使用 DATABRICKS、THOUGHTSPOT 和 PBI 從 EXCEL 到實時洞察.pdf

《使用 DATABRICKS、THOUGHTSPOT 和 PBI 從 EXCEL 到實時洞察.pdf》由會員分享，可在線閱讀，更多相關《使用 DATABRICKS、THOUGHTSPOT 和 PBI 從 EXCEL 到實時洞察.pdf（21頁珍藏版）》請在三個皮匠報告上搜索。

1、2024 Databricks Inc.All rights reservedGET VALUE FROM YOUR DATAGET VALUE FROM YOUR DATA1FROM EXCEL TO REAL FROM EXCEL TO REAL TIME INSIGHTS USING TIME INSIGHTS USING DATABRICKS,DATABRICKS,THOUGHTSPOT&PBITHOUGHTSPOT&PBI2024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reserved2Tho

2、mas Thomas NakkenNakkenHead of development,data&Head of development,data&analytcisanalytcis2Europris since September 2021Worked in the IT industry since y2k,with different roles such as:Backend developer in TelecomArchitect and manager of R&D for SaaS EDI and insight solutionsCTO for SaaS in HSE wit

3、h distributed team in Norway,Mexico and Philippines2024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reserved32024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reservedImport goods from Asia(deep sea)One main storage facility Drive sales with a well proved cam

4、paign machine Business Business modelmodel for Europrisfor Europris42024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reserved52024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reservedSystem landscapeSystem landscape6POSSystemERPStore divisionERPSupply chainP

5、IMWMSAutomaticOrderingWebAppEDISourcingInventoryStoreOrdering SystemsStore PlanogramCampaginplanningHR planningTransportDeep seaMarketingInventoryPIMHR planningTransportLand basedInsights&Analytics2024 Databricks Inc.All rights reserved7Were building a cutting-edge cloud-based analytics platform for

6、 real-timeinsights,data-drivendecision-making and AIutilization.2024 Databricks Inc.All rights reserved8CFO:How will it CFO:How will it look like in look like in EXCELEXCEL?2024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reservedDevelopers need to master our whole tech stackBui

7、ld a team for analytics in retailBuild a team for analytics in retail9Our focus when we choose our tech stackOur focus when we choose our tech stackNo one outside dev team master SQL or pythonControllersBusiness analystDevelopersIn the futureTeam of data scientists 2024 Databricks Inc.All rights res

8、erved2024 Databricks Inc.All rights reservedAnalytcisAnalytcis platformplatform10The way we structured itThe way we structured itRAW DATA LAYERTables with SCD2Aggregation tablesPredefined reports and semantic models2024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reservedAnalytc

9、isAnalytcis platformplatform11The way we use itThe way we use itData as product2024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reservedAnalytcisAnalytcis platformplatform12The way we built itThe way we built it2024 Databricks Inc.All rights reserved2024 Databricks Inc.All right

10、s reservedDelta Sharing(open)Python libs(whitelisted)UDF not supportedOnly delta tabels as sinksLimited streaming functionality13Delta Live Tables&Unity CatalogDelta Live Tables&Unity CatalogBe aware of limitationsBe aware of limitations132024 Databricks Inc.All rights reserved2024 Databricks Inc.Al

11、l rights reservedAnalytics platformAnalytics platform14Real time analytics of sales in storesReal time analytics of sales in storesEvent hubStreamingDelta Live TablesReceipt(json)Event hubStreamingWorkflowEvent hubAggregatedsales(json)Aggregatedsales(json)ChangeeventEvent hub2024 Databricks Inc.All

12、rights reserved2024 Databricks Inc.All rights reserved15Self service insightsSelf service insightsFor the Excel peopleFor the Excel people15With the use of PowerBI we have stopped circulations of weekly Excel spreadsheets.2024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reserved

13、Enabling self service analytics to“non technical users”Use of an”self explaining”solution to give product managers insight to sales and marginsEmbed reports to store managers simple usage16Self service insightsSelf service insightsHow business users can explore their dataHow business users can explo

14、re their data162024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reserved17Aggregated data productAggregated data product2024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reserved18Data product for Data product for analytcisanalytcis2024 Databricks Inc.All rig

15、hts reserved2024 Databricks Inc.All rights reservedHighly skilled developers can create everythingDevelop code from data collection to aggregated data productsDriven by the state-of-the-art technologyAggregated tables optimizedfor readingTotal control trough CI/CD and unit testsMinimized risk for bi

16、as introduced by analystsStatic report solutions for KPI focused peopleDynamic insight tools for those who need to understandPossibility to use fine grained analytics for data scientistUse software developersDesign aggregated tables Use software developersDesign aggregated tables for easy analyticsf

17、or easy analyticsDifferent tools for different Different tools for different peoplepeople19KEY LEARING POINTSKEY LEARING POINTS192024 Databricks Inc.All rights reserved2024 Databricks Inc.All rights reservedForecasting within product categoriesForecasting within product categoriesPredict what to sel

18、l where and when by product categoriesWhat characterizes the customers who fall into different groups in RFM?What characterizes the customers who fall into different groups in RFM?Understand why customers fall into different RFM segmentsWhat measures can be taken to save the customer?Price OptimizationPrice OptimizationWas the price right what should we have done what should the right price be.Product recommendationProduct recommendation20Next stepsNext steps202024 Databricks Inc.All rights reserved21Thank you!Thank you!