Time to Form a Business Intelligence Institute
|
There's always been a lot of frustration surrounding the whole concept of business intelligence. In fact, a lot of people consider it to be one of the ultimate oxymorons. A lot of that frustration stems from the fact that an IT organization can spend millions of dollars building, deploying and maintaining a business intelligence application only to discover that only a handful of users are getting anything out of it. This situation use to routinely perplex Ken Rudin, the CEO at LucidEra, a provider of business intelligence software as a service. But Rudin soon began to understand the source of the real problem once he implemented a business intelligence-as-a-service offering. As the entity now responsible for running the application, Rudin now gets to see first hand how end users use the system. His conclusion might shock some people. The basic problem with business intelligence applications is that outside of say the finance office nobody really knows how to use them. And that's largely because most people don't really know what kinds of questions to ask about their business. That issue has led Lucid to rethink its entire approach to offering business intelligence as a service. The company now spends a lot of time effectively coaching users how to ask questions about their business. What's troubling about all this is that it brings into question the whole value proposition of business intelligence. What's the point of spending millions of dollars on applications that most to the business community is too ignorant to use? To rectify this problem we need to create something that is roughly equivalent to a Business Intelligence Institute that end users can attend to learn how to ask penetrating questions using the data available in business intelligence applications. Short of that, all we're really doing is building complicated applications that only a small percentage of the end user population can ever really derive value from. A lot of fuss is made of the historic divide between business and IT. And more often than not IT is blamed for the problem. But when you get right down to it, you can help but wonder if the divide exists in the first place simply because the business side has no real idea what it's looking for in the first place. |
Comments (3)
To take this a step further, the actual culprits are time and commitment.
Those that 'probably know what - Biz' and those that 'could possibly do - IT' are typically too busy to take the time out to thrash a clear, measurable strategy with respect to corporate governance, business measurements and KPI's - including ensuring that the key resources will be available.
Both parties do not take too readily to an outsider trying to lead a new strategis direction, be it a full blown BI Roadmap and / or DWH Strategy, or a finite specific solution to help monitor a few critical KPI's.
These critical KPI's are normally an inner dragon's hoard that is guarded and mothered by the chosen few that have created them - and they have been through so many itterations that trying to replace or automate them is akin to destroying a private fiefdom.
Every site and organisation that I have worked for, and consulted to, has a few key personel, that due to their pivotal role(s), are either overwhelmed by their day to day workloads, or are treated with kid gloves in determining who they spend their time with, and on what.
And this is on both sides of the fence.
Arrogance is also a subtle opponent of any new endeavor, so take a lesson from Kimble's age old treaty that the softer issues will do more damage than any technical hurdle - and you have a situation that will always be around impacting these types of projects.
As to the 'few' vs the 'many' in receiving any benefit, well it's related to the 80/20 pyramid in delivering INTELLIGENCE to the key policy makers and business custodians. I see BI used too often as an umbrella that generally covers all REPORTING endeavors.
There will be some parts that might only feed one or three ppl - if they're your CEO and CIO, then you're doing a good job. But, the entire BI realm is about collating and placing the building blocks for the eventual effective use of the technology and the inherent use of the information therein.
Most just don't follow through with the initial plans and eventually entities like training, ongoing analysis, POC's (proof of concepts), feedback loops, etc that could be handled via a BICC are just left to fade away as the releative personel are inundated with year end's or critical work flows.
And yes - most corporations are far too busy to understand how they could make more effective use of the tool sets and most vendors are likewise far to busy trying to sell stuff instead of leading the way.
Catch33 - IT/BIZ/Vendor - and a few large ego's and lack of the time and commitment to make an effective intelligent platform over a period of time (that must be constantly revisited), instead of a flash chart and some drill though reports within minutes (that 'yawn' can still be done in excel).
Posted by MikeD | October 1, 2008 1:15 AM
Good post, and good follow up comment. Just wanted to add my two cents...
Traditional barriers to BI success are often quoted as complexity and price. Both are definitely true to a certain extent, yet it has not stopped companies from spending large amounts of capital and time to try to make it work. However, the SaaS model is bringing many startups to BI, and these companies are not constrained by previous versions or ways of thinking. For this reason, it seems this new blood could bring about a paradigm shift not only in the tools (the technology) but in the way people interact with BI.
At Good Data, we focus on two things: VISIBILITY and ALIGNMENT. Visibility covers the technical aspect of extracting insight from raw data (analysis). Alignment covers making the insight accessible to the people who need it in order to act on it.
Traditional packages force the user to accommodate the application. A familiar example of this can be seen at the airport. Is there a more un-intiutive system than the ones in place at the check-in desk? Good Data aims to turn this around, and have the application accommodate the user with a simple and familiar Web interface. Built-in collaboration features encourage sharing, comments and overall group think. We believe this fosters accessibility throughout the team, and creates the alignment required for BI success.
This collaboration is largely missing in traditional BI. We feel that empowering people to see the data, then provide multiple methods for interacting with the data (create feedback loops through comments, or provide simple tools to extend/refine analysis) will allow people to learn from one another and begin not only asking the right questions, but also finding the answers.
Posted by Ray Light | October 1, 2008 11:20 AM
I don't agree with MikeD's comment about time and commitment of the business side and the IT side being the main culprit. I've been on most sides of the BI market: vendor, consumer/user, and consultant, and though time and commitment are factors, they are really the result of the two primary issues with BI: 1) It's too hard for people to get answers to their questions, and 2) even when it's made easier to get answers, you'll discover that most people don't know the right questions to ask.
Regarding point #1: Yes, biz and IT are usually at odds when it comes to BI. How can you expect anything different when the people who have the information needs (biz) aren't able to get it for themselves, and the people who can get the info (IT) aren't the ones with the actual need?! If you can move to a new model where biz isn't reliant on IT for everything, and where IT isn't overwhelmed with requests from the biz folks, then you can remove the need for large amounts of time and commitment to make sure biz and IT are constantly aligned. That's why SaaS BI is becoming so popular. Biz gets a solution quickly, and IT's resources get freed up. Coupled with a focus on UI's designed to help business users (not business analysts or IT personnel), the need for time and commitment drops dramatically.
Regarding point #2: Once you remove the traditional barriers that required the large amounts of time and commitment and give users a solution that's simple to set up and simple to use, you run into the next problem: most people haven't ever been shown how to analyze different parts of their business, so they don't know the right metrics to be looking at and the right questions to be asking about those metrics. That's where Michael Vizard's comments about a Business Intelligence Institute come in. I'm not saying that we specifically need such a physical university, or even a specific organization, but we need to get people educated on what are the best ways to analyze different areas of your business. Otherwise, people are left to make it up as they go.
Posted by Ken Rudin | October 2, 2008 12:51 AM