What's Hot: Reflecting on Sticky Dots at the BI Executive Summit

Wayne W. Eckerson
Director of TDWI Research

Topic: Experts in Business Intelligence

Twice a year at TDWI's BI Executive Summit, we have attendees engage in "sticky-dot" voting. In the age of wireless and electronic devices, this paper-based approach is decidedly backwards, yet just as fun and informative (see photo). Our latest BI Executive Summit took place in August, 2010 in San Diego.

Perennial Favorite. We always ask the audience, which is comprised of 75 percent BI directors and 25 percent BI business sponsors, to place a sticky dot on three technologies that will be most important to them in the next three years. At the past four Summits, predictive analytics has captured the lion's share of the sticky dot votes, and this year was no exception. Predictive analytics scored 39 of the 202 votes cast.

Judging by the number of articles and events with "analytics" in the title, this certainly seems to be the "year of analytics." Although analytical techniques and algorithms have been around for decades, we are just now reaching the point where our organizations and BI teams are ready to embrace predictive analytics. We now have large volumes of data in data warehouses, high-performance analytic platforms to process big data in a cost effective way, and executives who think "analytics" can give their organizations a competitive advantage.

What remains to be seen is whether organizations can figure out where and how to apply analytical techniques and technology and develop the internal skills and organizational structures needed to succeed in this new endeavor. So, although interest in analytics has reached a crescendo, it's unclear whether analytics will succeed in the trenches of mainstream organizations.

Emerging Technologies. What was surprising about this Summit's voting is the technologies that finished in second and third place. Visualization jumped out of nowhere to capture the second spot with 30 votes, followed closely by master data management (MDM) with27 votes, which has been simmering steadily in the background and now looks ready to boil.

Visualization. I asked several colleagues why visualization may have scored so high. Some suggested that the growing visibility of visual analysis upstarts, such as QlikTech, Tableau, and Tibco Spotfire, may account for the increased interest in visualization. These same tools are also being given a boost by hype around "in-memory processing" which delivers speed-of- thought analysis. Because in-memory wasn't a choice, perhaps visualization was the next best stand in. SAP has also spent oodles to market its "Explorer Accelerated" technology which combines in-memory technology with a visualization interface to deliver speed of thought "exploration" of subject-area data marts.

Perhaps the most satisfying explanation is that every BI vendor has spent the past two years upgrading its graphical user interface with Flash and AJAX to deliver a more modern look and feel. Thus, almost every BI customer has been bombarded by vendors eager to display their new displays, so the cumulative effect is significant.

Mastering Data. MDM has been a "hot" topic for several years, but never scored high in our informal polling. It appears MDM has moved past the hype phase to the implementation phase where organizations now recognize that they can significantly reduce the costs of running their data warehousing operation by cleaning and standardizing data before it even gets into their BI environment. To implement MDM, however, organizations need to first implement a data governance program which defines the mechanics by which the business will treat "data as a corporate asset." Unfortunately, many companies struggle data governance, especially when implementing it as a top- down program. The best data governance programs evolve bottom-up through tactical projects aimed to solve immediate business problems.

Dashboards. Performance dashboards, which finished in fourth place with 25 votes, is the highest-rated technology to experience a downward slide in the sticky-dot polling. A regular second-place finisher, performance dashboards may have split its vote with visualization, which had never been on the list before. Obviously, dashboards offer a highly visual medium for displaying data, and organizations continue to deploy them in droves (even if they are not exactly sure what to call them). However, 2000 to 2010 was the decade of the performance dashboard as companies discovered the power of monitoring key business metrics, so, it's nearly time for performance dashboards to share the limelight with other emerging technologies.

Some of the emerging technologies exhibited strong growth in sticky-dot polling compared to previous results. Top contenders are BI in the cloud (fifth place with 24 votes), analytic databases (sixth place with 20 votes), and mobile BI (seventh place with 11 votes). Other notable technologies didn't make the list (an oversight by yours truly.) These include BI search (which still needs time to percolate) and operational BI (which is going mainstream at an accelerated pace).

Agile BI. We also had a second round of voting for "methods and techniques" that would grow in importance during the next three years. Not surprisingly given the theme of our conference, "agile BI" topped all contenders, grabbing 49 votes out of 186. Agile BI was followed by perennial favorite data quality (37 votes), BI governance (34 votes), talent recruitment (26 votes), and the BI Center of Excellence (20 votes).

Given the buzz around BI Centers of Excellence, I thought this topic would score higher. However, organizations are recognizing that a BI Center of Excellence really doesn't work until the BI program has reached a mature stage. Because most BI programs are still "teenagers" according to TDWI's BI Maturity Model assessment, it makes sense that most organizations have yet to step up and tackle this topic.

The next BI Executive Summit will be held at Caesar's Palace in Las Vegas on February 14 to 16. I hope to see you or your BI executive there!

Wayne W. Eckerson is director of TDWI Research. Wayne can be reached at [email protected] .

Vendor Q & A - Jaspersoft on Open Source Adopters

Q: What trends are you seeing among open-source adopters? For example, is its use growing faster in large enterprises than in SMBs? Is growth still coming from the bottom up (that is, early adopters are usually business users who want IT to adopt the software as a company standard)?

A: Growth in the BI sector is occurring in two markets. The first of these embeds BI in other software; these are usually ISVs who want to cost-effectively include enterprise-class BI in their solutions.

Within this category, open-source BI (OSBI) adopters seem to be organizations of all sizes -- from startups to larger vendors with mature solutions -- who are either adding new products or retrofitting established solutions in order to improve market appeal, reduce maintenance hassles, and cut costs.

Although an ISV's customers influence product planning and architecture, the impetus for embedding open source BI tends to come from internal product managers -- who want to bring BI functionality quickly to market -- and from development teams that prefer open standards and Web service architectures to simplify integration, allowing them to focus on their core technology.

Within the other market category -- stand-alone enterprises -- a different pattern emerges, and size matters. Adopters tend to be mid-size organizations or departments within large enterprises (smaller organizations seem less embracing of open source software to run and manage their business operations).

In most of these implementations, IT -- at both executive and technologist levels -- drives the choice of open source BI. As new projects arise, IT teams tend to look beyond incumbent tools in the interest of cost savings or because requirements make OSBI a better choice. When companies need a modern Web services-driven BI platform, or more flexibility, or a lightweight reporting/analytics engine, or want to embed BI in a homegrown solution, they choose OSBI.

Although end users certainly help drive the move to BI in many organizations, it's IT that's leading the charge to open source BI adoption.

Answer provided by Mike Boyarski, director of product marketing at Jaspersoft

Copyright 2010. TDWI. All rights reserved.