouns”ofthatparticulartransactionalsystemsuchasproducts,
markets,stores,employees,customers,anddifferentvariationsoftime.

Dimensionsaregroupsofhierarchiesanddescriptorsthatdescribethefacts.It
isthesedimensionalattributesthatenableanalyticexploration,attributessuch
assize,weight,location(street,city,state,zip),age,gender,tenure,etc.

Dimensionalmodelingisidealforbusinessusersbecauseitsupportstheirnatural
question-and-answerexplorationprocesses.DimensionalmodelingsupportsBI
conceptssuchasdrillacross(navigatingacrossdimensions)anddrillup/drill
down(navigatingupanddownthedimensionalhierarchiessuchastheproduct
dimensionhierarchyofproduct⇨brand⇨category).

Today,allBItoolsusedimensionalmodelingasthestandardwayforinteracting
withtheunderlyingdatawarehouse.

<b>DataModelingforDataScience</b>

<i>Intheworldofdatascience,Hadoopprovidesanopportunitytothinkdifferently</i>
abouthowwedodatamodeling.HadoopwasoriginallydesignedbyYahootodeal
withverylong,flatweblogs.Hadoopwasdesignedwithverylargedatablocks
(Hadoopdefaultblocksizeis64MBto128MBversusrelationaldatabaseblock
sizesthataretypically32Kborless).Tooptimizethisblocksizeadvantage,the
datascienceteamwantsverylong,flatrecordsandlong,flatdatamodels.1

</div>
<span class='text_page_counter'>(139)</span><div class='page_container' data-page=139>

<b>Figure5.10</b>UsingflatfilestoeliminateorreducejoinsonHadoop

AsanexampleinFigure5.10,insteadofthreedifferentstarschemaswith
conformedorshareddimensionstolinkthedifferentstarschemas,thedata
scienceteamwantsthreelong,flatfileswiththefollowingcustomerdata:

Customerdemographics(age,gender,currentandprevioushomeaddresses,
valueofcurrentandprevioushome,historyofmaritalstatus,kidsandtheir
agesandgenders,currentandpreviousincome,etc.)

Customerpurchasehistory(annualpurchasesincludingitemspurchased,
returns,pricespaid,discounts,coupons,location,dayofweek,timeofday,
weathercondition,temperatures)

</div>
<span class='text_page_counter'>(140)</span><div class='page_container' data-page=140></div>
<span class='text_page_counter'>(141)</span><div class='page_container' data-page=141>

visitastore,recencyofstorevisit,frequencyofstorevisitsinpast

week/month/quarter,howlongdoIstayatwhichstores(“passthru”or
“linger”),etc.

<b>Classifications.Nowwewanttocreatesome“classifications”aboutBill</b>

Schmarzo'slifethatmighthaveimpactonStarbucks'skeybusinessinitiatives
suchaslifestageclassification(longmarriage,kidincollege,kidathome,
weight/dietconscious,etc.),lifestyleclassification(heavytraveler,heavychai
teadrinker,lightexerciser,andsoon),orproductclassification(morning
coffee/oatmealconsumer,afternoonfrap/cookieconsumer,etc.).

<b>AssociationRules.Wemightalsowanttocapturesomepropensitiesabout</b>

Bill'susagepatternsthatwecanusetosupportStarbucks'skeybusiness
initiatives,includingpropensitytobuyoatmealwhenhebuyshisventichai
lattewhentravelinginthemorning,propensitytobuyacookie/pastrywhen

travelingintheafternoon,propensitytobuyproductinthechannel,etc.

<b>Scores.Wealsomaywanttocreatescorestosupportdecision-makingand</b>

processoptimization.Scoresthatwemightwanttocreate(again,depending
onStarbucks'skeybusinessinitiatives)couldincludeadvocacyscore(which
measuresmylikelihoodtorecommendStarbucksandmakepositivecomments
forStarbucksonsocialmedia),loyaltyscore(whichmeasuresmylikelihoodto
continuetovisitStarbucksstoresandbuyStarbucksproductsversus

competitors),productusagescore(whichisameasureofhowmuchStarbucks
productIconsume—andrevenueIgenerate—whenIvisitaStarbucksstore),
etc.

Aprofilecouldbemadeupofhundredsofmetricsandscoresthat—whenusedin
combinationagainstaspecificbusinessinitiativelikecustomerretention,

</div>
<span class='text_page_counter'>(142)</span><div class='page_container' data-page=142>

<b>Figure5.11</b>Samplecustomeranalyticprofile

Somemetricsandscoresaremoreimportantthanothers,dependingonthe

businessinitiativebeingaddressed.Forexample,afinancialservicesfirmfocused
oncustomeracquisition,disposableincome,retirementreadiness,lifestage,age,
educationlevel,andnumberoffamilymembersdatamaybethemostimportant
predictivemetrics.However,forthatsamefinancialservicesfirmfocusedon

customerretention,metricssuchasadvocacy,customersatisfaction,attritionrisk,
socialnetworkassociations,andselectsocialmediarelationshipsmaybethemost
importantpredictivemetrics.

</div>
<span class='text_page_counter'>(143)</span><div class='page_container' data-page=143>

<b>Figure5.12</b>Improvecustomerretentionexample

Theanalysisprocessworkslikethis:

<b>Step1:Establishahypothesisthatyouwanttotest.Inourcustomerretention</b>

example,ourtesthypothesisisthat“Premiumgoldcardmemberswithgreater
thanfivedayswithoutapurchaseormobileappengagementhave25to30
percenthigherprobabilityofchurnthansimilarcustomers.”

<b>Step2:Identifyandquantifythemostimportantmetricsorscorestopredicta</b>

certainbusinessoutcome.Inourexample,themetricsandscoresthatwe're
goingtousetotestourcustomerattritionhypothesisincludesCustomer

Tenure(inmonths),CustomerSatisfactionScore,AverageMonthlyPurchases,
andCustomerLoyaltyScore.Noticethatthemetricsdonothavethesame
weight(orconfidencelevel).Somemetricsandscoresaremoreimportantthan
othersinpredictingperformancegiventhetesthypothesis.

<b>Step3:Employthepredictivemetricstobuilddetailedprofilesforeach</b>

individualcustomerwithrespecttothehypothesistobetested.

<b>Step4:Compareanindividual'srecentactivitiesandcurrentstatewithhisor</b>

herprofileinordertoflagunusualbehaviorsandactionsthatmaybe
indicativeofacustomerretentionproblem.Inourcustomerretention

example,wemightwanttocreatea“CustomerAttrition”scorethatquantifies

thelikelihoodthatparticularcustomerisgoingtoleave,andthencreate

specificrecommendationsastowhatactionsor“nextbestoffers”canbe
deliveredtoretainthatcustomer.

<b>Step5:Continuetoseekoutnewdatasourcesandnewmetricsthatmaybe</b>

</div>
<span class='text_page_counter'>(144)</span><div class='page_container' data-page=144>

andscoresusingsensitivityanalysisandsimulationsliketheMonteCarlo
experiments.

<b>Step6:Integratetheanalyticinsights,scores,andrecommendationsintothe</b>

</div>
<span class='text_page_counter'>(145)</span><div class='page_container' data-page=145>

<b>Summary</b>

OrganizationsarerealizingthatdatascienceisverydifferentfromBIandthatone
doesnotreplacetheother.Bothcombinetoprovidethe“dynamicduo”of

analytics—onefocusedonmonitoringthecurrentstateofthebusinessandthe
othertryingtopredictwhatislikelytohappenandthenprescribewhatactionsto
take.

Bigdataisakeyenablerofanewdisciplinecalleddatascience.Datascienceseeks
toleveragenewsourcesofstructuredandunstructureddata,coupledwith

advancedpredictiveandprescriptiveanalytics,touncovernewvariablesand
metricsthatarebetterpredictorsofperformance.

Asdiscussedinthischapter,BIisdifferentfromdatascienceinthefollowing
ways:

Thequestionsaredifferent.

Theanalyticcharacteristicsaredifferent.

Theanalyticengagementprocessesaredifferent.
Thedatamodelsaredifferent.

Thebusinessviewisdifferent.

Thischapteralsointroducedtheveryimportantdatascienceconceptcalled
analyticprofiles.Organizationsarelearningthatmoreimportantthantryingto
createa360-degreeprofileofthecustomerisidentifyingandquantifyingthose
fewerbutmoreimportantmetricsthatarebetterpredictorsofbusinessor
customerperformancesuchasoptimizingkeybusinessprocesses,influencing
customerbehaviors,anduncoveringnewmonetizationopportunities.

</div>
<span class='text_page_counter'>(146)</span><div class='page_container' data-page=146>

<b>HomeworkAssignment</b>

Usethefollowingexercisestoapplywhatyoulearnedinthischapter.

<b>Exercise#1:DescribethekeydifferencesbetweenBIanddatascienceand</b>

whatthosedifferencesmeantoyourorganization.

<b>Exercise#2:Listsampledescriptive(Whathappened?),predictive(Whatis</b>

likelytohappen?),andprescriptive(WhatactionsshouldItake?)questions
thatarerelevanttothetargetedbusinessinitiativethatyouidentifiedin
Chapter2.

</div>
<span class='text_page_counter'>(147)</span><div class='page_container' data-page=147>

<b>Notes</b>

1_{ApacheHadoopisanopen-sourcesoftwareframeworkwritteninJavafor}

distributedstorageanddistributedprocessingofverylargedatasetson

computerclustersbuiltfromcommodityhardware.AllthemodulesinHadoop
aredesignedwithafundamentalassumptionthathardwarefailures(of

</div>
<span class='text_page_counter'>(148)</span><div class='page_container' data-page=148></div>
<span class='text_page_counter'>(149)</span><div class='page_container' data-page=149>

<b>Chapter6</b>

<b>DataScience101</b>

Therearemanyexcellentbooksandcoursesfocusedonteachingpeoplehowto
becomeadatascientist.Thosebooksandcoursesprovidedetailedmaterialand
exercisesthatteachthekeycapabilitiesofdatasciencesuchasstatisticalanalysis,
datamining,textmining,SQLprogramming,andothercomputing,mathematical,
andanalytictechniques.Thatisnotthepurposeofthischapter.

ThepurposeofChapter6istointroducesomedifferentanalyticalgorithmsthat
businessusersshouldbeawareofandtodiscusswhenitmightbemost

appropriatetousewhichtypesofalgorithms.Youdonotneedtobeadata

scientisttounderstandwhenandwhytoapplytheseanalyticalgorithms.Amore
detailedunderstandingofthesedifferentanalyticalgorithmswillhelpthe

</div>
<span class='text_page_counter'>(150)</span><div class='page_container' data-page=150></div>
<span class='text_page_counter'>(151)</span><div class='page_container' data-page=151>

startviralmarketingcampaigns.

Thischapterreviewsanumberofdifferentanalytictechniques.Youarenot

expectedtobecomeanexpertinthesedifferentanalyticalgorithms.However,the
moreyouunderstandwhattheseanalyticalgorithmscando,thebetterposition
youareintocollaboratewithyourdatascienceteamandsuggesttheartofthe
possibletoyourbusinessleadershipteam.

FundamentalexploratoryanalyticalgorithmsthatarecoveredinChapter6are:
Trendanalysis

Boxplots

Geography(spatial)analysis
Pairsplot

Timeseriesdecomposition

Moreadvancedanalyticalgorithmsthatarecoveredinthischapterare:
Clusteranalysis

Normalcurveequivalent(NCE)analysis
Associationanalysis

Graphanalysis
Textmining

Sentimentanalysis

Traversepatternanalysis

Decisiontreeclassifieranalysis

Cohortsanalysis

Throughoutthechapter,youwillcontemplatehowTheParkscouldleverageeach
ofthesedifferentanalytictechniques.

<b>NOTE</b>

ThroughoutthischapterIprovidelinkstositesthatcanhelpyouget

</div>
<span class='text_page_counter'>(152)</span><div class='page_container' data-page=152>

<b>NOTE</b>

</div>
<span class='text_page_counter'>(153)</span><div class='page_container' data-page=153>

<b>FundamentalExploratoryAnalytics</b>

Let'sstartbycoveringsomebasicstatisticalanalysisthatwaslikelycoveredin
yourfirststatisticscourse(yes,Irealizethatyouprobablysoldyourstatsbookthe
minutethestatsclasswasover).Trendanalysis,boxplots,geographicalanalysis,
pairsplot,andtimeseriesdecompositionareexamplesofexploratoryanalytic
algorithmsthatthedatascientistsusetogeta“feelforthedata.”These

exploratoryanalyticalgorithmshelpthedatascienceteamtobetterunderstand
thedatacontentandgainahigh-levelunderstandingofrelationshipsandpatterns
inthedata.

<b>TrendAnalysis</b>

Trendanalysisisafundamentalvisualizationtechniquetospotpatterns,trends,
relationships,andoutliersacrossatimeseriesofdata.Oneofthemostbasicyet
verypowerfulexploratoryanalytics,trendanalysis(applyingdifferentplotting
techniquesandgraphicvisualizations)canquicklyuncovercustomer,operational,
orproducttrendsandeventsthattendtohappentogetherorhappenatsome

periodofregularity(seeFigure6.1).

<b>Figure6.1</b>Basictrendanalysis

InFigure6.1,thedatascientistmanuallytestedanumberofdifferenttrending
optionsinordertoidentifythe“bestfit”trendline(inthisexample,using

MicrosoftExcel).Oncethedatascientistidentifiesthebesttrendingoption,the
datascientistcanautomatethegenerationofthetrendlinesusingR.

</div>
<span class='text_page_counter'>(154)</span><div class='page_container' data-page=154>

differentbusinessdimensions(e.g.,products,geographies,salesterritories,
markets)inordertoundercoverpatternsandtrendsatthenextlevelof

granularity.Thedatascientistcanthenwriteaprogramtojuxtaposethedetailed
trendlinesintothesamechartsothatitiseasiertospottrends,patterns,

relationships,andoutliersburiedinthegranulardata(seeFigure6.2).

<b>Figure6.2</b>Compoundtrendanalysis

Finally,trendanalysisyieldsmathematicalmodelsforeachofthetrendlines.
Thesemathematicalmodelscanbeusedtoquantifyreoccurringpatternsor

behaviorsinthedata.Themostinterestinginsightsfromthetrendlinescanthen
beflaggedforfurtherinvestigationbythedatascienceteam(seeFigure6.3).

</div>
<span class='text_page_counter'>(155)</span><div class='page_container' data-page=155>

<b>WARNING</b>

</div>
<span class='text_page_counter'>(156)</span><div class='page_container' data-page=156>

<b>TheParksRamifications</b>

TheParkscouldusetrendanalysistoidentifythevariables(e.g.,waittimes,
socialmediaposts,consumercomments)thatarehighlycorrelatedtothe
increaseordecreaseinguestsatisfactionforeachattraction,restaurant,retail
outlet,andentertainment.TheParkscouldleveragetheresultsfromthetrend
analysisto

1. Flagproblemareasandtakecorrectiveactionssuchasopeningmorelines,
promotinglessbusyattractions,movingkiosksthatareblockingtraffic
flow,andresituatingcharactersatdifferentpointsintheparks;

2. Identifythelocationandtypesoffutureattractions,restaurants,retail
outlets,andentertainment.

Formoreinformationabouthowtomakesimpleplotsandgraphs(line
charts,barcharts,histograms,dotcharts)inR,checkout

/>

<b>Boxplots</b>

Boxplotsareoneofthemoreinterestingandvisuallycreativeexploratoryanalytic
algorithms.Boxplotsquicklyvisualizevariationsinthebasedataandcanbeused
toidentifyoutliersinthedataworthyoffurtherinvestigation.Aboxplotisa

convenientwayofgraphicallydepictinggroupsofnumericaldatathroughtheir
quartiles.Boxplotsmayalsohavelinesextendingverticallyfromtheboxes

(whiskers)indicatingvariabilityoutsidetheupperandlowerquartiles,hencethe
<i>termsbox-and-whiskerplotandbox-and-whiskerdiagram(seeFigure6.4).</i>

</div>
<span class='text_page_counter'>(157)</span><div class='page_container' data-page=157></div>
<span class='text_page_counter'>(158)</span><div class='page_container' data-page=158>

<b>TheParksRamifications</b>

TheParkscanemployboxplotstodetermineitsmostloyalguestsforeachof
thepark'sattractions(e.g.,CanyonCopterRide,MonsterMansion,Space
Adventure,GhoulishGulch).TheParkscanusetheresultsoftheboxplot
analysistocreateguestcurrentandmaximumlifetimevaluescoresagainst
whichtoprioritizetowhomtorewardwithPriorityAccesspassesandother
couponsanddiscounts.

FormoreinformationaboutcreatingboxplotsinR,checkout
/>

<b>Geographical(Spatial)Analysis</b>

Geographicalorspatialanalysisincludestechniquesforanalyzinggeographical
activitiesandconditionsusingabusinessentity'stopological,geometric,or
geographicproperties.Forexample,geographicalanalysissupportsthe

integrationofzipcodeanddata.goveconomicdatawithaclient'sinternaldatato
provideinsightsaboutthesuccessoftheorganization'sgeographicalreachand
marketpenetration(seeFigure6.5).

<b>Figure6.5</b>Geographical(spatial)trendanalysis

IntheexampleinFigure6.5,geographicalanalysisiscombinedwithtrend

</div>
<span class='text_page_counter'>(159)</span><div class='page_container' data-page=159>

<b>TheParksRamifications</b>

TheParkscanconductgeographicaltrendanalysistospotanychanges(at
boththezip+4andhouseholdlevels)inthegeo-demographicsofguestsover
timeandbyseasonalityandholidays.TheParkscanusetheresultsofthis
geographicalplusseasonalityanalysistocreategeo-specificcampaignsand

promotionswiththeobjectiveofincreasingattendancefromunder-penetratedgeographicalareasbydayofweek,holidays,andseasonality.

<b>PairsPlot</b>

Pairsplotanalysismaybemyfavoriteanalyticsalgorithm.Pairsplotanalysis
allowsthedatascientisttospotpotentialcorrelationsusingpairwisecomparisons
acrossmultiplevariables.Pairsplotanalysisprovidesadeepviewintothe

differentvariablesthatmaybecorrelatedandcanformthebasisforguidingthe
datascienceteamintheidentificationofkeyvariablesormetricstoincludeinthe
developmentofpredictivemodels(seeFigure6.6).

<b>Figure6.6</b>Pairsplotanalysis

</div>
<span class='text_page_counter'>(160)</span><div class='page_container' data-page=160>

<b>TheParksRamifications</b>

TheParkscanleveragepairsplotanalysistocompareamultitudeofvariables
toidentifythosevariablesthatdrivegueststoparticularattractions,

entertainment,retailoutlets,andrestaurants.TheParkscanusetheresultsof
theanalysistodrivein-parkpromotionaldecisionsandoffersthatdirect

gueststounder-utilizedattractions,entertainment,retailoutlets,and
restaurants.

AdditionalpairedplotoptionsinR(e.g.,pairs,splom,plotmatrix,ggcorplot,
panelcor)canbefoundat

/>

<b>TimeSeriesDecomposition</b>

Timeseriesdecompositionexpandsonthebasictrendanalysisbydecomposing
thetraditionaltrendanalysisintothreeunderlyingcomponentsthatcanprovide
valuablecustomer,product,oroperationalperformanceinsights.Thesetrend
analysiscomponentsare

<b>Cyclicalcomponentthatdescribesrepeatedbutnon-periodicfluctuations,</b>

<b>Seasonalcomponentthatreflectsseasonality(seasonalvariation),</b>

<b>Irregularcomponent(or“noise”)thatdescribesrandom,irregularinfluences</b>

andrepresentstheresidualsofthetimeseriesaftertheothercomponentshave
beenremoved.

Fromthetimeseriesdecompositionanalysis,abusinessusercanspotparticular
areasofinterestinthedecomposedtrenddatathatmaybeworthyoffurther
analysis(seeFigure6.7).

</div>
<span class='text_page_counter'>(161)</span><div class='page_container' data-page=161></div>
<span class='text_page_counter'>(162)</span><div class='page_container' data-page=162>

<b>TheParksRamifications</b>

TheParkscandeploytimeseriesdecompositionanalysistoidentifyand
quantifytheimpactthatseasonalityandspecificeventsarehavingonguest
visitsandassociatedspend.TheParkscanusetheresultsoftheanalysisto

1. Createseason-specificmarketingcampaignsandpromotionstoincrease
guestvisitsandassociatedspend,

2. Determinewhicheventsoutsideofthethemeparks(concerts,professional
sportingevents,BCSfootballgames)areworthyofpromotionaland

sponsorshipspend.

FormoreinformationabouttimeseriesdecompositioninR,checkout

</div>
<span class='text_page_counter'>(163)</span><div class='page_container' data-page=163>

<b>AnalyticAlgorithmsandModels</b>

Thefollowinganalyticalgorithmsstarttomovethedatascientistbeyondthedata
explorationstageintothemorepredictivestagesoftheanalysisprocess.These
analyticalgorithmsbytheirnaturearemoreactionable,allowingthedatascientist
toquantifycauseandeffectandprovidethefoundationtopredictwhatislikelyto
happenandrecommendspecificactionstotake.

<b>ClusterAnalysis</b>

Clusteranalysisisusedtouncoverinsightsabouthowcustomersand/orproducts
clusterintonaturalgroupingsinordertodrivespecificactionsor

recommendations(e.g.,personalizedmessaging,targetmarketing,maintenance
scheduling).Clusteranalysisorclusteringistheexerciseofgroupingasetof

objectsinsuchawaythatobjectsinthesamegrouparemoresimilartoeachother
thantothoseinothergroups(clusters).

Clusteringanalysiscanuncoverpotentialactionableinsightsacrossmassivedata
volumesofcustomerandproducttransactionsandevents.Clusteranalysiscan
uncovergroupsofcustomersandproductsthatsharecommonbehavioral
tendenciesand,consequently,andcanbetargetedwiththesamemarketing

treatments(seeFigure6.8).

</div>
<span class='text_page_counter'>(164)</span><div class='page_container' data-page=164>

<b>NOTE</b>

</div>
<span class='text_page_counter'>(165)</span><div class='page_container' data-page=165>

<b>TheParksRamifications</b>

TheParkscanleverageclusteranalysistocreatemoreactionableprofilesof
thepark'smostprofitableguestclustersandhighestpotentialguestclusters.
TheParkscanusetheresultsoftheanalysistoquantify,prioritize,andfocus
guestacquisitionandguestactivationmarketingefforts.

FormoreinformationaboutclusteranalysisinR,checkout

/>

<b>NormalCurveEquivalent(NCE)Analysis</b>

Atechniquefirstusedinevaluatingstudents'testingperformance,normalcurve
equivalent(NCE),isadatatransformationtechniquethatapproximatelyfitsa
normaldistributionbetween0and100bynormalizingadatasetinpreparation
forpercentilerankanalysis.Forexample,anNCEdatatransformationisawayof
standardizingscoresreceivedonatestintoa0–100scalesimilartoapercentile
<i>rankbutpreservingthevaluableequal-intervalpropertiesofaz-score(seeFigure</i>
6.9).

<b>Figure6.9</b>Normalcurveequivalentanalysis

</div>
<span class='text_page_counter'>(166)</span><div class='page_container' data-page=166></div>
<span class='text_page_counter'>(167)</span><div class='page_container' data-page=167></div>
<span class='text_page_counter'>(168)</span><div class='page_container' data-page=168>

<b>TheParksRamifications</b>

TheParkscanemploytheNCEtechniquetounderstandpriceinflection

pointsforpackagesofattractionsandrestaurants.TheParkscanleveragethe
priceinflectionpointstooptimizepricing(e.g.,createapackageofattractions
andrestaurantsbyseasonality,holiday,dayofweek,etc.)andcreatenew
PriorityAccesspackages.

<i>Formoreinformationabouthowtousez-scorestonormalizedatausingR,</i>
checkout

Formore

insightsintotheNCEdatatransformationtechnique,see

/>

<b>AssociationAnalysis</b>

Associationanalysisisapopularalgorithmfordiscoveringandquantifying
relationshipsbetweenvariablesinlargedatabases.Associationanalysisshows
customerorproducteventsoractivitiesthattendtohappentogether,which
makesthistypeofanalysisveryactionable.Forexample,theassociationrule
{buns,ketchup}→{burger}foundinthepoint-of-salesdataofasupermarket
wouldindicatethatifacustomerbuysbunsandketchuptogether,sheislikelyto
alsobuyhamburgermeat.Suchinformationcanbeusedasthebasisformaking
pricing,productplacement,promotion,andothermarketingdecisions.

Associationanalysisisthebasisformarketbasketanalysis(identifyingproducts
and/orservicesthatsellincombinationorsellwithapredictabletimelag)thatis
usedinmanyindustriesincludingretail,telecommunications,insurance,digital
marketing,creditcards,banking,hospitality,andgaming.

</div>
<span class='text_page_counter'>(169)</span><div class='page_container' data-page=169>

<b>Figure6.11</b>Associationanalysis

Oneveryactionabledatasciencetechniqueistoclustertheresultingassociation
rulesintocommongroupsorsegments.Forexample,inFigure6.12,thedata
scienceteamclusteredtheresultingassociationrulesacrosstensofmillionsof
customersinordertocreatemoreaccurate,relevantcustomersegments.Inthis
process,thedatascienceteam

Runstheassociationanalysisacrossthetensofmillionsofcustomersto
identifyassociationruleswithahighdegreeofconfidence,

Clustersthecustomersandtheirresultingassociationrulesintocommon

groupingsorsegments(e.g.,Chipotle+Starbucks,VirginAmerica+Marriott),
Usesthesenewsegmentsasthebasisforpersonalizedmessaginganddirect
marketing.

<b>Figure6.12</b>Convertingassociationrulesintosegments

</div>
<span class='text_page_counter'>(170)</span><div class='page_container' data-page=170></div>
<span class='text_page_counter'>(171)</span><div class='page_container' data-page=171>

<b>TheParksRamifications</b>

TheParkscanleveragemarketbasketanalysistoidentifythemostpopular
andleastpopular“packagesofattractions.”TheParkscanusethis“packages
ofattractions”datato

1. CreatenewpricingandPriorityAccesspackagesforthemostpopular
packagesinordertooptimizein-parktrafficflowandreduceattraction
waittimes,

2. CreatenewpricingandPriorityAccesspackagesfortheleastpopular
“packages”inordertodrivetraffictounder-utilizedattractions.

FormoreinformationaboutassociationanalysisinR,checkout

/>

<b>GraphAnalysis</b>

Graphanalysisisoneofthemorepowerfulanalysistechniquesmadepopularby
socialmediaanalysis.Graphanalysiscanquicklyhighlightcustomerormachine
(thinkInternetofThings)relationshipsobscuredacrossmillionsifnotbillionsof
socialandmachineinteractions.

Graphanalysisusesmathematicalstructurestomodelpairwiserelationsbetween
objects.A“graph”inthiscontextismadeupof“vertices”or“nodes”andlines
callededgesthatconnectthem.Socialnetworkanalysis(SNA)isanexampleof
graphanalysis.SNAisusedtoinvestigatesocialstructuresandrelationships
acrosssocialnetworks.SNAcharacterizesnetworkedstructuresintermsofnodes
(peopleorthingswithinthenetwork)andthetiesoredges(relationshipsor

interactions)thatconnectthem(seeFigure6.13).

<b>Figure6.13</b>Graphanalysis

</div>
<span class='text_page_counter'>(172)</span><div class='page_container' data-page=172></div>
<span class='text_page_counter'>(173)</span><div class='page_container' data-page=173>

<b>TheParksRamifications</b>

TheParkscanemploygraphanalysistouncoverstrengthofrelationships
amonggroupsofguests(leaders,followers,influencers,cohorts).TheParks
canleveragethegraphanalysisresultstodirectpromotions(discounts,
restaurantvouchers,travelvouchers)togroupleadersinordertoencourage
theseleaderstobringgroupsbacktotheparksmorefrequently.

FormoreinformationaboutsocialnetworkanalysisinR,checkout

-r-using-package-igraph/.

<b>TextMining</b>

Textminingreferstotheprocessofderivingusableinformation(metadata)from
textfilessuchasconsumercomments,e-mailconversations,physicianor

techniciannotes,workorders,etc.Basically,textminingcreatesstructureddata
outofunstructureddata.

Textminingisaverypowerfultechniquetoshowduringanenvisioningprocess,
asmanybusinessstakeholdershavestruggledtounderstandhowtheycangain
insightsfromthewealthofinternalcustomer,product,andoperationaldata.Text
miningisnotsomethingthatthedatawarehousecando,somanybusiness

stakeholdershavestoppedthinkingabouthowtheycanderiveactionableinsights
fromtextdata.Consequently,itisimportanttoleverageenvisioningexercisesto
helpthebusinessstakeholderstoimagetherealmofwhatispossiblewithtext
data,especiallywhenthattextdataiscombinedwiththeorganization's

operationalandtransactionaldata.

</div>
<span class='text_page_counter'>(174)</span><div class='page_container' data-page=174>

<b>Figure6.14</b>Textmininganalysis

</div>
<span class='text_page_counter'>(175)</span><div class='page_container' data-page=175></div>
<span class='text_page_counter'>(176)</span><div class='page_container' data-page=176>

<b>TheParksRamifications</b>

TheParkscanmineguestcomments,socialmediaposts,ande-mailstoflag

andrankareasofconcernandproblemsituations.TheParkscanleveragethe
textminingresultstolocateunsatisfiedguestsinordertodrivepersonal
(face-to-face)guestinterventionefforts.

FormoreinformationabouttextmininganalysisusingR,checkout

/>

<b>SentimentAnalysis</b>

Sentimentanalysiscanprovideabroadandgeneraloverviewofyourcustomers'
sentimenttowardyourcompanyandbrands.Sentimentanalysiscanbea

powerfulwaytogleaninsightsaboutthecustomers'feelingsaboutyourcompany,
products,andservicesoutoftheever-growingbodyofsocialmediasites

(Facebook,LinkedIn,Twitter,Instagram,Yelp,Snapchat,Vine,etc.)(seeFigure
6.15).

<b>Figure6.15</b>Sentimentanalysis

InFigure6.15,thedatascienceteamconductedcompetitivesentimentanalysisby
classifyingtheemotions(e.g.,anger,disgust,fear,joy,sadness,surprise)of

</div>
<span class='text_page_counter'>(177)</span><div class='page_container' data-page=177>

keycompetitor'sperceivedperformanceandqualityofserviceissuffering).

Unfortunately,itissometimesdifficulttogetthesocialmediadataatthelevelof
theindividual,whichisrequiredtocreatemoreactionableinsightsand

recommendationsattheindividualcustomerlevel.However,leading

</div>
<span class='text_page_counter'>(178)</span><div class='page_container' data-page=178>

<b>TheParksRamifications</b>

TheParkscanestablishasentimentscoreforeachattractionandcharacter

andmonitorsocialmediasentimentfortheattractionsandcharactersinreal-time.TheParkscanleveragethereal-timesentimentscorestotakecorrective
actions(placateunhappyguests,openadditionallines,openadditional

attractions,removekiosks,movecharacters).

FormoreinformationaboutsentimentanalysisusingR,checkout

/>

<b>TraversePatternAnalysis</b>

Traversepatternanalysisisanexampleofcombiningacoupleofanalytic

algorithmstobetterunderstandcustomer,product,oroperationalusagepatterns.
Traverseanalysislinksacustomerorproductusagepatternsandassociationrules
toageographicalorfacilitymapinordertoidentifypotentialpurchase,traffic,
flow,fraud,theft,andotherusagepatternsandrelationships.

Theprocessstartsbycreatingassociationrulesfromthecustomer'sorproduct's
usagedata,andthenmapsthoseassociationrulestoageographicalmap(store,
hospital,school,campus,sportsarena,casino,airport)toidentifypotential
performance,usage,staffing,inventory,logistics,trafficflow,etc.problems.
InFigure6.16,thedatascienceteamcreatedaseriesofassociationrulesabout
slotandtableplayinacasino,andthenusedthoseassociationrulestoidentify
potentialfootflowproblemsandgamelocationoptimizationopportunities.The
datascienceteam

Createdplayerperformanceassociationrulesaboutwhatgamestheplayers
tendtoplayincombination,

</div>
<span class='text_page_counter'>(179)</span><div class='page_container' data-page=179>

<b>Figure6.16</b>Traversepatternanalysis

Theresultsofthisanalysishighlightsareasofthecasinothataresub-optimized
whencertaintypesofgameplayersareinthecasinoandcanleadto

</div>
<span class='text_page_counter'>(180)</span><div class='page_container' data-page=180></div>
<span class='text_page_counter'>(181)</span><div class='page_container' data-page=181>

<b>TheParksRamifications</b>

TheParkscanemploytraversepatternanalysistounderstandparkandguest
flowswithrespecttoattractions,entertainment,retailoutlets,restaurants,
characters,etc.TheParkscanusethetraversepatternanalysisresultsto

1. Identifywheretoplacecharactersandsituateportablekiosksinorderto
increaserevenues,

2. Determinewhatpromotionstoofferinordertodrivetraffictoidle
attractionsandrestaurants.

<b>DecisionTreeClassifierAnalysis</b>

Decisiontreeclassifieranalysisusesdecisiontreestoidentifygroupingsand
clustersburiedintheusageandperformancedata.Decisionclassifieranalysis
usesadecisiontreeasapredictivemodelthatmapsobservationsaboutanitemto
conclusionsabouttheitem'stargetvalue.

InFigure6.17,thedatascienceteamusedthedecisiontreeclassifieranalysis
techniquetoidentifyandgroupperformanceandusagevariablesintosimilar

clusters.Thedatascienceteamuncoveredproductperformanceclustersthat,
whenoccurringincertaincombinations,wereindicativeofpotentialproduct
performanceormaintenanceproblems.

</div>
<span class='text_page_counter'>(182)</span><div class='page_container' data-page=182>

<b>TheParksRamifications</b>

TheParkscanusedecisiontreeclassifieranalysistoquantifythevariables
thatdriveguestsatisfactionandincreasespendbyguestclusters.TheParks
canleveragethedecisiontreeclassifieranalysisresultstodeterminewhich
variablestomanipulateinordertodriveguestsatisfactionandassociated
guestspend.

FormoreinformationaboutbuildingdecisiontreesusingR,checkout“Tree-BasedModels”at />

<b>CohortsAnalysis</b>

Cohortsanalysisisusedtoidentifyandquantifytheimpactthatanindividualor
machineshaveonthelargergroup.
Cohortsanalysisiscommonlyusedbysportsteamstoascertaintherelativevalue
ofaplayerwithrespecttohisorherinfluenceonthesuccessoftheoverallteam.
TheNationalBasketballAssociationusesarealplus-minus(RPM)metricto
measureaplayer'simpactonthegame,representedbydifferencebetweenthe
team'stotalscoringanditsopponent's.Table6.1showstopRPMplayersfromthe
2014–2015NBAseason.
<b>Table6.1</b>2014–2015TopNBARPMRankings

<b>Rank Player</b> <b>Team MPG RPM</b>

1 StephenCurry,PG GS 32.7 9.34
2 LeBronJames,SF CLE 36.1 8.78
3 JamesHarden,SG HOU 36.8 8.50
4 AnthonyDavis,PF NO 36.1 8.18
5 KawhiLeonard,SF SA 31.8 7.57

6 RussellWestbrook,PG OKC 34.4 7.08
7 ChrisPaul,PG LAC 34.8 6.92
8 DraymondGreen,SF GS 31.5 6.80
9 DeMarcusCousins,C SAC 34.1 6.12
10 KhrisMiddleton,SG MIL 30.1 6.06

Source: />

</div>
<span class='text_page_counter'>(183)</span><div class='page_container' data-page=183></div>
<span class='text_page_counter'>(184)</span><div class='page_container' data-page=184></div>
<span class='text_page_counter'>(185)</span><div class='page_container' data-page=185>

<b>TheParksRamifications</b>

TheParkscanemploycohortsanalysistoidentifyspecificemployeesand
charactersthatincreasetheoverallpark,attractions,characters,customer,
andhouseholdsatisfactionandspendlevels.TheParkscanusetheresultsof
thecohortsanalysisto

1. Decidehowmanyandwheretosituatespecific,popularcharacters;
2. Rewardparkassociatesthatdrivehighercustomersatisfactionscores.
FormoreinformationaboutcohortsanalysisinR,checkoutthearticle
“CohortAnalysiswithR–RetentionCharts”at

</div>
<span class='text_page_counter'>(186)</span><div class='page_container' data-page=186>

<b>Summary</b>

TheobjectiveofChapter6istogiveyouatasteforthedifferenttypesofanalytic
algorithmsadatascienceteamcanbringtobearonthebusinessproblemsor
opportunitiesthattheorganizationistryingtoaddress.Thischapterbetter
acquaintedyouwiththedifferentalgorithmsthatthedatascienceteamcanuseto
acceleratethebusinessuseranddatascienceteamcollaborationprocess.Whileit
isnottheexpectationofthisbookorchaptertoturnbusinessusersintodata
scientists,itismyhopethatChapter6willsetthefoundationthathelpsbusiness
<i>usersandbusinessleadersto“thinklikeadatascientist.”</i>
Thischapterintroducedawidevarietyofanalyticalgorithmsthatthedatascience
teammightuse,dependingontheproblembeingaddressedandthetypesand

varietiesofdataavailable.Italsointroducedafictitiouscompany(Fairy-Tale
ThemeParks)againstwhichyouappliedthedifferentanalytictechniquestosee
thepotentialbusinessactions(seeTable6.2).
<b>Table6.2</b>CaseStudySummary

</div>
<span class='text_page_counter'>(187)</span><div class='page_container' data-page=187></div>
<span class='text_page_counter'>(188)</span><div class='page_container' data-page=188></div>
<span class='text_page_counter'>(189)</span><div class='page_container' data-page=189>

<b>HomeworkAssignment</b>

Usethefollowingexercisestoapplywhatyoulearnedinthischapter.

<b>Exercise#1:Revieweachoftheanalyticalgorithmscoveredinthischapter</b>

andwritedownoneortwousecaseswherethatparticularanalyticalgorithm
mightbeusefulgivenyourbusinesssituations.

<b>Exercise#2:Revisitthekeybusinessinitiativethatyouidentifiedin</b>Chapter
2.Writedowntwoorthreeoftheanalyticalgorithmscoveredinthischapter
thatyouthinkmightbeappropriatetothedecisionsthatyouaretryingto
makeinsupportofthatkeybusinessinitiative.

<b>Exercise#3:Writedowntwoorthreebulletpointsaboutwhyyouthink</b>

</div>
<span class='text_page_counter'>(190)</span><div class='page_container' data-page=190>

<b>Notes</b>

1_{Risaprogramminglanguageandsoftwareenvironmentforstatistical}

</div>
<span class='text_page_counter'>(191)</span><div class='page_container' data-page=191></div>
<span class='text_page_counter'>(192)</span><div class='page_container' data-page=192></div>
<span class='text_page_counter'>(193)</span><div class='page_container' data-page=193></div>
<span class='text_page_counter'>(194)</span><div class='page_container' data-page=194>

<b>NOTE</b>

Itistypicalthat40to60percentofthedatawarehouseprocessingloadis
performingETLwork.Off-loadingsomeoftheETLprocessestothedata
lakecanfreeupconsiderabledatawarehouseresources.

UnhandcufftheBIanalystsanddatascienceteamfrombeingreliantonthe
summarizedandaggregateddatainthedatawarehouseasthesinglesourceof
datafortheirdataanalytics(andmitigatetheunmanageableproliferation
“spreadmarts”1thatarebeingusedbybusinessanalyststoworkaroundthe
analyticlimitationsofthedatawarehouse).

Thedatalakesolvesagreatmanyproblems.However,itcanalsoraisealotof
questions.Inapapertitled“BewaretheDataLakeFallacy”

(Gartnerraisedcautionsaboutthe

datalake,specificallyaroundtheassumptionthatallenterpriseaudiencesare
highlyskilledatdatamanipulationandanalysis.Gartner'spointwasthatifadata
lakefocusesonlyonstoringdisparatedataandignoreshoworwhydataisused,
governed,defined,andsecuredorhowdescriptivemetadataiscapturedand
maintained,thedatalakerisksturningintoadataswamp.Andwithoutan
adequatemetadatastrategy,everysubsequentuseofdatameanstheanalysts
muststartfromscratch.

Theabilityofanorganizationtorealizebusinessvaluefrombigdatareliesonthe
organization'sabilitytoeasilyandquickly:

Identifythe“rightand/orbestdata”

Definetheanalyticsrequiredtoextractthevalue

Bringthedataintoananalyticsenvironment(sandbox)suitedforadvanced
analyticsordatasciencework

Curatethedatatoapointwhereitis“suited”foranalysis

Standuptherequiredinfrastructuretosupporttheanalyticsinaccordance
withthedesiredperformanceandthroughputrequirements

Executetheanalyticmodelsagainstthecurateddatatoderivebusinessvalue
Deploytheanalyticsintotheproductioninfrastructure

</div>
<span class='text_page_counter'>(195)</span><div class='page_container' data-page=195>

<b>NOTE</b>

</div>
<span class='text_page_counter'>(196)</span><div class='page_container' data-page=196>

<b>CharacteristicsofaBusiness-ReadyDataLake</b>

Thedatalakeisnotanincrementalenhancementtothedatawarehouse,anditis
NOTdatawarehouse2.0.Thedatalakeenablesentirelynewcapabilitiesthat
allowyourorganizationtoaddressdataandanalyticchallengesthatthedata
warehousecouldnotaddress.

Therearefivecharacteristicsthatdifferentiateabusiness-readydatalakefromthe
datawarehouse(seeFigure7.1):

<b>Figure7.1</b>Characteristicsofadatalake

<b>Ingest.Abilitytorapidlyingestdatafromawiderangeofinternaland</b>

externaldatasources,includingstructuredandunstructureddatasources.The

datalakecanaccomplishrapiddataingestionbecauseitcanloadthedataas-
is;thatis,thedatalakedoesnotrequireanydatatransformationsorpre-buildingadataschemabeforeloadingthedata.

<b>Store.AsingleorcentralrepositoryforamassingALLtheorganization'sdata</b>

includingdatafrompotentiallyinterestingexternalsources.Thedatalakecan
storedataeveniftheorganizationhasnotyetdecidedhowitmightusethe
data.AstheDirectorofAnalyticsandBusinessIntelligenceatStarbuckswas
quoted:“AfullquarterofStarbuckstransactionsaremadeviaitspopular

loyaltycards,andthatresultsin“hugeamounts”ofdata,butthecompanyisn't
surewhattodowith[allthatdata]yet.”Thesamegoesforsocialmediadata,
asStarbuckshasateamwhoanalyzessocialdata,but“Wehaven'tfiguredout
whatexactlytodowithityet.”2

<b>Analyze.Providesthefoundationfortheanalyticsenvironment(oranalytics</b>

</div>
<span class='text_page_counter'>(197)</span><div class='page_container' data-page=197>

internalandexternaldatasourceswiththegoalofuncoveringnewcustomer,
product,andoperationalinsightsthatcanbeusedoptimizekeybusiness
processesandfuelnewmonetizationopportunities.

<b>Surface.Supportstheanalyticmodeldevelopmentandtheextractingofthe</b>

analyticresults(e.g.,scores,recommendations,nextbestoffer,businessrules)
thatareusedtoempowerfrontlineemployees'andbusinessmanagers'

decisionmakingandinfluencecustomerbehaviorsandactions.

<b>Act.Enablestheintegrationoftheanalyticresultsbackintotheorganization's</b>

operationalsystems(callcenter,directmarketing,procurement,store

</div>
<span class='text_page_counter'>(198)</span><div class='page_container' data-page=198></div>
<span class='text_page_counter'>(199)</span><div class='page_container' data-page=199>

Asadatawarehousemanager,Ihatedtheanalyticsteam.Why?Because

wheneveritsmembersneededdata,theyalwayscametomydatawarehousefor
thedatabecausetheyweretoldthatthedatawarehousewasthe“singleversionof
thetruth.”Andtheanalyticteam'sdataandqueryrequestsusuallyscrewedupmy
productionSLAsintheprocess(seeFigure7.2).

<b>Figure7.2</b>Theanalyticsdilemma

</div>
<span class='text_page_counter'>(200)</span><div class='page_container' data-page=200>

<b>Figure7.3</b>Thedatalakelineofdemarcation

Thedatalakeprovidesa“lineofdemarcation”betweentheproduction

</div>

<!--links-->
<a href=' /><a href=''>SeekingAlpha.com</a>
<a href=''>Realtor.com</a>
<a href=' /><a href='http:// gains-high-food-costs-dent-margins/2013-10-21'> </a>
<a href=' /><a href=' /><a href=' /><a href=''>Mint.com</a>
<a href=' /><a href=' /><a href=' /><a href=' /><a href=' /><a href=' /><a href=' /><a href=' /> MBA (International Business): PROSPECTUS 2013-15 potx

• 16
• 263
• 0