Practical Fundamentals for Master Data Management

Practical Fundamentals for Master Data Management A DataFlux White Paper Prepared by David Loshin Introduction Enterp prisewide programs are oft designed to improve b ten business processes and busine decisions through enha ess t anced informa ation sharing – based on a more reliable founda ation of data. For this purp pose, these in nitiatives prom mise significan nt organi izational bene efits along a number of dim n mensions of v value, but, as with any busine project, th can also carry significan risks and co ess hey c nt omplexity. Organ nizations that have successffully deployed programs t d that span the enterprise ar re prime candidates fo a successfu enterprise m or ul master data mmanagement (MDM) initiativ those lack ve; king this expe erience are at greater risk f MDM failu However, for ure. , since the early tang t gible value of an MDM stra ategy comes f from standard dized seman ntics, consolidated metad data, and mos critically, d ata quality im st mprovement , focusin on these fundamental aspects of MD reduces t risks while providing ng a DM the e practic utility to th data user community. cal he c In this paper we loo at the level of maturity and capabilit necessary f deploying ok ls ty for enterp prisewide initiatives, such as comprehen a nsive MDM, and ponder why these initiativ stall. In tu we will co ves urn, onsider the typ of value d pes derived from MDM and which tangible benefits can be achieved early in the proce Since man of these a y ess. ny concen ntrate on simplifying and standardizing semantics, m s managing met tadata, and improv ving data quaality, this pape suggests th starting w tasks that address thos er hat with se fundam mental needs will add value and prepar the organiz s re zation to take the steps e neede to incrementally build th master dat capability. In conclusion some ed he ta n, concre actions can be taken to position tho fundamen ete o ose ntals as the fir step in rst growin a long-term business-or ng m riented enterpprise informa ation strategy. Con nsidering MDM Benefits g s New te echnologies such as maste data manag s er gement are ooften positioned as “silver bullets when it com to addressing long-sta s” mes anding system challenge Every mic es. corporrate technology of the preceding decad – data war de rehouses, enterprise resour planning (ERP), customer relationshi manageme (CRM) – arrived with rce ( ip ent similar levels of fanf r fare to solve the problem o managing data across the enterprise t of e. It shou not be sur uld rprising that the numerous benefits sug s ggested by the hype surrounding MDM might lead to inflated exp ectations. Yet when the du has settled o t ust d, the too are installe and the data integratio is done, or ols ed, d on rganizations o often find that they are still stalled in progressin towards th elusive “36 ng hat 60-degree vie of the ew” custom product, or other data element. mer, a Perhap too much faith is placed in the conve ps f d entional wisdo surrounding the value om of MDM? When rev viewing the di ifferent white papers, vend briefs, ana dor alyst reports o or the ma articles on the topic, it is valuable to consider the business drivers for – and any n o e d purported benefits of – implemeenting MDM. 1 In general, the pres sumptive benefits achieved using MDM techniques c be d M can orized into the three clas catego ese sses:  Vague stateements with no measurabl e characterist or perform n tics mance metric cs (such as “becoming a smmarter organiz zation”)  Valuable buusiness benef for which MDM is nece fits essary but not sufficient t (such as “re educing merg and acqui sition costs”) ger  Data impro ovement bene from better data quality an data efits derived f nd governance practices e They Sound Good, Bu y G ut… Senior business exe r ecutives are fo ocused on the organization’s strategic o e objectives and how measurable pe m erformance metrics demon m nstrate busine value. In th light, the ess hat value proposition fo any propos technolog must be co p or sed gy oupled with hhard metrics that ca be used to baseline the current state as well as sh an o e how measured d improvvement. In the examples abo ove, statemen like “enab nts bling people t make decisions more to quickly or “making your organiz y” g zation smarte both soun desirable. H er” nd However, neither has discrete measures of quantificatio and you m e f on, might ask whether either haas any rea business meaning at all. After all, it is unclear whether concepts like “the al m s s speed of decision making” or “o m organizationa l smartness” contribute to achieving an o n organi ization’s strateegic objective Other exa es. amples are eq qually devoid of substance, such as “reducing the count of vendor record or “breaking down org t v ds” ganizational silos.” These are mo likely to be outcomes of more direc business pr ore b ct rocess improv vements focused on revenue generatio reducing c on, costs or decre easing operat tional risk, but do not add value in and of themselves.