Delivering Useful BI Tools for the Masses

A closer look at BI technologies available for operational employees. By Dr. Larry Harris, Progress EasyAsk

Today’s business pressures require operational employees to work more efficiently. As companies consolidate positions and as time and efficiency become more and more important, executives and upper-level managers are looking to empower operational employees with the information they need to do their jobs. It’s critical that employees in all ranks of an organization get access to the data they need to quickly make informed decisions that can drive actions and improve revenue, productivity and efficiency.

Traditional BI tools cater to the needs of executives and analysts. While organizations recognize the benefits of BI tools at these levels of the organization, there is a growing desire to push the benefits down beyond executives and analysts to operational users. But there are limitations in the two primary information discovery technologies in use today: Traditional BI tools are too complex for wide-spread use, and enterprise search is too simplistic to deliver the needed analysis.

Traditional BI Tools: The ‘Usability’ Issue

In spite of massive investments, it’s estimated that less than 20% of employees with BI tools can actually use them – and those who can are typically the analysts hired for such a purpose. That leaves 80% of the people without a way to gain insights that can help drive business – no doubt effecting efficiency and the bottom line.

Most large companies provide executives and management with business intelligence information in the form of intuitive and user-friendly dashboards, scorecards and reports for monitoring those data points established as Key Performance Indicators (KPI). Any trends or issues highlighted with the KPIs must be investigated via ad hoc reports or queries to pinpoint the problem or uncover the opportunity. Because traditional BI tools are so complex, business users of the dashboards or existing reports aren’t able to take that next step on their own. Even if they had the skills, time and inclination, they would still need to know where to find the data to build the reports. Analysts, of course, have less of a challenge because they understand both the tools and the structure and parameters of the underlying data.

The result is an analyst-oriented process that presents a fundamental barrier to the true deployment of operational BI across the enterprise. Even if employees or partners know whom to ask for the information, the already over-burdened analysts can’t keep up with the demand. Compound the problem with frequently misinterpreted or poorly articulated specifications, and you have a lengthy, inefficient process for fulfilling ad hoc reporting requests. At the operational level when time is critical, delays are costly – either directly or through missed opportunities.

I think we would all agree (analysts included) that analysts should focus on the most challenging and important issues of the business. If their time is strained by generating a user’s simple ad hoc report, they cannot give full attention to interrogating underlying business data and uncovering opportunities.

The solution to the problem could be the same it is for every similar resource bottleneck issue – increase access to self-service tools and in doing so, unburden the analysts and reduce the users’ time-to-report so that they can act more quickly to maximize the potential business improvement.

Enterprise Search: Simplicity at a Price

Clearly, enterprise search solutions address part of the problem because they eliminate the usability issue. They are simple to use, and they understand who the users are and what they are authorized to see. Every employee can use enterprise search tools to quickly search and get a list of documents that may or may not be relevant – wherever they reside within the organization or elsewhere in the extended enterprise.

Enterprise search looks for things that others have already created. This limitation makes it fall short on delivering answers to every employee, partner or customer. It lacks the inherent precision and ability to analyze the data and answer the question. With enterprise search, users may find reports that executives or analysts have already developed and deem important, but they won’t be able to get answers to specific operational questions.

At this point we have identified the gap between enterprise search and BI tools: BI is highly sophisticated and provides powerful analysis capabilities but remains challenging to use; search is highly intuitive and widely adopted but doesn’t analyze.

Filling The BI Gap: Usability Meets Meaningful BI Data

As an industry, we’ve figured some things out. We know how to query data stores – BI; and how to get lots of people to use something – search. To work well and to fill the gap between the difficulty of BI tools and the simplicity of enterprise search, operational BI must deliver on three key goals:

• Simplify data access for ALL
• Speed up information access
• Increase access to existing BI assets

One approach is to combine natural language ad hoc query and report search technologies to make business intelligence as easy as Web search. Using a search box as a query tool is a familiar concept to all and requires no training. This achieves the goal of simplifying data access for all.

By using a single, simple search box users can ask questions just the way they would in a business conversation. They can ask a question such as “revenue this quarter for toys”; then build on that question and enter, “Compare revenue for board games and dolls”; or narrow down the question by appending “for items made in the U.S”; or display as a graphic “in a pie chart”. Anyone in the company can pose questions of these types and get answers back in seconds.

Using this approach, business intelligence is made available to employees, partners and customers without any training or knowledge of data structures. They can get access to data and reports that are inherently specific in nature – “Show me my orders from the last quarter”, which can mean different things based upon the fiscal reporting calendar of the person asking the question; for example, an employee sales person versus a customer of that sales person.

The user can save information from inquiries in a formatted BI report, share it with colleagues, schedule to have the report distributed on a regular basis, or publish it to a repository for use in the future. Employees can share and leverage valuable information and reduce redundancy within their organizations. Some companies may implement the IT report certification option before publishing.

Naturally, if the question asked has already been answered in an existing report, it is important to surface the answers, or information assets, in addition to providing the ad hoc response. In many cases, existing BI reports incorporate additional useful insight. Greater visibility of these reports increases the return from that investment.

Implementing a Solution to Fill the Gap

A simple-to-use, natural language approach to ad hoc query opens up many possibilities. Users can find quickly find information and build or refine queries on the fly; business line managers, employees, partners and even customers can get an immediate view of business the way they want to see it. They can react more quickly and take action to improve their personal contribution to driving greater efficiency, revenue and productivity.

Financial barriers have restricted the adoption of both BI solutions and enterprise search software in the past. While addressing the complexity and usability issues is important, it is only meaningful if the resulting technology can be implemented quickly – in weeks versus months – and at an accessible and scalable price point. After all, this is a complementary solution to bring greater value to those existing investments and deliver the promise of BI to a wider audience.

About the Author

Dr. Larry Harris is General Manager and Vice President of the EasyAsk Division of Progress Software. He is internationally recognized as an expert on database systems and computerized natural language processing. Dr. Harris received a PhD in Computer Science from Cornell University and is the author of "AI Enters the Marketplace", published by Bantam Dell Publishing Group.

About Progress EasyAsk

Progress EasyAsk delivers powerful, easy-to-use software for information discovery and analysis. Used widely by industry leaders, EasyAsk leverages its natural language query technology in two distinct offerings: EasyAsk for Operational BI and EasyAsk for eCommerce. EasyAsk for Operational BI software delivers BI to the masses. It can be used by virtually any authorized user within or beyond the enterprise to easily query, analyze and share business intelligence information on demand. EasyAsk for eCommerce is the retail industry's most intuitive Website search, navigation and merchandising software. It helps merchandisers to deliver immediate ROI through increased conversion rates, sales revenue and customer satisfaction. EasyAsk is a division of Progress Software (NASDAQ: PRGS).

Contact Information

Email: info@easyask.com

Telephone: 781.280.4077

Website: www.progress.com/easyask

© 2008 Progress Software Corporation. All rights reserved. Progress and EasyAsk are trademarks or registered trademarks of Progress Software Corporation or one of its affiliates or subsidiaries in the U.S. and other countries. Any other trademarks contained herein are the property of their respective owners. Specifications subject to change without notice.