predictive

“Our business is unique, and we don’t think we can use ***”. A common response we get the first time we talk to a company about analytics. The “***” is usually some analytics based process. For example, one manufacturing company insisted that the demand for their products is not predictable. Therefore they would have no use for predictive analytics. Today, this same company is relying on monthly forecasting models to assess demand and validate production schedules.

The more scientific assessment of their situation is that the demand is uncertain, but not wholly unpredictable. This scenario is perfectly suited for predictive analytics, because it is all about “converting future uncertainties into usable probabilities”. The forecasting models do precisely this. Models provide confidence intervals around next month’s demand: “80% confidence that product 1 will sell between 1200 and 1300 cases. 95% confidence that it will sell between 1000 and 1500 cases”. 

Non Manufacturing predictive analytics example

Here is another example of a totally different company that manages donations for non-profit companies. It has a business need that can be addressed by a similar approach. This company creates campaign plans (which are basically a collection of Projects) for all of their customers.  This includes information such as the description of a project, what premium will be used, the date the project is scheduled, costs, etc.  They then would like to use this information to try to predict their cash flow for the coming year.

Once a project is started, it is entered into a financial tracking system. Once in there they record all the bills and invoices associated with the project.  In addition to the project data, they also have bills and invoices for internal expenses, as well as salaries, taxes. What they would like to do is see if they can use historical data to accurately project cash flow. Also look for areas where they can improve.  For example, if they notice that a particular employee takes longer to make invoices than another employee. They can address this with them to improve cash flow.  If they see that specific customers are trending toward paying their bills later, they can address this as well.

The main difference between this and the manufacturing company is the product. One deals with a manufactured commodity. The other deals with an intellectual entity (a project). The ebbs and flows of these entities both affect the top line and bottom line of the business. Understanding these changes will help in better preparing the business for uncertainties. This is where analytics, both descriptive and predictive, will play a valuable role.

There are really two situations where analytics can play an important role and it helps to categorize application areas of analytics into these two broad themes: internal facing and external facing.

Internal facing analytics

Internal facing analytics applications are all those cases which help to make the business more efficient and resilient. The two examples cited above are both internal facing analytics applications. Here are some more in no particular order.

External facing analytics

External facing analytics applications will help businesses increase their top line by driving in more customers, helping to understand customers better and thus enabling business to attract and retain higher value customers. Examples of external facing analytics applications – again in no particular order:

  • Understanding existing customers by proper customer segmentation to determine how to best market to them
  • Prospecting for new customers: which prospects have a higher propensity to turn into customers quickly
  • How valuable are existing customers over the long term by assessing customer lifetime values
  • How to retain high value customers by reducing customer churn addressing customer retention issues

Analytics, can help any business whether it is manufacturing or media. Every business has a balance sheet and an income statement. Analytics can help you logically and rationally address issues which can cast a negative light on these. While on the surface, it would indeed appear that every business is unique, from the point of view of applying analytics, they are really the same.

Original date of posting Jan 14, 2014

Photo by Franki Chamaki on Unsplash

No responses yet

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

simafore.ai