Technology is not the fundamental challenge that organizations face when it comes to implementing business intelligence (BI) solutions. Certainly a lack of alignment between IT and business is a problem, but it’s really just a byproduct of the fact that BI solutions have been so complex in the past that business users were required to bring in IT to create for them the analysis they wanted. Of course that will lead to alignment problems. Is anyone really surprised by this? When the people asking the questions can’t answer them for themselves, and the people who can answer the questions are not the ones asking them, how could we really have expected that everything would be rosy without a lot of effort?
It’s not lack of executive commitment. Certainly, when BI fails, lack of executive commitment plays a part. But, we have to ask why the executives weren’t committed? They certainly care about growing their business, and analytics are required to do that effectively. So why aren’t they committed?
It’s not even the complexity of BI solutions. There have been tons of articles and blog posts written about how BI is too complicated. And often it’s suggested that complexity really is the core problem. It’s true that if you make things simpler, then you get rid of the great divide between business and IT because business can now be self-sufficient and answer their own questions. And BI will no longer fail because of lack of internal expertise to keep a complex system running.
Complexity is certainly a huge issue, but there is overwhelming evidence that it’s not really at the core. First, for years people have been focusing on trying to make BI more accessible by simplifying the user experience. More recently, several companies, both old and new, have offered hosted, or software as a service (SaaS) BI solutions to get rid of the complexity that is involved with managing a BI solution on premise. The Holy Grail has been to make it simple so we can deliver “BI for the masses.” But many BI initiatives still fail to get adopted and have impact on companies.
We’re chasing the wrong grail. Simplicity is absolutely required, but simplicity alone is not enough. Once you peel away the barrier that was created by complexity and give users the ability to answer whatever questions they have, you uncover the real culprit preventing companies from being successful with BI: People have no way of knowing which questions are meaningful ones to ask, and which are meaningless.
I’m not aware of any college classes that teach people how to use numbers to analyze sales, or customer support, or human resources, or suppliers. I don’t know of any business schools that teach this either. And companies don’t teach their employees — most people just end up doing something very similar to whatever they learned from their managers, if anything. To be clear, I wouldn’t be surprised to discover that such classes and training programs exist somewhere, but they’re certainly not common.
Imagine what would happen if you hired someone to do accounting for you who had no background in finance. They’d be forced to make it up as they go. What’s the likelihood that they’d come up with the key financial metrics like ROI, Cash Flow, Liquidity Ratio, Debt to Equity Ratio, Current Ratio, Inventory Turns, EBIT, Days Sales Outstanding, etc? It’s pretty unlikely. So, the financial analysis this person would do wouldn’t be very useful in helping the business, so it would fall by the wayside.
This is exactly what we do with BI solutions. We ask business users to define the key metrics they need to run their business. They don’t have basic background in how to use analytics to improve a business, so they have to make it up as they go. Though they come up with some good metrics, they usually miss others that are absolutely critical. So, management doesn’t see much value coming from their BI initiative. Since it’s not really moving the needle in their business, support wanes.
We need to give people some basic exposure about how to effectively use analytics. As a society, we teach people to read and how to interpret the meaning of what they’ve read. That’s basic literacy. People also need exposure regarding what numbers to look at in their business, and how to interpret the numbers. We need basic numeracy to complement literacy.
How do we move people towards numeracy? Part of it will be up to analytic vendors and user communities to define and share best practices for analyzing a business. But we can greatly accelerate that process by providing users with analytic solutions that provide the key metrics for analyzing the various areas of their business built into the solutions. That’s why analytic applications are so useful. They get us away from generic BI tools, and move us towards real solutions focused on addressing specific business problems and which include embedded best-practice metrics and analyses.
If you don’t think this is feasible because you believe that each company does business differently and each therefore must define their own KPI’s, then I can prove it to you. I can take your sales pipeline data, apply generic best practice analyses to it, and I guarantee you that I’ll find important issues and opportunities regarding your pipeline. Don’t believe me? Then accept my challenge and I’ll prove it to you.
By removing the barrier of BI complexity, we then expose the fundamental issue that we’re providing people with tools to get questions answered, but not with the context to identify which questions have the most impact. People need to demand that their BI vendors provide them not only with the ability to get answers, but also with the right questions too.