What is a Dashboard

An easy to read graphical presentation of the current status and historical trends of an organization’s key performance indicators (KPIs) to enable instantaneous and informed decisions to be made at a glance.

Business Processes and KPIs

The key performance indicators (KPIs) of your dashboard will result from multiple business processes.

Examples of KPIs are:

• New Customers Acquired
• Revenue by Segment
• Cycle Time
• Inventory Levels

The above KPIs result from the following business processes:

• Acquire Customer
• Sell Product
• Manufacture Product
• Maintain Product Inventory

If your business intelligence project is to deliver this dashboard, you will need to completely build the analytic foundation for 4 business processes in your data warehouse. If this is your first BI project, this will be a real challenge. A common recommendation is to build your data warehouse by focusing on 1 business process at a time so that you have frequent deliverables and can show progress.

A common response to this warning is that you are only going to build summary data for these business processes so that it will be much easier and faster to build. Continue reading for a discussion on grain and atomic data.

Dashboards and Grain

If you decide to only build summary data for your business processes to support your dashboard application, you are missing out on a fundamental tenet of dimensional modeling.

You should build your fact table at the most atomic level of data possible

With summary data, the executives at your organization are undoubtedly going to ask for more details. Without atomic data, you will not be able to answer these questions. Therefore I highly recommend that you do not build your fact tables with summary data only. Summary data can be created for performance reasons, but you should still have the detailed data available.

When is a good time for Dashboards

After you have built the foundation for multiple business processes with atomic data you can tackle building dashboards. Dashboards are unique projects in that they require different skill sets for their visual and artistic nature.

For more information on business intelligence project planning for success you should buy The Data Warehouse Lifecycle Toolkit.

The post The Challenge of Dashboard BI Projects appeared first on Business Intelligence and Data Warehousing.

Create security table

The first step is to create a security table and add it to your universe data foundation. The security table has 2 fields:

1. BOUSER – the business objects user id that the user uses to access business objects
2. DEPARTMENT – the department this user should have access to

Join the security table to the relevant dimension or fact table in your data foundation

In this example the user can have access to multiple departments.

Create a security filter in your business objects universe business layer

The SQL definition is SECURITY.BOUSER = @Variable(‘BOUSER’)

On the properties tab you will need to check Use filter as mandatory in query and choose the Filter Scope Apply on Universe. This will make this filter used on all queries using this universe.

Test

Here is the query that will be created when selecting the department and department name object from the universe:

SELECT
DEPARTMENT.DEPARTMENT,
DEPARTMENT.DEPARTMENT_NAME
FROM
DEPARTMENT,
SECURITY
WHERE
( SECURITY.DEPARTMENT=DEPARTMENT.DEPARTMENT )
AND
( SECURITY.BOUSER = @Variable('BOUSER') )

In the query generated above the business objects user id of the user executing the query will be substituted for the @Variable(‘BOUSER’) statement. As a result, the user will only have access to the departments that have a row in the security table associated with their business objects user id.

Please leave a comment with your experience in implementing row level security in a business objects universe.

The post Business Objects Universe Row Level Security appeared first on Business Intelligence and Data Warehousing.

Absent or Weak Business Sponsor

A strong business sponsor is the most important ingredient for a successful business intelligence sponsor. This person needs to be well-respected within your organization and be able to make decisions for the organization. If the CIO is your business sponsor, you are in for a world of trouble. A CIO does not have influence over the business community and will not be well-respected when business decisions need to be made. Finding a strong business sponsor will significantly increase your projects probability of success.

The BI Technology to be used was purchased Prior to Gathering Requirements

Gathering requirements for a business intelligence project is one of the first steps that should influence all subsequent design and development decisions. If you have already decided on a technology platform, the technology that was chosen will influence the requirements, which may translate into business needs not being met. On future BI projects you will definitely leverage the existing technology but you should not skip this step to make sure that the technology chosen will meet the business goals.

The Business does not have Time to Participate

Business participation is mandatory in a business intelligence project. If the business is too busy to take part in the BI project, it is probably a good idea to postpone the project. A BI project cannot be successful without intimate participation by members of the business community in your organization. I would estimate that, at minimum, half of your project team needs to be from the business community.

Poor Data Quality

I have discussed the importance of data quality in an earlier post and some practical ideas on what can be done about poor data quality. Ultimately the success of your business intelligence project will rely on having quality data. It will be very difficult to develop a good framework for decision-making with underlying data of poor quality.

Lack of Data Definitions

Defining business terms for your organization is an important process. Without consistent and agreed upon definitions for business terms you will not be able to consistently give the same answer to the same question. This part of your BI project should be started as early as possible as there will be many meetings that need to be held to get your organization to come to consensus on data definitions. It will be very difficult to code your ETL process for consistency without first having consistent business definitions of terms.

I hope that this post has given you some ideas on ways to improve your next business intelligence project so that it is successful and delivers value to your organization. If you have other common reasons BI projects fail, please leave a comment.

The post 5 Reasons Why Your Business Intelligence Project Will Fail appeared first on Business Intelligence and Data Warehousing.

To understand how to diagnosis a business objects web intelligence document, you need to understand how filters work in web intelligence. Think of filters as a funnel with the top of the funnel allowing the most data through the filter and gradually reducing the data set.

Here is the hierarchy of the funnel:

1. Data Provider Query Filter
2. Report Filter
3. Section Filter
4. Block Filter
5. Individual Cell Formulas

To visualize the filters, web intelligence provides a document structure and filters panel that shows the filters applied to your document.

Here is the document structure and filters panel that produced the report data block above:

You can see that the document is filtered for:

• account level 2 equal to expense at the report level
• account level 3 equal to personal service at the section level
• account level 4 equal to benefit at the vertical table block level

Another way that data is filtered in a web intelligence document is with input controls.

View the input control below:

This input control produced the filter shaded in gray above (account level 1 equal to use of funds). Input controls can be setup to filter reports, sections, and data blocks.

I hope this helps you unravel the mysteries of business objects web intelligence filters. Please leave a comment with any comments or tips you may have.

The post Business Objects Web Intelligence Filters appeared first on Business Intelligence and Data Warehousing.

In Reengineering the Corporation, Hammer states:

Seemingly small data quality issues are, in reality, important indications of broken business processes

I completely agree.

In the Kimball Group Reader, Kimball responds by saying:

Not only does this insight correctly focus our attention on the source of data quality problems, but it also shows us the way to the solution

So when you hear people say they have a data quality problem, you know that they have a business process problem. The challenge is what to do about fundamentally broken business processes, since you are the business intelligence / data warehouse person. This opens up a whole new world for BI professionals. You may be asked to spend significant time involved in business process improvement so that the output of the business process is quality data that can lead to improved decision-making. At minimum you need to educate executives in your organization that business intelligence can not be successful without quality data. Specifically, that poorly designed business processes result in poor data quality. Hopefully the executives at your organization will understand and focus efforts on improving business processes. Only then can a successful business intelligence project begin.

You have to be careful to do due diligence on every candidate business intelligence project so that you have confidence that you have quality data to work with. If you find that you don’t, you need to educate the sponsors of the project on the importance of quality data, and why this candidate BI project should be put on hold until the business process issues are resolved. You just have to make sure you don’t say “no” to every candidate BI project. If you do, you may be out of a job.

If you find that poor data quality is a constant problem at your organization you may want to acquire some skills on business process improvement so that you can help solve the problem. A great book on redesigning business processes is Workflow Modeling: Tools for Process Improvement and Application Development.

The post Data Quality and Business Process appeared first on Business Intelligence and Data Warehousing.

• units_sold is a fact
• date, store, and product are dimensions
• all the other data elements are attributes of a dimension

Dimensional models are typically represented using a star schema.  Below is a very simple star schema.  However, star schemas are typically built for simplicity to aid in understanding.

The benefits of dimensional modeling are:

1. Understandability – The dimensional model is very understandable by the business. In addition, this model is completely based on business terms, so that the business knows what each fact, dimension, or attribute means.
2. Query performance – Dimensional models are denormalized and optimized for fast data querying. Many relational database platforms recognize this model and optimize query execution plans to aid in performance.
3. Extensibility – Dimensional models can easily accommodate change. Dimension tables can have more columns added to them without affection existing business intelligence applications using these tables. Fact tables are also extensible, but need more planning when adding new dimensions so that they do not disrupt existing queries that are using the fact table.

One of the best books on dimensional modeling is The Data Warehouse Toolkit: The Complete Guide to Dimensional Modeling.  I highly recommend that you buy a copy if you are interested in learning more about dimensional modeling.  Over the next few months, I will be discussing the process of developing a dimensional model. Feel free to bookmark this site if you are interested in following along or choose one of the follow options listed at the top of the navigation sidebar on the left side of this page.

The post What is Dimensional Modeling appeared first on Business Intelligence and Data Warehousing.

My favorite definition is provided by Forrester Research.

Business Intelligence (BI) is a set of methodologies, processes, architectures, and technologies that transform raw data into meaningful and useful information used to enable more effective strategic, tactical, and operational insights and decision-making.

This definition of business intelligence includes data warehousing as part of business intelligence. This is the primary reason I like the Forrester definition.  I don’t believe that BI can be done in isolation without considering the data warehouse processes.

Here is a list of things that business intelligence isn’t:

• BI is not just a Data Warehouse
• BI is not just Dashboards
• BI is not just about Technology
• BI is not just for Executives/Leadership
• BI is not just for Large Institutions

There are many different types of business intelligence applications.  Here is my list of the 5 primary types of BI applications:

1. Standard Reports – Reports with the ability to enter a parameter, drill-down to lower level of detail, and link to related reports
2. Ad Hoc Query Access and Reporting – Often used as the development tool for standard reports
3. Dashboards and Scorecards – At-a-glance views of key performance indicators (KPIs)
4. Data Mining – Data exploration with the intent to find patterns or relationships that can be made useful to the organization
5. Operational Intelligence – Real-time insight into business operations

A great book you should buy for more information on business intelligence is The Data Warehouse Lifecycle Toolkit.

The post What is Business Intelligence appeared first on Business Intelligence and Data Warehousing.

“What should I do to learn how to develop ETL mappings using Informatica PowerCenter?”

Here are the steps that I would take to learn how to do extract, transform, and load with Informatica PowerCenter.

I assume that you already have knowledge of SQL. If you don’t know SQL or need to learn more about SQL, I would recommend that you buy Learning SQL by Alan Beaulieu.

You also should have access to the Informatica PowerCenter tool in your environment. If not, you should get the Informatica server and client tools installed on your desktop.

A great place to start to learn Informatica is the Getting Started Guide found in the documentation folder of your PowerCenter client tools installation. This guide is really good at teaching you the basics. You will need to have access to create tables in a database, as the guide teaches you by loading some sample tables/data and teaches you all the basic ETL concepts using the sample data.

The tutorials in the guide will teach you how to:

• Create Source Definitions
• Create Target Definitions
• Create Target Tables
• Create a pass-through Mapping
• Create Sessions and Workflows
• Use the Workflow Monitor
• Create Aggregate Transformations
• Create Expression Transformations
• Create Lookup Transformations
• Create Filter Transformations

Make sure that you do all the tutorials and not just read them. You will learn the concepts better by following the tutorials step by step. All of the tutorials are very basic, but they will give you the foundation you need to develop ETL processes using Informatica PowerCenter.

The next guide I would skim is the Transformation Guide. This is a great reference guide to all the transformations available in Informatica PowerCenter. I would not suggest you read the guide from cover to cover, but to instead read the Working with Transformations chapter and the overview section for each Transformation. This will give you a good idea of the transformations available that you can use in your mappings, so that when you are trying to develop a mapping you will have some background on what transformations are available. When you need to do a new transformation for the first time, you can read the entire section of the Transformation Guide for that transformation for more details.

Next, you should actually develop your first mapping in Informatica PowerCenter.  There is no better training than actually doing it.

A great book you should buy for learning the Extract, Transform, Load process is The Data Warehouse ETL Toolkit: Practical Techniques for Extracting, Cleaning, Conforming, and Delivering Data.

The post How To Learn Informatica appeared first on Business Intelligence and Data Warehousing.

I have tested this on the following Mac environment:

• Mac OS X Lion 10.7.3
• Java Version 1.6.0_31-b04-415

There are 3 different clients available to use Web Intelligence. Only the Web and Java clients can be used on a Mac. The desktop client is a Windows only installation. I used the following web browsers to test Business Objects Web Intelligence on Mac OS X:

• Safari 5.1.5
• Firefox 11.0
• Chrome 18.0.1025.163

Of the above web browser, Firefox is the only browser that provides a decent user experience for Web Intelligence. The major issues with the other 2 browsers is that they don’t support drag and drop to move blocks, columns, and add new objects to existing blocks. Also they don’t support creating new documents in the java client and editing data providers. The java client is not great in Firefox, but it does allow you to create new documents, and edit data providers. All other functions that you need to do should be done in the regular web client.

The SAP Business Objects version used for this analysis was Version 4.0 Support Package 2 Patch 13.

In conclusion, if you are going to use SAP Business Objects Web Intelligence on Mac OS X, you should use Firefox as your web browser. I hope this has been helpful to you. If you have any questions or helpful advice to share others about your experience using Web Intelligence on Mac OS X, please leave a comment.

The post Business Objects Web Intelligence Mac OS X appeared first on Business Intelligence and Data Warehousing.

1. Web
2. Java – Rich Internet Application
3. Desktop – Desktop (Windows only)

Each of the Business Objects Web Intelligence clients have unique features that are not available in the other clients.

There isn’t a Business Objects Web Intelligence client that has all the features.

The only difference between the Web Intelligence Rich Internet Application (Java) and the Web Intelligence Desktop client is that in the Windows Desktop client you can use excel, text, and web services as a data provider.  If you don’t need these data provider sources, the java client has all the functions you need.

The Web client does not allow you to create new documents and data providers.  However there is one significant feature of the Web Intelligence web client in that it allows you to create document links between documents.

This analysis is with Version 4.0 Support Package 2 Patch 13 of SAP Business Objects Web Intelligence.

If you know of a feature difference that I missed, please leave a comment.

The post Business Objects Web Intelligence Client Differences appeared first on Business Intelligence and Data Warehousing.