BI Dashboards with SQL Server 2008 R2
0 likes3,106 views
The document discusses the implementation and capabilities of Microsoft Business Intelligence (BI) tools, particularly through Microsoft SQL Server 2008 R2 and related platforms like Excel and SharePoint. It emphasizes the integration of BI with organizational processes to enhance decision-making, user collaboration, data visualization, and reporting functionalities. Key topics include dashboard design principles, user experience, and the importance of effective visual communication of data for business users.
BI Dashboards with SQL Server 2008 R2
- 2. Using Business Intelligence (BI) Introducing Microsoft BI Delivering BI with Microsoft SQL Server 2008 R2
- 3. A category of methodologies and technologies for gathering, storing, analyzing and providing access to data to help enterprise users make business decisions.
- 4. Summarize business critical operations Purchasing, finance, investments, etc… Used at multiple organizational level Executive, Accounting, Personal Often embedded in larger processes
- 6. To improve organizations by providing business insights to all employees, leading to better, faster, more relevant decisions • Microsoft has a long-term commitment to delivering a complete and integrated BI offering • SQL Server has led innovation in the BI space for more than a decade • There is widespread delivery of intelligence through Microsoft Office • The platforms are enterprise-grade and affordable
- 7. DELIVERY SharePoint Server Reports Excel Dashboards Workbooks Analytic Scorecards PowerPivot Views END USER TOOLS Excel PowerPivot DATA PLATFORM SQL Server SQL Server Reporting Services Analysis Services SQL Server DBMS SQL Server Integration Services StreamInsight Events
- 8. Familiar User Experience Business User Experience Self-service Access and Insight Data Exploration and Analysis Predictive Analysis Data Visualization Contextual Visualization Integrated Content and Collaboration Business Collaboration Platform Thin Clients Dashboards BI Search Content Management Compositions Information Platform Information Platform Analysis Services Reporting Services Integration Services Master Data Data Mining Services Data Warehousing
- 9. Specialized BI for BI Tools Everyone
- 10. Database Integration Analysis Reporting Master Data StreamInsight Engine Services Services Services Services Relational Classic Managed Data OLAP Data ETL Reporting Management Management Processing Complex Self Service Master Data Replication Event Data Mining Reporting Management Processing Data Profiling PowerPivot* Full Text Embedded Self Service Search Reporting Analytics
- 12. Cubes in SQL Server 2008 R2
- 13. Provides an interface familiar to most information workers Supports basic, advanced, and embedded analysis Integrates with SQL Server and Analysis Services as well as other data sources Includes a new add-in for highly-performing data integration and analysis Enables collaboration through server workbooks shared across the Web Supplements portals and dashboards
- 14. Manages compressed storage of, and extremely fast querying and reporting from, large volumes of related tabular data Data can be loaded from a variety of data sources Leverages familiar Excel features Uses Data Analysis Expressions (DAX) for sophisticated transformation of data Supports publication of PowerPivot workbooks to SharePoint
- 15. In-memory Data Cubes in Excel Excel addin allows creation of cube Cube persisted in the Excel Workbook Consumers of workbook have access to cube Excel Services Gemini Browser Excel Workbook
- 16. Server hosting of Excel calculation engine Provides a web thin client for Excel workbooks Excel aggregates data from multiple sources Exposes data using REST and JavaScript Browser Excel 2007 Custom applications
- 17. Excel Services exposed using Web Parts Web Parts display named items Parameters exposed as Web Part connections Parameter changes will update charts Allows real time feedback to users
- 18. Creating a BI Dashboard with Excel 2010
- 20. Performance Status Chart Web Point Indicators Parts Services Excel Visio BI Search Services Services
- 21. Simple web parts that provide visualization Quick summary of data Can access multiple data sources Fixed Values Lists and External Lists Excel Services Other Web Parts SQL Analysis Services Lists and External Lists Excel Services
- 22. Context-driven dashboards across systems Provides transparency and accountability Interactive access using browser Can be created/updated by IT professional
- 23. Professional report authoring environment Designed for developer or IT professional Does not ship with SharePoint 2010 SP 2010 has Reporting Services Web Parts Used to create professional reports for BI apps
- 24. Extends the reach of search to BI elements Adds Excel and Reporting Services to search Presents results based on dimensions Links directly to specific view of the report
- 25. Evolution of the Reporting Center template Predefined lists for PPS, Excel Services Starting point for BI portals
- 27. Visual displays of mission critical analysis Answers fundamental business questions Single screen display of information “Real Time” summary of data Interactive links to details
- 28. Provides an overall view of status indicators Modeled after the business not the data Manages key performance indicators (KPIs) Often included as part of a dashboard Balanced Scorecard Certified
- 29. Dashboard designer used to create content ClickOnce application launched from browser Manages content directly in SharePoint Lists for content, data sources, dashboards
- 30. Data access is managed with Data Sources Data is accessible from multiple sources Ex. Analysis Services, Excel, Lists, SQL, etc… Accessed with service account or current user Stored in list as an XML file (*.ppsdc)
- 31. Business users view content in the browser Stored as Web Part pages Web Part connections used to relate content Ex. Item choice in a scorecard updates reports
- 32. Decomposition trees allow data exploration New in SharePoint 2010 Available by right clicking data in browser Analyze -> Decomposition Tree Allows filteringexpanding across dimensions
- 35. Visual Display of the most important information needed to achieve one or more objectives which fits entirely on a single computer screen so it can be monitored at a glance Information Dashboard Design By Stephen Few
- 36. Some are Better Than Others
- 37. Specific – Clear on what is being measured Measurable – Accurate and consistent Actionable– Influence to alter the outcome Relevant– Significance and leverage Timely – Appropriate intervals
- 38. Key Data/Metric s In Context Dense Zero Information Learning Not Curve Cluttered Single Page Graphically Oriented Access to Highlight Supporting Exceptions Details
- 39. “Single Page” “Graphical” No Scrolling Displays Patterns Nothing Hidden Abstract Data Linked Content Minimal Grids Better Dashboard All Relevant Info Design Domain Aware One Place to Look Leverages Not Data Driven Knowledge “Disparate Data” “Guided Analytics”
- 40. Key Data/Metrics High level performance metrics Generally tied to goals and bonuses Focus on actionable data Disparate sources Context The 3Ts = Target, Trend, Typical Similar/Related Data/Metrics Basis for interpretation
- 41. Dense Information All relevant data/metrics Different perspectives Not Cluttered – “At a Glance” It is not how much, but how well Easy to view,The Visualan eyeTufte Information not Display of Quantitative Edward R. chart Data-Ink Ratio ( ) Non-essential ink should be removed Supporting ink (axis lines) = low visual impact
- 42. Dashboard viewing is not a person’s job Oriented at casual users, line mgt, & execs Little/No accumulated experience Each viewing is the first time Dashboards are not analytical tools But they provide guided analytics Targeted to a Person/Role/Problem Set
- 43. Outliers should POP out Visual Cues Highlight Both Color Problems Size Opportunities Shape What is an Exception? Boldness > 1 Std Dev, 2…. Five9s Position When action needs be taken Icons
- 44. Dashboards are not an Associated end unto themselves Layers Contextual Guided Content Updates Analytics Update Dashboard Webparts Links to New Pages Hierarchical Drill-Down Root Cause Analysis Ad Hoc Analysis Detail Reports
- 45. Scorecards Diagrams Grids Charts Gauges Maps Tree Map Decomp Tree And So Much More……
- 46. Color Use Color Sparingly – Generally Soft Colors Highlighting – Bold Colors Different Colors = Different Meaning Try shades of grey Use a Single Hue Increase the intensity for low to high values Remember - 7% of men are color blind A dashboard is not a painting Usefulness is Paramount
- 47. When to use: Values are familiar Need to lookup a value Small differences are important Mixed units of measure Design Considerations Try to keep square-ish Comparisons are done by column Limit length with Top/Bottom
- 48. Highlights Items of interest Data Bars Helpful to read magnitude Color Scales/Icons Icons Can overwhelm the data are best used to: Arrows Replace Used for change over time data to save space Show variance or score
- 49. Are a special type of grid Created for KPIs and Objectives Can be methodology oriented Balanced Scorecard, 6 Sigma, … Used as a navigation aid A list with data Scorecards are less about data and more about process
- 50. Score Catalog Budget Audited Plan Forecast Accurate Scorecards Tolerances Benchmarks Approved Perspectives Initiatives KPIs Organize Metrics Measures Data
- 51. Lines Charts – Time Series oriented Highlights trend, patterns or variability Highlights relationships between series Trend lines & projections can be added Bar Chart – Comparison Oriented Categorical comparisons within a dimension http://www.symcorp.com Nominal comparisons across data (sales to cost) Pie Charts – Part to Whole Only works with small sets Hard to visually quantify A lot of space for not much data
- 52. If the dashboard is not visually appealing, 1 Page – 6#s people will not go to it If the dashboard is not effective and practical, people will not use it Pretty Yes! Useful? What is fun the first time Might be boring the 100th time Too Dense?
- 53. Horizontal Long labels Pattern comparisons Long-ish lists Vertical Groups with in groups Stacked and 100% Bars Pareto Charts Doesn’t work for large numbers of groups Bars are easily drilled into
- 54. Effective Great Mixing Of Data-Ink Charts & Ratio Trellis Values Small Charts Multiples Introduction of Sparklines & Bullet Graphs
- 55. A bullet graph is a variation of a bar graph developed by Stephen Few. Seemingly inspired by the traditional thermometer charts and progress bars found in many dashboards, the bullet graph serves as a replacement for Symmetry’s dashboard gauges and meters Variation Sales Costs Black Bar = Value Green/Red Bar = Target Blue Pointer = Last Year
- 56. Useful in showing regional components Maps are great for demographics Impact grids can be effective Be careful of charts within a map
- 57. When using a set of Hard to set gauge scale gauges expected value for the general case should be straight up Best for continuous process Poor Data-Ink Ratio
- 58. Balanced Scorecard Process Diagram Strategy Map Diagrams are underutilized Graphical Representations Impact Diagram Root Cause/Fishbone Diagram
- 59. Show the data Focus on the substance of the numbers Don’t distort the data Present many numbers in a small space Encourage comparisons Show data in several perspectives The Visual Display of Quantitative Information Edward R. Tufte
- 60. Number of Frames Up to 4 is good, no more than 6 Frame Proportions Size = implied importance Uniform otherwise Location Top left = primary focus Bottom right = supporting detail Off page = irrelevant Printing The dirty little secret
- 61. Different frame on same page Must fit on the page Can drill from here to new page New page in same browser Easy to get back Full page available New browser tab/window Allows for comparisons Users must close windows/tabs
- 62. A tiered set of dashboards based on role or function A set of grids, graphs, or visualizations that examine a single metric A set of detailed reports or ad-hoc analytics http://www.symcorp.com
- 63. Identify the key dimensions for the metric Customer, product, department Drill into one or more of these Top 10 rankings Identify the audience Marketing manager – Drill to product Sales Manager – Drill to customer
- 65. • Business Intelligence (BI) Primer • BI Tools in SharePoint 2010 • Excel Services • Performance Point Services
- 67. Show Me the Numbers: Designing Tables and Graphs to Enlighten Stephen Few Information Dashboard Design: The Effective Visual Communication of Data Stephen Few The Visual Display of Quantitative Information, 2nd edition Edward R. Tufte http://www.perceptualedge.com/ http://dashboardspy.com/ Symmetry Corp – www.symcorp.com