Gartner Research Director Toby Bell offered a few ways to tame the proliferation of e-mail use in his presentation Thursday "The Information Explosion and What to Do About It."

The first step is to understand that most e-mail is not worth keeping. It's about 98% junk, including duplications and personal correspondence, and 2% important stuff, such as HR records, legal documents and client information.

Enterprises then need to impose discipline on e-mail. Many enterprises are hesitant to enforce policies for purging old e-mails or setting limits on mailboxes. Users whine about size restrictions or archiving, and whiners win, Bell said. People think that their worthless e-mail is actually important. Stop that cycle by bringing in an expert if you have to who can identify valuable stuff and set a policy to get rid of the junk.

"If you don’t have good guidance in taming e-mail, you will continue on as you have done, which is bad, or you will keep everything, which is worse," Bell said.

Other tactics to cure the e-mail addiction are:

· Discourage the use of wireless e-mail devices. Most people do not need to be connected all the time, but those who are connected can make the disconnected feel they are out of the club. Allow for reasonable lags in response time to e-mails.

· Try open-source technologies and portals. Many users like them.

Submit a Comment | Permalink

Research VP Michael Smith urged organizations to choose metrics that really drive business value in his talk "How to Define the Right Metrics for Corporate Performance Management" this afternoon in Swan. But he warned IT professionals against trying to claim that benefits, such as better sales cycles, are solely the result of IT.

"That will put you at odds with the people you work with," Smith said, advising IT leaders to remember that a combination of things can lead to business benefits.

A big benefit will come from presenting performance management as a united-team approach. High-level executives like to see IT people, business people and finance people collaborate. They will reward that behavior by following the direction the group recommends to improve business process. They also may provide more funding.

"Don't claim the bread crumbs," Smith said. "Just be part of the team, and the rewards will come."

Submit a Comment | Permalink

These burning questions emerged at the business intelligence workshop Monday:

• How do we get started with BI?
• How do we get from a culture of "get it done" to one that plans?
• How do you deal with a competency center that does not engage IT?

About a dozen participants in the workshop then tossed around ideas to answer these questions.

Among the ideas were:
• To get started, hire a consultant with expertise in your industry and consider starting with simpler areas, such as financials.

• To get it done, BI managers should have a sponsor in the business and should educate the sponsor about the value of BI.

• To engage the competency center, let it participate in BI efforts and let it have some of the responsibility for the outcomes.

What do you think?

Submit a Comment | Permalink

Research VP Mark Gilbert stopped by the Blog Lounge earlier today to discuss his presentation, "Towards a New Workplace Architecture: The Acid Test." He made a couple of points:

Microsoft productivity tools still totally dominate the productivity tool landscape, at least domestically. Everyone in the session said their organizations use Office. There were questions about open source alternatives, but no users. That's not the case in Europe, Gilbert said.

Architecture specialists, not surprisingly, are overwhelmingly (95%) deployed in central IT staff, not within business units.

Gilbert plans to discuss architecture issues further in Wednesday's presentation, "ECM: Content Infrastructure in Your Future."

Submit a Comment | Permalink

Gartner Research Director Kurt Schlegel used two real-world case studies to show business intelligence (BI) leaders and prospective BI users some ways BI can bring real value to businesses.

Schlegel discussed how the Oakland A's baseball team used BI to make it to this year's Major League Baseball playoffs. The A's are a small-market team that does not have nearly the resources of a large market team, such as the New York Yankees. Small-market teams normally do not win many games, since they cannot afford star players, but the A's found a way to build a team and play the game to maximize what it could afford. Among the A's tactics were:

• Using different metrics to measure players' performances, such as on-base percentage instead of batting average, to find underrated players it could sign at lower prices


• Using decades of statistics to debunk the myth that teams must take many risks to win games, such as stealing bases


• Ultimately, the A's BI effort allowed it to predict how many games it would win, and how many games its competitors would win, to come up with a strategy for the season.

Euro Disney also used BI to predict ways to improve its business. A major problem at Euro Disney was lines - people who waited in lines to see attractions, eat or check in to hotels were often unhappy and were not spending as much money as they would if their time was not spent waiting. Euro Disney used BI to:

• Generate real-time operations level reporting that determined how long the lines were and generated alerts if lines got too long.


• Predict where lines would get longer, so it could deploy staff to reduce lines in busy areas.


• Check its predictions against actual line-length data to ensure its predictive model was working properly.

"The result is people are more satisfied, people spend more money, people recommend Euro Disney, people come back year after year," Schlegel said.

These examples show direct correlations between BI and the strategic goals of enterprises. These examples demonstrate the vision of how BI will affect business transformation.

Submit a Comment | Permalink

At this year’s Symposium, I am excited to have the opportunity to present a new scenario for the business intelligence and information management area (see Future of Business Intelligence and Information Management: Activities, Competencies and Organizations). This research is based an a lot of the conversations I am having with clients and my Gartner peers regarding the increasing interest in supporting new business models and people with business intelligence, performance management, content management, data management, data quality, collaboration and workplace applications.

I am finding three key — but diverging — shifts that are driving client interest:

1. More dependencies between business models: So, as an IT Leader, you are probably finding that you are having to balance more complex business processes and are more dependent on other people and their defined business processes.

2. Simpler user interfaces: The knowledge, expectation and access to technology for the average worker — not just knowledge workers — are increasing dramatically. To add to this, the most favored way of communicating with people and applications is via search, instant messaging interfaces, e-mail and visual tools.

3. More-complex technologies and architectures: With the advent of service-oriented architecture (SOA), diverse data and content stores and a plethora of tools and applications, the business intelligence and information management applications and infrastructure are becoming more granular and complex.

So, we are dealing with more complexity as business and IT people. But, our users are driving more simplicity. This might be good for business and people, but it certainly leaves the IT leaders in a difficult position.

If you agree that this is where the world is going, what can you do about this?

For one thing, we need to think differently about how we approach supporting our businesses, users and processes. Turn your thinking upside down. This means thinking of business, people and processes before technology (see "Key Steps to Making Business Intelligence Strategic").

Also, understanding the people and processes that have the highest impact on your business strategy is critical for your sanity and success. Otherwise, you risk spending time on people and projects that don’t impact the business.

Last, but not least, it is important to have a methodology for approaching the support of your business (see "Activity Cycle Overview: Business Intelligence and Information Management,") in a logical way.

The role of an IT leader is changing from that of “providing and implementing technology” to that of “trusted advisor.” This is a great opportunity for you — if you take the time to understand what is changing and how you can get ahead with a clear methodology and approach to strategically support your business, people and processes.

I look forward to seeing many of you at Symposium. If you will not have the opportunity to be there, feel free to schedule an inquiry with me to discuss this research and how you might take advantage of these shifts.

1 Comment | Submit a Comment | Permalink

It’s not new or fancy or flashy, but data integration provides the lifeblood of key business initiatives such as business intelligence or master data management. If you’re like most organizations, you’re spending tons of time and money trying to improve the mechanics of how you move data around the business and make sense of it in the context of critical business processes. Unfortunately, also like most organizations, you probably don’t spend enough time worrying about the quality of the data you’re integrating and delivering. And that leads to some significant issues. Here’s how we see it playing out, in the form of a couple of predictions:

• Through 2012, only 15 percent of organizations will proactively and comprehensively include data quality competencies into their data integration work (0.7 probability).


• Through 2012, organizations that fail to introduce data quality into their data integration activities will see their costs increase by 100 percent or more (0.8 probability).

Think about how much you spend on data integration activities today. What if you could cut that expenditure in half, or protect yourself from having to double that investment over the next several years? And the implications of not focusing on data quality are, by themselves, quite significant. Those organizations that do nothing and continue to ignore the issue of poor data quality will at best continue to operate inefficiently, miss opportunities, and invest their resources in the wrong areas. At worst, they will experience a slow decline in their ability to compete, eventually leading to failure.

So here’s the choice you face – either follow the path of growth and competitive advantage, or follow the downward slope toward failure! Of course you want to do the former. View data integration and data quality as inseparable – identify current areas of waste and inefficiency due to poor data quality, and apply these principles to demonstrate the advantages of a holistic data delivery strategy. Several of the sessions in our Information Management track at Symposium will show you how, inlcuding Data Integration and Quality: Nucleus of the Data Delivery Evolution and Business Intelligence, Business Process Management and Information Management Market Directions.

Submit a Comment | Permalink

Several clients have asked me if there are security issues when using the software-as-a-service (SaaS on-demand) offerings from Web analytics vendors, available from most Web analytics vendors.

Leaking your Web analytics reports to your competition could reveal useful information about your company, but there are security issues for an in-house deployment and also for buying SaaS. Each organization should determine the appropriate balance between security and convenience for each model. The SaaS model in Web analytics is very convenient, and the security threat is low for several reasons:

1. The highest threat that I'm aware of is caused by bad account management. Your analytics users retrieve analytics reports from the Internet via a personal account and password. If the accounts aren't carefully managed, it's possible for a former employee to gain access and retrieve unauthorized reports. One SaaS vendor in this space addresses this issue by allowing domain and IP address filters to be placed on any access to your reports. When the Web analytics server is behind your firewall, only those employees who have access to your private network will also have access to the reports.

2. There could be a security breach at your SaaS Web analytics vendor — either a security hole that enables exploiters to have access to your data, or unethical employees using unauthorized access. Although these scenarios are possible, your vendor's reputation is at stake, and most leading vendors take security very seriously. There have been no breaches that I'm aware of. On the other hand, if you run Web analytics tools in-house, there could also be unauthorized access going on behind your firewall.

3. Network spying is the least-likely security problem scenario. The information you send to the Web analytics data collection system contains information such as Web page titles and product purchase amounts. Like any Internet traffic, these tags could be "sniffed" by your Internet service provider, a government agency or another network-level snooper. It's unlikely, however, that only this aspect of your Internet traffic would be exploited. In other words, if you're being spied on, you have more to worry about than just your Web page activity tags.

Submit a Comment | Permalink

As of this afternoon, the Space Shuttle Atlantis is scheduled to return to Kennedy Space Center in Florida tomorrow. Think back to the September 8 press conference held after the Atlantis flight takeoff was once again postponed. NASA officials discussed the reason for the scrub and the thinking that went into the decision. There are many lessons that can be learned from the way NASA operates, but in this case, we can learn a lesson in how to react to significant events and failures in business process instrumentation. Many organizations are just starting out on the path to becoming a “real-time enterprise,” and should look to NASA for lessons already learned.

The scrub on that Friday was caused by a failed fuel sensor. There are four sensors placed low in the main fuel tank, which sense if the fuel is running low. The rocket motors will explode if they are running when the fuel runs out, and o the sensors act as a safety switch to cut off the motors while there is still fuel remaining. While the other three sensors were still working as expected, the bad sensor incorrectly indicated “wet,” and could not be trusted to sense low fuel. Predefined procedures dictated that with one bad sensor, the mission must be delayed, and the fuel tank must be drained and refilled, so that the sensor behavior would be well understood.

Business intelligence tools used for operational tasks, along with event driven business activity monitors can present a similar challenge to enterprises. Sensors that gather business data and events in real-time may be faulty, and trigger false alarms. Much like NASA’s approach, sensors for important data should be backed up with spares, they should have self-test mechanisms to verify that they are working, and procedures should be in place for actions to take when sensors fail.

After the scrub, NASA was reconsidering the wisdom of scrubbing the flight when three of four sensors were still working, but defended its actions in following its predefined procedures. They were developed without the pressures of operations, when safety might be compromised. Similarly, enterprise processes for addressing significant events and untrustworthy data should be defined before the business operations staff is faced with the situation. Exceptions to the standard operating procedures may be discussed at the time of an incident, but there is value in having standard operating procedures to fall back on.

Taking the step towards building a real-time enterprise is one step of a long journey. There are many new best practices that organizations need to discover, but fortunately, NASA has already traveled far down the road and can offer us some good lessons. We've put together a custom agenda for the Business Intelligence and Information Management community at Fall Symposium. See you there!

Submit a Comment | Permalink