Realizing Business Insights with PowerPivot 4/20/2017 1:05 PM Realizing Business Insights with PowerPivot © 2007 Microsoft Corporation. All rights reserved. Microsoft, Windows, Windows Vista and other product names are or may be registered trademarks and/or trademarks in the U.S. and/or other countries. The information herein is for informational purposes only and represents the current view of Microsoft Corporation as of the date of this presentation. Because Microsoft must respond to changing market conditions, it should not be interpreted to be a commitment on the part of Microsoft, and Microsoft cannot guarantee the accuracy of any information provided after the date of this presentation. MICROSOFT MAKES NO WARRANTIES, EXPRESS, IMPLIED OR STATUTORY, AS TO THE INFORMATION IN THIS PRESENTATION.

About the Presenter Chris Campbell ccampbell@blue-granite.com Principal Consultant with BlueGranite Certified Professional in Microsoft BI Development Microsoft Virtual Technology Specialist Business Intelligence Technical Architect at the Microsoft Technology Center, Detroit ccampbell@blue-granite.com b-chcamp@microsoft.com http://www.blue-granite.com http://twitter.com/ckcampbell

Agenda Introduction What is PowerPivot? Data Modeling for Analysis Demo Resources Questions

About BlueGranite Established 1996 in Kalamazoo, Michigan Member Microsoft BI Partner Advisory Council

What is PowerPivot?

Trends in Business Intelligence BI 2.0 Discoverability Collaboration Self Service Business Insights Data Access Structured & Unstructured Data Business Owned, IT Managed Agile Flexible Collaborative BI 1.0 Silo Silo Unstructured Data Structured Data IT Owned & Managed Slow to Respond Monolithic Not Interactive

A solution that re-defines the line between I.T. and end-user roles What is PowerPivot? A solution that re-defines the line between I.T. and end-user roles Users Model Analyze Personalize Share BI solutions I.T. Provision Administer Secure Track data ... using familiar tools without requesting support … managing usage and resources involved

Why Use PowerPivot? Process massive amounts of data in seconds Take advantage of familiar Excel tools and features Load even the largest data sets from virtually any source It is FREE! It’s sort of free…anyone with Excel 2010 can download and install the Excel add-in. To take advantage of the SharePoint publishing capabilities, you must install PowerPivot for SharePoint on a SharePoint server. The PowerPivot for SharePoint license is included with SQL Server 2008 R2 but you must consider server licensing for both SharePoint and SQL Server of whatever server you install it on.

Build Rich Interactive Dashboard Applications in Excel PowerPivot for Excel 2010 Build Rich Interactive Dashboard Applications in Excel With a few mouse clicks, a user can create and publish intuitive and interactive self-service analysis solutions.

Import and Extend Data Models Visual Data Modeling Import and Extend Data Models Create relationships between “corporate” data and external data

Import Massive Data Volumes PowerPivot for Excel: Process massive amounts of data directly in Microsoft Office Excel Broad Support for Data Sources xVelocity Columnar Compression

Interactive Slicers in Excel and in SharePoint Excel 2010 and Excel Services Interactive slicers enable users to look at the data from various directions in Excel 2010 and in the browser through PowerPivot for SharePoint and Excel Services.

Understand What Data is Important Increase IT efficiency: Familiar Technologies for Authoring, Sharing, Security, and Compliance Customizable IT Dashboard Visualize usage with animated charts IT knows author, data sources, and users, great, but what about the refresh settings? These settings seemed to bother some IT pros. Can you prevent users/owners of BI solutions from changing these settings? Operations Dashboard, how does it help IT folks to stay on top of the self-service BI game? Bring details along the lines of customizations. Audience might want to change the default reports provided. Server Status CPU and Memory Utilization System Capacity and Performance Workbook and User Activity over Time Quality of Service Query Response Times Current Server State Largest and Most Popular Solutions Solution Statistics and Monitoring Document Information and Data Sources Activity and Top Users Query Performance per Workbook Simplify management of SSBI content using IT Operations Dashboard for SharePoint

Download and Install PowerPivot Go to http://www.powerpivot.com and choose a version

Download and Install PowerPivot Make sure to select the appropriate version for your version of office (32-bit or 64-bit)

Finding out what version of Excel You Have In Excel, Click “File” then Help. The version of Excel is noted under “About Microsoft Excel”

Installing PowerPivot PowerPivot installs as an “add-in” to Excel. It appears as a new tab on the Ribbon.

Data Modeling for Analysis

Online Transaction Processing OLTP vs. OLAP Online Transaction Processing Designed for transaction based systems (Sales, Inventory, Manufacturing, etc.) Most LOB, ERP and CRM systems fall into this category Inserting and retrieving data for single or small subsets of records is very fast Summary reporting against large data sets is slower Complex data structures are difficult to understand and use by end-users OLTP Databases are “traditional”, transactional databases that are generally tied to Line of Business systems like Point of Sale, Inventory, Manufacturing, Financial, etc. There were designed with the application in mind and are very good at getting data in one record at a time. They contain all of the information we need but they are not very good at summarizing data or generating reports for large volumes of data. The structures are hard to understand by end-users.

Online Analytic Processing OLTP vs. OLAP Online Analytic Processing Designed specifically for reporting systems Built to consolidate and summarize data from OLTP systems Usually built to bulk load data in an off-line mode from back-end systems Summary reporting against large data sets is very fast Data Structures are built with the end-user in mind to allow for easy data analysis with familiar tools OLAP databases on the other hand, are designed specifically with the end-user in mind. They are meant to be bulk loaded and are tuned for fast summary reporting. Data structures are simplified with clarity being the primary goal.

OLAP Data Modeling OLAP Data Models generally take the form of a “Star” Schema Two kinds of tables Dimension Fact Fact tables have a 1:M relationship to all Dimension tables. In both types of databases, data is related between the tables in one of three ways: 1:1 means that for every record on the main table there will be only one record on the secondary table. Example: A Customer has only one home address. 1:M means that for every record on the main table, there may be more than one record on the secondary table. Example: A Customer has made many purchases. M:M means that for every record on each table, there may be more than one record on the secondary table. Example: A Customer has many Accounts and each Account may have many Customers related to it. M:M relationships are complex to model and are out of scope for this session. In an OLAP database, there are two types of tables: Fact tables contain the lowest level transactions for our purposes. For instance, purchase transactions might be considered facts. Dimension tables contain the attributes that we might want to report on. For example: Customer, Product, Store, Date, etc. Fact tables are always related to one or more Dimension tables using 1:M relationships. Each table is related to another using “Key” values.

OLAP Data Modeling We think of an OLAP data model as a “Cube” An OLAP cube allows you to “slice and dice” data in many ways. It is a data structure built on top of the OLAP database and allows us to easily manipulate the data. The cube concept gives us an easy way to visualize how the data is stored and accessed. In reality, we are not constrained to 3 dimensions but it’s not easy to visually represent more than that. There are two basic types of data in an OLAP solution: Measures are the numbers that we report “on”…Sales, quantity, cost, headcount, etc. Measures are quantitative in nature and can usually be summed (or aggregated in some other fashion…Avg, Min, Max, etc.) Dimensions are what we report “by”…for example, by Date, by Region or by Product. Combining the Measures with the Dimensions allows us to build up the reports that answer our day-to- day business questions. We think of an OLAP data model as a “Cube” Dimensions are what we report “by” Measures are what we report “on”

Data Modeling for Analysis PowerPivot blurs the line between dimensions and facts. Only requires that a user understand where the numbers are and how they want to slice them. Can join together many disparate data sources. DAX allows you to manipulate the data and build custom calculations. PowerPivot helps you build an OLAP model from your data sources so that you can perform the same types of “slice and dice” reporting that is available from cubes built in traditional Business Intelligence implementations. To build this model, you must understand the basic fundamentals of relating data.

DEMO Building a Data Model and Dashboard with PowerPivot

DAX Evaluation Context Row Context “The current row” Determined by the content of the TABLE Includes ALL columns May include “RELATED()” columns from other tables Query Context What row(s) and column(s) are selected? Determined by what are on rows/columns in your pivot table Filter Context What rows are left after the filter is applied? Row Context Query Context Filter Context Result

DAX Evaluation Context – Example Row Context Product Category Quantity Unit Price Amount 15” Tablet Computers 5 $500 $2500 17” Laptop 1 $1200 Halo 4 Games $60 Query Context Category Amount Computers $3700 Games $60 In the example above, Amount is a calculated column…the product of Quantity and Unit Price. In the first table, it is said to have Row context because it is calculated for each row of the table. In the second example, the table has be summarized by Category and restricted to just Category and Amount as would have been done in a pivot table. This is the result of Query Context. By having the pivot table summarize our data, we have altered the result and caused a calculation to occur. In the final example, we have filtered to only Computers thus applying Filter context to obtain the final result. Filter Context Category Amount Computers $3700

PowerPivot Management Dashboard DEMO PowerPivot Management Dashboard

PowerPivot Management Dashboard

Management Dashboard – Server Activity

Management Dashboard – Workbook Activity

Management Dashboard – Workbook Details

Resources

Resources: Microsoft BI/SharePoint Microsoft Business Intelligence http://www.microsoft.com/bi SharePoint BI website http://sharepoint.microsoft.com/product/capabilities/bi/ Business Intelligence TechCenter http://technet.microsoft.com/en-us/bi/default.aspx SQL Server 2008 R2 http://www.microsoft.com/sqlserver/2008/en/us/R2.aspx

PowerPivot Resources Web sites BOL http://msbiacademy.com http://powerpivotpro.com http://www.powerpivot.com http://powerpivotgeek.com http://powerpivot-info.com BOL http://msdn.microsoft.com/en-us/library/ee210682(SQL.105).aspx

Blogs SharePoint Insights blog Excel blog PowerPivot blog http://blogs.microsoft.com/SharePointBI Excel blog http://blogs.msdn.com/excel PowerPivot blog http://blogs.msdn.com/powerpivot PerformancePoint Services blog http://blogs.msdn.com/performancepoint Reporting Services blog http://blogs.msdn.com/b/sqlrsteamblog

Questions? ccampbell@blue-granite.com b-chcamp@microsoft.com http://www.blue-granite.com http://twitter.com/ckcampbell