First Step in Data Analytics
Data and prediction analytics projects never start out with the goal of building a prediction model. Instead, they are focused on ….
Gaining new customers
Selling more products, and
Adding efficiencies to a process, etc.
Unfortunately, predictive analytics models really do not do any of these things. These models are built on patterns of historical data extracted from sources inside and outside the organization if available at all. These predictions do not solve business problems. Instead, they provide insights to direct organizations towards solving them.
Step 1: Business Understanding
Step one in any data or prediction analytics project is to fully understand the business (or organizational) challenge in order to architect a proper data analytics model to help solve it.
This involves -
An analysis of the needs of the current business challenge or opporunity
The data available for use
And the ability of the business to use the data and subsequent analytics
In general, translating a business problem into an analytics solution involves answering the following key questions: For those lean six sigma practitioners, this is analagous to translating a practical problem into a statistical problem.
What is the business problem or challenge?
What are the goals that the business wants to achieve?
How does the business or value stream currently work?
In what ways can a data or predictive analytics model help address the business problem and/or provide insights to help solve it.
The answers to these questions are used to build the intelligence around the potential analytics or prediction model which is step 2 of a data or predictive analytics project.
At CiKATA we start this step with a Kaizen Event to define the analytics map using Post-It's. This activity maps the business problem or challenge to influential business processes, behaviors or demographics. See image below.
Figure 1 Analytics Feature Map
The pink “Feature” Post-Its in figure 1 above are the descriptive features used in the subsequent data analytics table. Once complete, the data analyst or project leader will compare these features with the different data sources and data cateogories that are available to project team.
If you enjoyed this article, click the like button and share with your network. If you would like more information about this topic or related initiatives contact us at succeed@cikata.com. Subscribe to receive value adding content and training discounts not open to the public.
