Data Modeling and Analysis- Delivering Unexpected Value and Improved Marketing

Posted by Margot Tollefson on July 20th, 2017

The magnitude of the cost to tolerate problems due to variation is not usually fully understood, even in financial terms. A study conducted on over 1000 mechanical drawings demonstrated that over half of all the changes requested to be made after the design was approved for production were the direct result of dimensioning and tolerancing errors. In each case, these errors could have been avoided if variation issues had been better assessed earlier in the design process. Most companies realize that these changes can be expensive, but few seem to have quantified this cost.

One of the great things that a good financial data modeling and analysis model can define different business scenarios. A good model should also test how sensitive the results can be to changes in the assumptions. A great way to tackle both of these goals is to build a sensitivity table.

To demonstrate how a sensitivity table works, let's build a very simple model that will calculate the return on a hypothetical investment. We will assume a certain investment amount, forecast annual cash flows and calculate an exit value. From these calculations, we can calculate an internal rate of return (IRR). Our sensitivity data modeling and analysis will look at a couple inputs in the model and alter their values to see how it impacts the IRR.

Over the next year, expect accelerated adoption of predictive analytics within and across business functions as organizations seek to find greater value from their data. Predictive analytics will help organizations uncover previously hidden patterns, identify classifications, associations, and segmentations, and make much more accurate predictions from structured and unstructured information. This will dramatically impact on data modeling and analysis strategies and planning processes as enterprises and service provider rely on real-time analysis of current activity and anticipate what will happen next. By identifying the key drivers of various business outcomes, organizations will be further enabled to deliver more personalized and contextual customer experiences.

Open standards are important attributes that now make data modeling and analysis more accessible and affordable. A caveat is that you cannot make decisions from data that you don't have. It is important to have a good POS system that collects detailed data by store, by customer, by time frame, by payment method, etc. Security and privacy of the data is also a show-stopper if you do not have it. One of the biggest benefits of a retail data model comes from understanding the customer.

The data model can provide: campaign and promotion information with cross purchase behavior along with customer attrition, complaints, credit risk, delinquency, interaction, lifetime value, loyalty, profile movement, profitability and market basket analysis.