Web Intelligence is about the analysis of information through easy ad hoc query access and intuitive interactive viewing. This combination not only empowers information consumers to explore answers to new business questions, but makes it easier on information designers as they do not need to spend time designing reports for every possible question that could arise - nor trying to guess what every user could possibly want to see.
This sounds like a great achievement, but what does this really mean?
It is often difficult to strike the right balance between guided analysis and empowering the information consumer to perform analysis on their own. As a comment on a previous post referred to, we need to understand workflows where "on report analysis" in Web Intelligence makes sense and where it creates overhead in performance and scalability in an SAP BusinessObjects Enterprise deployment.
The goal of this post is to provide a series of workflows where analysis is the central theme. Each workflow will be different in its implementation of analysis. In addition, each workflow will have its own benefits and challenges. The objective is to provide understanding behind the need to implement Web Intelligence for each workflow and then further discuss how each workflow could be improved - or how Web Intelligence could be improved to further fit the workflow.
Please keep in mind that these workflows are real customer scenarios. Some of the details have been removed to keep the customer's identity hidden. In addition, some of the workflows are still under review to determine the specific challenges being faced and how the workflows could be improved (if at all).
I ask that you review each of the workflows, think about how you are using Web Intelligence, and provide comments on how you think each workflow can be improved (or your opinion on how Web Intelligence may be improved).
Case Study #1: Web Intelligence for HR Benefits Investment Planning
The business need for this workflow is to deliver investment plan performance details to both plan managers internally and HR managers externally. This then drives those information consumers to perform further analysis on the information presented so that they can take action to make improvements in the plan or continue with the current focus.
- Data sources contain millions of rows with each report returning thousands of rows. The data being queried comes from multiple sources and together shows plan details for all employees of a given company.
- Information consumers are both external customers and internal business users / financial analysts
- Web Intelligence reports designed as templates with core objects that are important for the specific customer and answer common business questions for that customer's investments
- Web Intelligence reports make use of prompts to restrict initial data to what is important to a specific customer or set of customers
- The Web Intelligence templates (base reports) are scheduled to reduce the amount of time waiting on the data to return
- On report analysis achieved via inclusion / exclusion of objects, sorting of data, filtering of data, creating alerts, adding charts/tables alongside existing report components and performing drills on existing data
- Security controlled via rights to object folders
Benefits (Why use Web Intelligence?)
- Base reports provide the majority of what the information consumers need
- Interactive Viewing capabilities make it easy to manipulate the information in order to create links between multiple information views and derive further insight
- Web-based analysis and ease of use via right-click menu and Microsoft-like interface (important because users are familiar with performing analysis in Excel)
- Easily and quickly scalable based on server architecture
- Scheduling allows reports querying against multiple or large data sources to be refreshed unattended by the user
Challenges
- Base reports contain hundreds of thousands of rows in some cases; the queries take hours to run via scheduling
- Initial report display takes several minutes due to large number of rows displayed and formatting utilized within the document
How can this workflow be improved?
- Revisit the use case to further understand the workflow and reduce the initial number of rows returned for the base reports; make better use of analysis features of Web Intelligence
- Review the formatting of the Web Intelligence documents to ensure top performance
- Potential use of Live Office to allow for pulling Web Intelligence content into Excel
Case Study #2: Web Intelligence for Account Management
The business need for this workflow is to provide technical analysts, sales consultants and product managers with access to deployment details for customers. The base reports provide overview information for all customers. The information consumers then create customized views derived from the base reports.
- Deployment of the Web Intelligence Rich Client to allow for offline report creation and analysis
- Makes use of personal data providers in combination with corporate data to enhance view of account knowledge
- Uses base reports as templates to provide high-level view of all accounts
- Analysis performed via filters, new tables / charts alongside existing report components, and inclusion of new objects to get to specific account details
- Uses ad hoc queries to create new reports and include objects not in base report
Benefits (Why use Web Intelligence?)
- Able to combine personal and corporate data sources in one place
- Templates provide overview of information and starting point for analysis
- Quick analysis through Interactive Viewing provides insights on specific customers
- Scheduling of reports with corporate data about customers
- Ad hoc allows for determining trends by incorporating new information from the same data source or other data sources
Challenges
- Use of personal data providers can negate one version of the truth when sharing reports, but provides a catalyst for discussion
How could this workflow be improved?
- This is actually a good use of Web Intelligence. I'm sure there are some areas that could be improved; however, the workflow identified is good.
Case Study #3: Web Intelligence for Call Center Management
The business need served by this workflow is the ability to reassign headcount between departments, modify employee schedules and determine which teams are performing the best. The goal is to ensure that the call center is operating at its peak.
- OLAP data source with a Universe for a front-end
- Web Intelligence used on the Web (rather than using the Web Intelligence Rich Client)
- OLAP cube includes significant number calculations; however, some formulas used within Web Intelligence by users who want to see different types of calculations
- OLAP provides high-level information on call center, department, team, individual, product and service performance with regions, states, countries, cities/provinces/towns
- Switch to SQL based reporting for specific details on department schedules/training, team schedules/training, individual performance/schedules/training/tenure, product quality, service details
Benefits (Why use Web Intelligence?)
- Combination of OLAP, data warehouse, and OLTP data
- Use of filters, drag / drop interface for inclusion / exclusion of objects to round data in OLAP cube
- Ability to explore OLAP data in new ways by moving objects around, creating new report components and combining OLAP data with data warehouse and OLTP data
Challenges
- No challenges have been identified as this time.
How could this workflow be improved?
- Further review of the workflow is required before suggestions can be made for improvement.
What is your take on each of these workflows? How can they be improved? Is Web Intelligence the right solution? Do you have your own deployment stories to share?
