《DBP：2022年在AI和BI中应用语义层的商业影响分析报告（英文版）（28页）.pdf》由会员分享，可在线阅读，更多相关《DBP：2022年在AI和BI中应用语义层的商业影响分析报告（英文版）（28页）.pdf（28页珍藏版）》请在三个皮匠报告上搜索。

1、Data Leader Study&Research ResultsResearch Conducted byPrashanth Southekal,PhD.,MBA,MSManaging Principal,DBP InstituteForeword byJon FrancisChief Data and Analytics Officer at PayPalConclusion byMegan C.Brown,Ph.D.Director of Data Literacy at Starbuckswww.dbp-The Business Impact of Using a Semantic

2、Layer for AI&BIby 100+Enterprise Data and Analytics Professionals2ForewordIntroductionExecutive SummaryOverview of the Semantic LayerDesign of the Research StudyKey Research Study FindingsConclusionThe Research AnalystReferencesAppendix 1:Data and Analytics MaturityAppendix 2:Semantic Layer Use-case

3、sAppendix 3:AcronymsAppendix 4:Glossary34684272829This research helps data and analytics professionals discover what may be one of the best-kept secrets for improving AI and BI that using a Semantic Layer significantly improves the speed,scale,and cost to create actionable insights and an

4、alytics for decision-makers.Having developed and deployed data and analytics capabilities and teams at Paypal,Amazon,Microsoft,T-Mobile,Nike,and Starbucks,what really matters is the ability to deliver actionable insights across the enterprise where speed and scale drive impact.To drive speed and sca

5、le while managing costs,organizations need to move from the traditional“get in line”delivery model,to a“get it when you need it”approach by applying a“hub-and-spoke”delivery model.We do this by embracing modern cloud-based data platforms,tools and processes,and decentralizing insights and analytics

6、creation supported by centralized data management.The Semantic Layer was created to enable enterprises to develop actionable insights using AI and BI at scale,supporting decentralized insights creation with centralized data governance and data management.The Semantic Layer is a key piece of technolo

7、gy that enables companies to develop and deploy data products(e.g.data models,metrics,and insights)rapidly and effectively at scale.As the research points out,consistent with my experience using a Semantic Layer,the benefits of a Semantic Layer to scale AI and BI are both significant and extensive,i

8、ncluding time to insights,scale,and savings in cost.Spee?Accelerating time to insights?Accelerating data preparation time?Accelerating time to analytics model creation Scale?Expanding the number of users for self-service insights creation?Expanding the number of data sources available?Enabling metri

9、c definition and usage consistency Cost Saving?Improving ROI?Reducing Cost?Improving EfficiencyI hope that this research helps other data and analytics leaders and practitioners understand and embrace the business benefits of using a Semantic Layer to deliver actionable insights from AI and BI with

10、improved speed,scale,and cost savings.Enjoy!Chief Data and Analytics Officer,PaypalJon FrancisA Semantic Layer Helps Enable Actionable Insights for EveryoneForeword34IntroductionCompanies are in the midst of making major investments in modern cloud-based data platforms,which includes applying advanc

11、ed analytics to improve business speed,scale,and performance.Questions concerning“if an investment to enable data,insights,and analytics is warranted have been answered with an overwhelming response of“yes”by most organizations.Now,companies are asking“how to scale”the delivery of insights and analy

12、tics beyond a few initial use cases.As a recent study Tipping the Scales in AI indicates,companies that scale insights and analytics deliver about+8%pts more EBIT(3.4x better)than companies who have not achieved scale.This research,based on actual experience from senior enterprise data and analytics

13、 participants who have chosen to deploy the use of a Semantic Layer,quantifies the actual benefits of using such a solution to realize the benefits of achieving speed,scale,and greater cost savings to deliver actionable insights from AI and Business Intelligence(BI).While there are many potential wa

14、ys to improve the speed,scale,and impact of data and analytics projects,one solution which has been used effectively by enterprises over the past ten years is a Semantic Layer.The Semantic Layer is a proven,trusted software component within the modern cloud data platform technology stack that focuse

15、s on improving the speed to actionable insights for BI and AI/ML.The Semantic Layer enables data analysts and data scientists to more rapidly,effectively,and consistently access and create data products that deliver actionable insights.Although the Semantic Layer has been in the marketplace for over

16、 ten years,the current level of awareness of using a Semantic Layer,including accelerating time to insights,scale,and cost savings are not widely understood among many business,IT,and data and analytics professionals.To address the Semantic Layer business impact awareness gap,this research captures

17、the actual business benefits from using a Semantic Layer as identified and quantified by senior enterprise data and analytics professionals whose organizations have deployed and are currently using a Semantic Layer.Their feedback is presented in this report,including quantifying the actual tangible

18、business benefits of using a Semantic Layer to increase speed to actionable insights,rapidly expand scale,and reduce costs.McKinseyYours Sincerely,Prashanth Southekal,PhD.,MBA,MSManaging Principal,DBP InstituteAdjunct Professor at IE Business School,SpainE:psouthekaldbp-W:www.dbp-5Semantic Layer Stu

19、dy-OverviewThis study focuses on understanding the business impact of scaling the delivery of data,insights,and analytics using a Semantic Layer for AI and BI.A Semantic Layer provides a single,common definition and representation of all data sources,metrics,and models,which enables the rapid creati

20、on of consistent insights for AI and BI without any data movement,coding,and waiting.A Semantic layer creates a way to define data as a product that can be consumed consistently and efficiently across the enterprise,enabling decentralized insights creation supported with centralized data management

21、platforms and governance.This study summarizes the actual results experienced by over 100 data and analytics professionals who have chosen to use a Semantic Layer to scale their data and analytics capabilities and impacts.DBP Institute,a data and analytics consulting,research,and education organizat

22、ion,launched a research study to assess the value of the Semantic Layer from industry practitioners.The study was fielded online in April 2022 and only data from validated respondents who were data and analytics experts using a Semantic Layer were accepted.Executive SummaryOverall,companies using a

23、Semantic Layer cite a 4.2x improvement(i.e.,a magnitude of 4.2 times improvement over the base level of performance from not using a semantic layer)in performance with less than half the effort required(e.g.savings in both number of resources,hours,project time/duration,and overall cost).This is a s

24、ignificant order-of-magnitude improvement in performance as well as a reduction in effort and cost.It means that a typical project taking 4 months to complete could be done in just 4 weeks using a Semantic Layer!Performance improvement was significant and consistent across every measur?4.4x improvem

25、ent in Time-to-Insights(e.g.,insights&analytics development?4.4x improvement in number of self-service users,data sources,metrics consistenc?4.2x improvement in Cloud Analytics performanc?3.7x improvement in cost savingsUsing a Semantic Layer Generates 4.2x Improvement with Half the Effort6Figure 1:

26、Overall improvement for speed,scale,and cost savingsPerformance Improvement in ScalePerformance Improvementin Time to InsightsPerformance Improvementin Savings in CostOveral PerformanceImprovement4.4X4.4X3.7X4.2XData and analytics professionals quantified the benefits of using a Semantic Layer based

27、 on actual experience for nine(9)individual measures.These nine measures were assigned to one of three major benefit drivers,summarized as follows?Accelerate Time to Insights-4.4x magnitude of improvement in time to insights.This means that using a Semantic Layer significantly accelerates speed to a

28、ctionable insights.For example,if a project took one month to make insights available without a semantic layer,using a semantic layer would reduce that time from 30 days to just one week,or a quarter of the time required.Time to insights addresses additional measures of performance,including 4.2x im

29、provement in data preparation time,4.6x improvement in time to analytics creation,and 4.5x improvement in time to insights creation?Increase Scale-4.4x overall improvement in scale.Specifically,companies using a Semantic Layer say that they have increased the number of self-service AI and BI users b

30、y 4.5x,increased the number of data sources by 4.8x,and improved data metrics consistency by 4.0 x?Reduce Cost and Resource Effort-3.7x magnitude reduction in cost.Additional performance measures impacted by using a Semantic layer include a 3.1x reduction in compute cost,3.9x improvement in AI and B

31、I project ROI,and 4.2x improvement in resource efficiency(e.g.productivity).Further,companies cite a 46%reduction in the number of hours required to implement a data and analytics project using a Semantic Layer vs.not using a semantic layer.That means for a typical 1,000 hour project,using a Semanti

32、c Layer reduces the number of hours to just 540,or a reduction of 460 hours,which is a very significant reduction in hours of effort and cost of effort.7The impact of using a Semantic Layer is significant across every performance measure,including each key business driver,such as speed to insights,i

33、ncreased scale,and reduced cost and effort.The benefits of using a Semantic Layer add up quickly for a typical 1,000 hour project that costs$200/hour for resources(i.e.,$200k),the level of effort is reduced by close to half(actual results cited are 46%reduction)using a Semantic Layer.This translates

34、 to a cost savings of nearly half,or a cost savings of$92k for the typical 1,000 hour project.For a company implementing just 25 typical 1,000 hour projects a year using a Semantic Layer,the direct,realizable cost savings would be$2.3mm annually.The return on investment using a Semantic Layer is sig

35、nificant and obvious deploying the use of a Semantic Layer should yield immediate direct savings and pay for itself well within a year.?4.4x improvement in Scal?4.4x improvement Speed to Insight?3.7x Savings in Cos?46%reduction in the number of hours required to implement a typical Data and Analytic

36、s project4x Faster4x Faster Speed-to-InsightsWith Semantic LayerWithout Semantic LayerWith Semantic LayerWithout Semantic Layer3.7x Reduction in Total CostWith Semantic LayerWithout Semantic LayerIncreasing Scale of Users&Data by 4xSignificant,Rapid Return on Investment(ROI)8Overview of the Semantic

37、 LayerTurning Data into Actionable Insights for EveryoneThe fundamental goal of data and analytics is to provide decision-makers with actionable insights.To deliver actionable insights more effectively,particularly in the context of digital transformation and the digitization of data,companies are i

38、nvesting in modern cloud-based data platforms so that all of their data is available in one centralized location.Companies realize that this is a critically important starting point,but being able to turn data into actionable insights means being able to structure,transform,and integrate data often

39、from multiple disparate sources and systems then publish it for consumption by AI and BI tools,including for analysis or report development and analytics development.The“Last Mile”of Data,Insights,and AnalyticsThe process of making data available and turning it into actionable insight and analytics

40、is often referred to as the“last mile”of effort because for many organizations,the last mile process is slow,manual,laborious,and inconsistent.Last mile activities involve as many as seven(7)steps to create and consume structured,integrated data so that it can be queried and analyzed,synthesized,sum

41、marized,visualized,presented,and narrated to make it relevant,understandable,and ultimately actionable and impactful.Companies seeking to increase the speed,efficiency,and reliability of actionable insights creation must find a way to increase the speed and scale of these“last mile”activities.Many c

42、ompanies turn to using a Semantic Layer to increase the speed,efficiency,and reliability of actionable insights creation while also reducing the level of effort and cost.The Semantic Layer delivers improvement in speed to insights,scale or efficiency,and costs by improving this“last mile”process thr

43、ough simplification,automation,standardization,and optimization.It does so by enabling the self-service creation of data products,including creating and publishing data products for AI and BI consumption,as well as enabling BI query optimization.Accelerating and Scaling Actionable Insight CreationSe

44、mantic Layer Solution Overview-ComponentsThe Semantic Layer improves the time to insights for AI and BI by simplifying,automating,standardizing,and optimizing how data products are created,consumed,and queried for AI and BI.The Semantic Layer consists of seven(7)components?Consumption Integration-Op

45、timizes access to data via the consumption layer for AI and BI using pre-built connections to tools like Excel,PowerBI,Tableau,Looker,and others?Semantic Modeling-Provides the ability to create data products rapidly via point-and-click data modeling that can be done through self-service by business

46、users,data scientists,and data engineers easily with no coding?Data Preparation Virtualization-Creates an easy approach for defining and reusing data transformations in a way that is transparent,easily modifiable,and operationally consistent with no coding and no data movement?Multi-Dimensional Calc

47、ulation Engine-Provides a built-in capability to rapidly perform massive multi-dimensional calculations at scale across a large number of concurrent BI tools,queries,and users?Performance Optimization-Optimizes query performance,enabling savings in cost?Analytics Governance-Provides a built-in abili

48、ty to manage data governance for data access as well as data product creation(data models,metrics),refinement,and publishing(access via BI Tools)?Data Integration-Enables rapid,automated connection to source data with no data movement required.9Semantic Layer-Position within Modern Data StackSemanti

49、c Layer-Sits Between Data Source Layer and AI/BI Consumption LayerThe Semantic Layer is a key component of the modern cloud-based data stack.The Semantic Layer sits between the curated data layer these are data sources that have been reviewed or screened for data quality and accuracy and as such,hav

50、e been approved and deemed acceptable for enterprise use.For the data sources deemed ready and available for enterprise use,the Semantic Layer makes all of those sources available for insights and analytics creation(with access and sharing permissions also being managed by the Semantic Layer).The Se

51、mantic layer creates and stores the business definition of the data often called a data product,which consists of the logical data model,data pipeline(virtualized),and metric store.The business definition is essentially a structured,multi-dimensional data set made ready for consumption by the AI and

52、 BI layer to be translated into actionable insights and analytics.10Decision SupportExpert SystemsData&InsightConsumptionDataCatalogBI Tools&ReportsAI/ML ToolsTransactional ReportsData AccessLayerSemantic LayerCurated DataData WarehouseData Lake Query EngineData Transformation&OrchestrationDV/ETL En

53、gineRaw DataRaw Data from Transactional SystemFigure 3:Semantic Layer Position Within Data and Analytics Technology StackFor more details about the Semantic Layer refer to the article “Demystifying the Semantic Layer for Smarter,Faster AI and BI”Southekal,2022.Research Study Design11The purpose of t

54、his research study is to help the data and analytics community better understand the value of using a Semantic Layer to create and deliver actionable,reliable insights that facilitate making better decisions and taking more effective actions faster with improved focus,clarity,and confidence.The rese

55、arch focuses on understanding the actual,empirical and quantifiable business impact of using a Semantic Layer to deliver actionable insights that address the“last mile”of data analytics,which makes data available for rapid insights creation and consumption.The foundation of research is to generate e

56、ffective questions that are easy to understand clear,concise and mutually exclusive and that are directed to a representative,qualified audience that is capable of providing accurate,useful answers.Against this backdrop,the survey questions were designed around three value drivers associated with us

57、ing a Semantic Layer?Speed to Actionable Insights-Reduce time to insights via improvements in efficiency and effectiveness,such as reducing the time to prepare and publish data,insights and analytics?Scale of Actionable Insights-Increase the throughput of insights creation,including the number of us

58、ers,data sources,and consistency of metrics results?Cost Savings and Reduction-Reduce costs by eliminating or automating key steps,including reducing the number of and amount of time from resources required to transform data sources into actionable insights for decision makers.To derive meaningful i

59、nsights,it was determined that the number of sample respondents should be between 30(as per the Central Limit Theorem)and 385(Data Sampling calculation based on 95 percent Confidence Level and 5 percent Margin of Error).In addition,the survey looked for qualified respondents belonging to a select,sp

60、ecialized group of experienced Business Intelligence and Analytics Developers,Data Analysts,Data Scientists,Machine Learning(ML)Engineers,Managers,Directors,Vice President and CDOs(Chief Data Officers),who have used a Semantic Layer for doing AI and BI.12The study was launched on 10th April,2022 and

61、 260 people accessed the survey across the globe in a span of 3 weeks.After validating the respondents and their credentials,108 qualified respondents were validated as being qualified to be included in the research results and detailed analysis.The percentage of respondents being qualified is in li

62、ne with the expected or desired range of respondents required for deriving meaningful insights from a survey.A summary of the respondent profiles for the 108 qualified respondents across three different dimensions or views(i.e.,company size,role,and industry sector)is shown below.Key Research Study

63、FindingsProfile of Respondents10000+Employees5001-10000 Employees1001-5000 Employees501-5000 Employees1-500 Employees12%34%12%6%35%Figure 4:Survey Respondents by Company Size13Figure 5:Survey Respondents by Role37%23%40%Middle ManagementIndividual ContributorsExecutive/C-suiteFigure 6:Survey Respond

64、ents by Industry Sector19%24%24%22%5%6%Financial ServicesConsultingTechnologyOthersOil/GasHealthcare14The survey response data was then analyzed to synthesize and present the resulting quantified results in a way that is easy to understand,enabling effective reader cognition,inference,and conclusion

65、s concerning the value of using a Semantic Layer.Specifically,the survey questions addressed three(3)performance areas?Impact of using a Semantic Layer on business performance across a variety of measures?Difference in performance using a Semantic Layer versus not using a Semantic Layer?The level of

66、 Data and analytics project implementation effort using a Semantic Layer.“The Semantic Layer provides business users with an easy way to understand the data.”“A Semantic Layer simplifies data preparation and feature creation with no/low code feature design.”“A Semantic Layer drastically reduces the

67、time to market analytics,insights,visualizations etc.to the business users.”Director,Data and Analytics,Standard Chartered BankChief Data Scientist,SiemensSenior Manager,MIS,Market Access“A Semantic Layer plays a crucial role to provide an abstract layer which can harmonize the taxonomies and provid

68、e a central hub which is aligned with data governance and data catalog.”Director,Data Architecture&Data Engineering,Loblaw Companies“This research helps data and analytics professionals discover what may be one of the best-kept secrets for improving AI and BI that using a Semantic Layer significantl

69、y improves the speed,scale,and cost to create actionable insights and analytics for decision-makers.”Chief Data and Analytics Officer,Paypal“I recommend having a Semantic Layer be part of your Data and Analytics architecture to achieve better business performance”Systems Performance Architecture,Met

70、a(formerly Facebook)“Including a Semantic Layer in your data democratization platforms translates quant information into business meaning.Instead of teams using faulty tribal knowledge or wading through many pages of technical documentation,those definitions are connected directly to the data and an

71、alytics.”Director of the Global Center of Excellence for Analytics and Data Science,Starbucks15Measuring the actual business impact of using a Semantic Layer is addressed by nine(9)metrics which were assigned to one of three main benefit driver groups:speed,scale,and savings in costs.The nine metric

72、s are listed below according to grouping:Business Impact of the Semantic Layer Improve Time to Actionable Insights(reduction in time required?Time required to create actionable insight?Time required to prepare data for analysis?Time required to develop and deploy new analytics modelsIncrease Scale o

73、f Insights and Analytics Creation and DeploymentReduce Costs to Create and Deploy Actionable Insights and Analytics4.Number of users able to perform self-service data prep,insights and analytics5.Number of data sources available for creating actionable insights and analytics6.Improvement in metric c

74、onsistency and efficiency(create once and reuse)7.Savings in cost to create and deploy data and analytics projects 8.Efficient use of resources to create and deploy data and analytics projects9.Improve ROI for data and analytics projects created using a Semantic LayerFigure 7:Overall Improvement for

75、 Speed,Scale and Savings in CostPerformance Improvementin ScalePerformance Improvement in Time to InsightsPerformance Improvementin Saving in CostOverall PerformanceImprovement4.4X4.4X3.7X4.2XImprove Time to InsightsTime to actionable insights performance is a measure of how fast an enterprise can p

76、repare and deploy actionable data and analytics using a Semantic Layer.In this regard,the survey responses demonstrate a 4.4x improvement in time to actionable insights using a Semantic Layer to create actionable insights and analytics.Speed to insights was measured across three(3)key measures,inclu

77、ding data preparation time,analytics deployment time,and time to generate actionable insights as shown below:Improvement in time togenerate insightsReduction in analytics deployment timeReduction datapreparation timeAverage Time Performance4.6X4.5X4.2X4.4XFigure 8:Speed to Actionable Insights Using

78、a Semantic Layer16Increase Scale of Insights and Analytics Creation and DeploymentScale is a measure of how much more capacity a Semantic Layer can generate in terms of users,data sources,and metrics,including metric creation and reuse.In this regard,survey respondents realized a 4.4x increase in sc

79、ale using a Semantic Layer to create and deliver actionable insights and analytics.Scale was measured across three(3)key measures,including improvement in metric consistency,increase in the number of data sources,and increase in the number of self-service users as shown below:Figure 9:Scale Expansio

80、n Impacts Using a Semantic LayerImprovement in the number of self-service usersIncrease in the numberof data sourcesImprovement in metricsconsistencyAverage ScalePerformance Value4.8X4.5X4.0X4.4X17Reduce CostsCost reduction is a measure of how much an enterprise can save by using a Semantic Layer.In

81、 this regard,survey respondents identified a 3.7x reduction in costs(cost savings)when using the Semantic Layer to create actionable insights and analytics.Cost reduction was measured across three(3)key measures,including resource efficiency,direct cost savings,and improved ROI as shown below:Figure

82、 10:Cost EffectivenessEfficiency in Data&Analytics projectsCost SavingROI from Data&Analytics projectsAverage Saving in Cost4.2X3.9X3.1X3.7X18Impact from Using a Semantic LayerImpact from Using a Semantic LayerResearch study results demonstrate an overall 4.2x improvement in business impact from usi

83、ng a semantic layer as identified and quantified by data and analytics professionals.The chart below shows the value of each variable for comparison.The results show that the level of improvement from using a Semantic Layer is consistently above a magnitude of 4x.For cost savings,the magnitude is al

84、so significant,above 3x.Further,the level of improvement demonstrates that the impact of using a semantic layer is significant and consistent across each measure.Figure 11:Business Impact of Using a Semantic Layer by Measure-MagnitudeMagnitude of Improvement in the number of self-service usersROI fr

85、om data&analytics projectsSaving in costEfficiency in data&analytics projectsReduction data preparation timeReduction in analyticsdeployment timeImprovement in time togenerate insightImprovement in metricsconsistencyIncrease in the numberof data sources0.03.01.04.02.05.019Reduction in Effort When Us

86、ing a Semantic LayerWhether analyzing retail sales data by product category,predicting the failure of a critical oil and gas asset,or helping insurers identify potential markets,it takes careful planning and execution to derive actionable insights from data and analytics using the Semantic Layer.The

87、 benefit is that using a Semantic Layer correctly can significantly shorten the cycle time and reduce the effort to create and deploy data and analytics projects.To gauge the level of effort,including the number of hours required to prepare data and derive insights when using vs not using a Semantic

88、 Layer,the survey asked the respondents the average time it takes to implement a typical data and analytics project across five(5)phases.The results showed that there is a 46%decrease in the number of hours required to create actionable insights using a Semantic Layer vs not using a Semantic Layer.R

89、esearch respondents note that a“typical”project not using a Semantic Layer would require approximately 903 hours,while doing the same“typical”project using a Semantic Layer would take only 484 hours,a reduction of 419 hours or a reduction in effort of 46%.This means that the effort would be almost h

90、alf as much!For example,if a company implements 200 data and analytics projects in a year,the direct annual savings in resource effort from using the Semantic Layer would be$8.38 million(83,800 hours of effort saved)assuming a typical hourly blended labor rate in the US is$100.The reduction in resou

91、rce effort in hours with and without the Semantic Layer across a typical data and analytics project is as shown below.The results suggest that projects not using a Semantic Layer take 46%more hours of resource effort to complete compared with a project that uses a Semantic Layer.20In addition,the re

92、duction in resource effort in hours with and without the Semantic Layer across a typical set of five data and analytics project phases is as shown below.With SlWithout Sl904 Hrs486 HrsFigure 12:Data and Analytics Typical Project Hours With&Without Using a Semantic Layer21Figure 13:Typical Data&Analy

93、tics Hours by Activity With&Without Using a Semantic LayerWith SlWithout SlData&Analytics Project PhasesWithout SLWith SL%ReductionInfrastructure1717854%Data Engineering18010940%Modelling19310844%Consumption1848653%Deriving Insight17510540%TOTAL90448646%infrastructureData EngineeringModellingConsump

94、tionDeriving Insights104 Hrs175 Hrs184 Hrs86 Hrs193 Hrs108 Hrs180 Hrs109 Hrs171 Hrs78 Hrs22The table below shows the level of effort in hours for a typical project by activity,representing the five typical activities involved and the corresponding predictions in effort.23Total Cost Savings From Usin

95、g a Semantic LayerThe survey also asked respondents to identify the total cost savings from using a Semantic Layer or a typical data and analytics project.The results show again that using a Semantic Layer reduces costs across each and every activity involving data and analytics.For example,the stud

96、y shows that using a Semantic Layer reduces data preparation cost by 18%,reducing a typical 350 hour data preparation activity effort to just 287 hours.The survey also asked respondents to identify the total cost savings from using a Semantic Layer or a typical data and analytics project.The results

97、 show again that using a Semantic Layer reduces costs across each and every activity involving data and analytics.For example,the study shows that using a Semantic Layer reduces data preparation cost by 18%,reducing a typical 350 hour data preparation activity effort to just 287 hours.“For a data sc

98、ientist,a Semantic Layer simplifies data preparation and feature creation with no/low code feature design.This enables the governed exploration of model generated insights,thereby accelerating the time to value for common and critical business processes in the enterprise.”“Having a Semantic Layer dr

99、astically reduces the time to market analytics,insights,visualizations etc.to the business users.Building a Semantic Layer is time consuming,but well worth the effort in the longer run.”“As organizations have started to concentrate on data literacy,they are planning to follow more mature models like

100、 data mesh,data fabric,or hub&spoke model.And for these models to be successful organization-wide,Semantic Layer plays a crucial role to provide an abstract layer which can harmonize the taxonomies and provide a central hub which is aligned with data governance and data catalog.”“A data consumer nee

101、ds to be able to easily discover,understand,and utilize the data.The Semantic Layer provides business users with an easy way to understand the data.“It takes time to build the Semantic Layer and familiarity with the business is highly required.But the ROI is 10 x worth it.”“I would definitely recomm

102、end to have the Semantic Layer in your Data and Analytics architecture for overall better business performance”Chief Data Scientist,SiemensSenior Manager,MIS,Market AccessDirector,Data Architecture&Data Engineering,Loblaw CompaniesDirector,Data and Analytics,Standard Chartered BankData Architect,Hyp

103、hen GroupSystems Performance Analyst,Meta(formerly Facebook)24Respondent Comments-Using a Semantic LayerThere were a total of 18 questions in the survey.17 of the 18 questions were directed and utilized numeric or Likert scale responses.One open ended question was included,which asked the respondent

104、 to explain their experience using a Semantic Layer in a descriptive way.Below are some of the actual responses from respondents who are also industry experts.Organizations need to scale business impact from their investments to modernize how they create and use data and analytics.The next step towa

105、rds showing value is to make it as simple as possible for business leaders to apply insights from data to their efforts.This requires us to look in two directions:to making data our democratization platforms speak business language and to supporting our leaders data literacy growth.Including a Seman

106、tic Layer in your data democratization platforms translates quant information into business meaning.Instead of teams using faulty tribal knowledge or wading through many pages of technical documentation,those definitions are connected directly to the data and analytics.This also shrinks the level of

107、 effort leaders must put into becoming ever more data literate.At Starbucks,being able to read,write&speak with data is a top priority.We want our data to speak the language of our business,and our business to understand and act upon insights we receive from our customers and partners.In fact,these

108、are necessary for us to meet the standards weve set for ourselves in our mission and values.In particular,we must use data to Deliver our very best in all we do and hold ourselves accountable for results.The closer we connect data and insights to a semantic layer,the more actionable our findings bec

109、ome.Ultimately,data literacy and the semantic layer are how we embed analytics in our business and create value from our data and analytics practices.Some of the benefits weve realized include the following?Expanding data and insights access so that support partners receive better feedback from our

110、stores?Improving consistency in how we use self-service data and metric?Increasing the productivity of insights and analytics creators through shared definitions,code,and dat?Solidifying our data governance practice so we responsibly acquire,share,and consume dat?Creating push and pull momentum for

111、the application of actionable insights from data.Data have become the fuel for our business strategies and decisions.Creating connections between data and a Semantic Layer and data literate leaders makes us more nimble and competitive.The business value is never in the data itself,instead it comes f

112、rom building meaningful and impactful insights that are quickly and easily acted on by our business leaders.It is necessary to build a Semantic layer and invest in data literacy to achieve the returns on data investments to which we all aspire.ConclusionA Semantic Layer Helps Data and Analytics Spea

113、k the Language of your Business25Director of the Global Center of Excellence for Analytics and Data Science,StarbucksMegan BrownDr.Southekal has worked and consulted for over 80 organizations including P&G,GE,Shell,Apple,and SAP.He is the inventor of the DEAR Model,a systematic and structured approa

114、ch for data-driven decision-making.Dr.Southekal is the author of two books Data for Business Performance and Analytics Best Practices and writes regularly on Data,Analytics,and Machine Learning in Forbes and CFO.University.His book“Analytics Best Practices is ranked#1 in the list of 100 best analyti

115、cs books of all time by BookAuthority.Org.Apart from his consulting pursuits,he has trained over 3,000 professionals worldwide in data and analytics.He is also an Adjunct Professor of Data and Analytics at IE Business School(Madrid,Spain)and CDO Magazine included him in the top 75 global academic da

116、ta leaders of 2022.He holds a Ph.D.from ESC Lille(FR)and an MBA from Kellogg School of Management(U.S.).Dr.Southekal lives in Calgary,Canada with his wife,two children,and a high-energy Goldendoodle dog.Outside work,he loves juggling and cricket.Dr.Southekal can be reached at psouthekaldbp-.26This r

117、esearch report was designed,analyzed,and prepared by Dr.Prashanth Southekal,Founder and Managing Principal of DBP Institute and Adjunct Professor of Data and Analytics at IE Business School,Spain.The Research Analyst27References?AtScale,202?BCG,Data Dominates,https:/ away:The secrets to scaling anal

118、ytics,https:/ Best Practices,Technics Publications,202?Southekal,Prashanth,“Demystifying the Semantic Layer for Smarter,Faster AI and BI”,Demystifying the Semantic Layer-White Paper AtScale,Mar 202?SurveyMonkey,202?VentureBeat,“Why do 87%of data science projects never make it into production?”,https

119、:/ 201?WEF,“How much data is generated each day?”,https:/www.weforum.org/agenda/2019/04/how-much-data-is-generated-each-day-cf4bddf29f/,2019https:/ 1:Acronyms?AI-Artificial Intelligenc?BI Business Intelligenc?CDO-Chief Data Office?CDW-Cloud Data Warehouse?CDS Citizen Data Scientist?CRM Customer Rela

120、tionship Managemen?D&A Data and Analytic?DLC-Data Lifecycl?CLT Central Limit Theore?DV Data Virtualizatio?ERP Enterprise Resource Plannin?IT Information Technolog?KPI-Key Performance Indicato?ML-Machine Learnin?PLM Product Lifecycle Managemen?ROI-Return on Investmen?SL Semantic Laye?SoR-System of Re

121、cor?SSA Self Serve Analytic?TCO-Total Cost of Ownershi?VUCA-Volatile,Uncertain,Complex and AmbiguousDBP(Data for Business Performance)Institute is a Consulting,Research and Education firm,that helps organizations leverage Digital technologies,Data and Analytics for business results.For more information,visit https:/www.dbp-.