Register for "A Guide to Data Products: Everything You Need to Understand, Plan, and Implement" - Friday, May 31,1:00 p.m. ET

Part III: The Holy Grail of Enterprise BI - User Adoption

This is the third in a four-part series on enterprise business intelligence (BI). 

Read Part I - The Battle for BI is Over: Now the Hard Part Begins

Read Part II - Ten Characteristics of a Modern Enterprise BI Tool

User adoption is the holy grail of BI. All BI managers want 100% adoption of their BI tools, adoption rates still hover between 25% to 30%. Clearly, most organizations have a long way to go to make BI pervasive and a key driver of business value.

There are dozens, if not hundreds, of factors that contribute to user adoption. Some are technical, many are not. But BI managers must get each one right to ensure high levels of BI adoption.

If there is a common thread among the factors driving BI adoption, it’s knowing your business users—what they do, what information they need, what incents them and how they make decisions. Without this knowledge, BI managers can never deliver products or solutions that meet their needs.

Casual and Power Users. Business users fall into two camps: casual users who use information to do their jobs, and power users who are paid to analyze data. These two groups have very different information requirements. This article present ten factors that drive adoption for casual users and power users.

Adoption Keys for Casual Users 

Casual users want their data down and dirty. They want BI tools that are intuitive to use and give them the data they need, when they need it, and how they want it—and nothing else. Here are five adoption keys for casual users.

1. Fast Performance. In the age of Google, casual users expect immediate response to on-screen clicks and gestures.  If users have to wait ten seconds or more on a regular basis, they grow impatient and stop using the BI tool. Successful BI programs track query response times and seek to average less than three seconds per click. This is a tall task for complex dashboard displays that encapsulate dozens of queries per screen. But thanks to in-memory processing scale-out computing, and columnar databases, fast response times are possible and achievable.

2. Trustworthy Data. Casual users won’t use a BI tool if they feel it delivers inaccurate or incomplete data. They will force BI managers to validate or reconcile the BI data with the other sources of data before they trust it, a laborious process that can take months or years. Integrating business-oriented metadata with a BI tool can remedy this problem, allowing users to view the origins and lineage of data, among other things. BI teams can also apply a watermark or seal to reports containing accurate, trustworthy data so users can differentiate between spreadmart and IT-validated data.

3. Actionable Dashboards. Casual users need a BI tool that gives them the information they need at a glance to monitor core processes and take immediate action. Actionable dashboards fill this role perfectly. Tailored to each user’s role and processes, an actionable dashboard enables users to quickly view performance at a top level and then, if desired, drill to lower levels of data to identify the root cause of a problem. This ability to follow an issue to its cause makes a dashboard actionable: it not only tells users the problem, it prompts them to get more information and then craft a solution to address the issue.

4. Modifiable Reports. Although most casual users don’t author reports from scratch, they often want to modify an existing report or dashboard by re-sorting a column, pivoting a dimension, changing a chart type, or even adding a new calculated column, total, or chart. This is the definition of self-service BI for casual users. In some cases, casual users might want to pull objects from a library—predefined charts, dimensions, filters, and metrics—and assemble them into new views, as long as the “assembly” happens auto-magically so objects recognize each other and link together without scripting or user input.

5. Mobile Delivery. Casual users have become big fans of mobile technology—and they expect reports and dashboard to follow them wherever they go. This puts a premium on creating mobile BI applications that are task centric and map to the way users perform their jobs. Screens must be simple, workflows intuitive, and interactivity obvious. Even better than a mobile dashboard is an alert that notifies a casual user when there is something important to view in a report or dashboard. Here, there is no need to continuously check a dashboard for relevant information—the information finds the user.

Adoption Keys for Power Users 

Unlike casual users, power users need wide-ranging access to data and a complete spectrum of BI features and functions to answer ad hoc questions from business executives and managers. Here are five adoption keys for power users.

1. Data Mash Ups. To respond quickly to questions from business managers, power users need to connect to any data source or application, quickly ingest the data, and then format, clean, transform and integrate the data so they can analyze it. Traditionally, this mash up process consumes the lion share of an analyst’s time, leaving scant time to analyze data. In response, BI products now incorporate data preparation or blending functions that expedite the creation of data mash ups. These functions use tacit rules and machine intelligence to identify data types, link tables, and suggest transformations, taking much of the guesswork and labor out of preparing data for analysis.

2. Visual Analysis. Although Excel was once was the tool of choice for power users to both prepare and analyze data, it has been replaced by a new breed of BI tool called a visual analysis or visual discovery tool. These tools enable power users to visually plot data on a canvas so they can rapidly discover trends, relationships, and patterns in the data. The tools are highly interactive and support a wide range of chart types, including maps and animations. As a bonus, many of the tools generate attractive dashboards that meet the needs of departmental colleagues. This is one reason why visual discovery tools are the fastest growing segment of the BI market today. (See "Part I: The Battle for BI Is Over: Now the Hard Part Begins.”)

3. Analytic Functions. Besides visualizing data, power users also want to analyze data. Beyond simple sums and averages, they want to apply mathematical, statistical, financial, forecasting, and machine learning algorithms to tease key patterns and relationships out of the data. At the very least, power users need to forecast sales, taking into account seasonality and past performance. Many BI tools now offer a point-and-click calculation or expression engine and can import models created with statistical tools, such as R or SAS, to augment their own calculations.

9. Reusable Logic. Power users often need to create custom business logic, such as a custom calculations, data sets, groupings, or hierarchies. Rather than wait for the IT department to create the logic in a database or ETL tool, power users need to create this logic in the BI tool, store it in a library, and then reuse it in subsequent applications, if desired. In addition, BI tools now support personal sandboxes where power users can mingle their own data with corporate data and apply their own custom calculations. Power users can only share their analyses with authorized users to prevent the proliferation of spreadmarts.

10. Content Sharing. When power users discover something, they want and need to share it. Effective BI tools support a wide range of publishing and collaboration features, including selected sharing, storytelling, annotation, and social media techniques, such as personal timelines, comments, likes, and ratings. BI collaboration saves power users time and improves the quality of their analysies. Rather than reinventing the wheel, power users can see and leverage each other’s work, becoming more efficient and productive. At the same time, the BI team can watch what the power users do and bake their information needs into the next iteration of the data warehouse environment.


Organizations can deploy the most advanced BI tools and design the most elegant BI architectures, but if business users don’t use BI tools to analyze data, make decisions, and take action, then all this money and effort is for naught.

The keys to BI adoption are numerous, but vary by type of user. Casual users need BI tools that are fast, easy and intuitive, while power users need BI tools that are powerful, flexible and functional.

Read - Part IV: Seven Keys to a United BI Environment

Wayne Eckerson

Wayne Eckerson is an internationally recognized thought leader in the business intelligence and analytics field. He is a sought-after consultant and noted speaker who thinks critically, writes clearly and presents...

More About Wayne Eckerson