If I told you in September that the Baltimore Ravens would win the Super Bowl in February and you did nothing with that information ( i.e. place a bet, announce your prognostication skill, etc. ), was my prediction worth anything? Much like “if a tree falls in the forest and no one is there to hear it, did it make a sound?”, you need action from a piece of analysis, prediction, etc. to make it worthwhile.
Thomas Jefferson once said:
For Business Intelligence, this is a much under served concept. For many years, Business Intelligence has focused, quite rightly, on the analysis, or the prediction, but not as much the “call to action”. The infamous cartoon implies that to provide true business value, we need to be more “explicit” in how we link Action to our Intelligence:
Here are my thoughts on the state of Actionable Intelligence in the field of Business Intelligence for 2013:
- Alerts Via Email Aren’t Really Action: Much like the above cartoon, assuming that someone is “on the other side of the line” when you send an asynchronous email is using the “miracle occurs here” approach. Many platforms and solutions have built their action on these types of alert emails. Some have developed escalation rules to be more sophisticated, but overall it is still an asych email…
- Workflow is where it is AT: When you start adding in workflow management to analytical platforms, you are starting down the road of unrelated product conglomeration ( or empire… err… technical stack building… ) for business intelligence companies. Until you realize that, you NEED to complete the feedback loop with end users and business stakeholders to provide the next level of value. I hope to see more of related product diversification activity in 2013 from the industry either via partnership or feature/function inclusion.
- Collaboration Continues: The inclusion of collaboration features in business intelligence platforms is a good step on the path to Actionable Intelligence. However, Collaboration can often be interpreted as part of the decision process vs. being the action or result of the process. In our society driven by risk avoidance and millennial workers, we might be to guide the collaboration toward action. However…
- Dynamic vs. Static: …one of the big issues in collaboration is how to handle the ‘static’ view of what the system represented (i.e. a screenshot, snapshot of data) when a collaborative comment/question is made and what the systems ‘actively’ says (i.e. real-time dashboard, current view of data) when someone comments or answers the question. Static vs. Dynamic views of analytic or predictive results in collaboration functionality needs to be addressed since the disparity between those two ‘states’ can be the difference between action and a false positive. Unfortunately, I don’t have a strong answer to what the solution should be other than “yes”…
- Business Intelligence Becomes Process Intelligence? Too often Business Intelligence is mapped to a ‘data’-based as opposed to a ‘process’-based solution. By that, I mean the data is oriented on the data than the operational process that the data came from originally. This isn’t a major change, but a change in focus. Data should be acquired, stored and presented with the metadata of its process vs. stripping that information away and attempting to rebuild it later on.
- Actionable Intelligence needs its own BI: Just as we have metadata about data and often processes, we must have metadata about the results derived from analysis. We are closing the loop in a more meaningful way with predictive models. We should now close the loop on actions and their value to an organization. Having this metadata, and ability to analyze on actions, will feed into business optimization concepts like Operational Excellence.
What say the readers?
Is adding robust workflow into business intelligence platforms folly? Do you already link your business intelligence analysis to workflow platforms? If so, do your workflow platforms feedback to your analytic solutions? Have you mastered Operational Excellence without the feedback loop?
Next week, I will take a more in depth look at a Data Management topic – The wonderful world of in-memory databases: Can you be in-memory and ACID compliant? Can you be in-memory and have rotating disks? Does it matter how you mix your in-memory with your spinning, and non-spinning (aka flash), disks?
NOTE – For those unfamiliar with the song “88Lines about 44 Women” by the Nails, I highly recommend you give it a try. At the very least, it was the inspiration for this series of blogs.