Business Intelligence (BI) is such a great term. The marketing folks really got it right with that one. I mean who doesn’t want to be more intelligent about their business? Potential audience isn’t the problem however. The question is if “Business Intelligence” is really making business smarter? The answer to that may spell trouble ahead for the big pureplay BI vendors like Cognos, BO, etc.
A bit of a disclaimer here, I did work for one of the big two BI firms for about 5 years, so my insight might be slightly biased. But I also think my experience gives me some unique perspective as well.
There are some issues that the BI market is struggling with, even as the market itself continues to grow at double digit rates:
- Perceived value by the enterprise
- Implementation approach and success percentage
- Breadth of data access
- Consolidation and the ERP vendors
I think the most prominent is the realization of benefit from an implementation of BI. Vendors, systems integrators, and customers are having a tough time implementing a large scale BI solution that brings value to the masses. I think this is due in large part because implementers are treating BI much like they do transactional line of business solutions.
BI is hard to implement because people don’t know what information they need. Implementers want to define projects based on a tight set of requirements. I think you can see the trouble that lies ahead. It is another case of our tendency as technologists to focus too much on the technology, and not enough on the possible benefit to the people using it. It takes a fair amount of time for people using BI solutions to truly grasp the possibility the technology presents, and to understand what they need their solution to do and therefore what enterprise BI looks like. Of course, no LOB VP or CIO is going to start a project with an open checkbook and say “After you figure out how to use this stuff, let me know how much it is going to cost and what the ROI is.” This leads to structured implementations that focus on moving existing reports, dashboards, or cubes from one legacy technology to a brand new shiny BI platform. In doing so there are generally some efficiency gains and fewer headaches for IT after a while. But what’s the real value? Chances are the existing reports that are being migrated weren’t serving the business needs anyway. Here we go again, if we automate an inefficient existing process or solution, it will be all better.
Another problem the BI industry faces is the breadth of information that BI solutions touch. First, BI only aggregates data from behind the firewall. Of this internal data, probably only about 40% of what people need is actually in a form that BI systems can talk to you, the remainder in unstructured documents, emails, and heaven forbid people’s heads. Then when you look at the 40% of data that is touchable, BI probably can only get to half of that because of internal silos, technology restrictions, and accuracy of data. Then there is the whole idea of external information that knowledge workers need access to, business news, market research, supply chain data - BI can’t touch that.
This leads to a problem that the entire Information Technology domain is struggling with. That is the ability to turn data into information (I’ve talked about this a little in previous posts). Yes, I can see that I sold 100 widgets last month, and even see trends over time in a really spiffy BI solution. What I can’t see is if my customers are going to need more or less widgets in the future. I can’t see why the northeast sells more widgets than the west. I likely have no visibility into what other products from my company are purchased by my widget customers. I can’t see what new technology is coming out that will make my widgets obsolete.
Of course there is a fair deal of success that can be achieved with the proper implementation of BI today - it wouldn’t be such a large industry with a high growth rate if not. Point solutions that provide visibility into data that was previously not available in a timely or accurate manner can have great impact on an organization. And people are learning from these point solutions on how to implement BI so it can be valuable across the organization. But it is a slow learning process, and one that will take time.
As they get mired down in perfecting their technology and competing with one another, instead of focusing on how to solve the value problem, the ERP vendors that own the data are getting into the BI game. Oracle’s purchase of Hyperion for example. The enterprise is figuring out that they can really only implement small point BI solutions for the time being. As they do that, they realize something else; they let their ERP vendors off the hook ten years ago. They spent tens or hundreds of millions of dollars implementing ERP, and changing their business process and in the process capturing all manner of data that was previously floating in the breeze. But they never worried about what to do with the data they were capturing, or how to provide visibility to it.
Combine this with the fact that he enterprise has realized that the promise of Enterprise BI, or Corporate Performance Management, or whatever the term du jour may be, is not an attainable state yet. You can see why the enterprise is looking to their ERP vendor to provide them a much more cost effective and integrated point solution to provide data access to their users. ERP vendors are happy to do this, because they see the opportunity to drive retention and adoption of their core products with this fancy “Business Intelligence” thing.
Does this spell trouble for the pureplays? Trying to sell a difficult to implement, high priced, low value technology against tightly integrated, close to free solutions being provided by the incumbent ERP vendors. Ever try to sell a brand new house with a big deck to a guy that really likes his existing house and doesn’t want to move, but just wants to add a deck out back?
In ten years, when the enterprise has the experience and wherewithal to really understand what Enterprise BI or CPM is, and the technology has caught up to allow it, the ERP vendors will own the solution. The pureplays that blazed the trail will have been acquired or reshaped themselves into something else to stay alive. Until then, the “Business Intelligence” vendors will go on, making hay while providing nothing more than data visibility and broken dreams of information transparency throughout the enterprise to their customers.
Sure, this sounds like heresy to the BI nation that really believes they are the next ERP. But it all smells too familiar to me. An entire market made from building its infrastructure and value proposition on top of something that somebody else owns. We’ve seen how that model works out. How many of you signed on using AOL today?
I like my house thanks, now I need to go get the permit for my new deck.
Share This
Sphere: Related Content
Info Chat
