Next Generation DSRs - An Analytic name is not enough

You need not always build your analytic tools, sometimes you should buy in. If the chosen application does what you need that often makes good economic sense... as long as you know what you are buying.

Let's be clear, an Analytic name does NOT mean there are any real Analytics under the hood.

For many managers, Analytics is akin to magic. They do not know how an analytics application works in a meaningful way and have no real interest in knowing. At the same time, there is no business standard for what makes up "forecasting", "inventory optimization", "cluster analysis", "pricing analysis", "shopper analytics", "like products" or even (my favorite) "optimization".  Don't buy a lemon!

Clustering with a destination in mind

I've posted before on Cluster Analysis, in an attempt to demystify one of the more accessible and useful analytical approaches for CPG/Retail teams (see Cluster Analysis - 101) .

Finding groups of similar stores (for example) can be a very effective way to manage the complexity of offering each store group what they really need without having to deal with each one individually, a mammoth task.   Whether you are looking to find groups of stores, shoppers, regions, products or even sales patterns a very similar approach can work for you.

Clustering is part of the journey it's not a destination.  If you don't know and understand what decisions your analytic work should enable  (your destination) how can you build a good model?

Point of Sale Data – Category Analytics

If you haven’t already read the previous entries in this series, you may want to go back and check out [Point of Sale Data – the basics] to see why you really need a DSR to handle this data, and  [Point of Sale Data – Sales Analytics]  for some thoughts on analyzing sales drivers that are equally relevant to Category Management,

 As Category Manager you’re working with the retailer to help drive sales for the entire category.  You hopefully have access to the full data for your category (which could be substantially more than your account manager colleagues).  Let’s see how predictive analytics and modeling could help address some of your challenges:  How well are current planograms performing?  What is the best product assortment for each store?  How can you best balance customization of assortment by store with the work required to create that detail?

Point of Sale Data – Sales Analytics

I’m assuming that you now have a DSR (see [Point Of Sale Data - Basic Analytics] ) so you can manipulate the large quantities of data necessary to do this work, you have your routine reports automated and use the DSR for ad-hoc queries against the POS data. 

The DSR provides a great foundation for analytic work: use it to integrate multiple data sources, clean the data, handle very large data volumes as though it was all sat on your desktop and it will help you build reports that summarize that history with ease. Typically, the DSR does not provide much help for you with predictive-analytics. 

Let’s look at an example related to what really drives sales.   Do you know?  Can you quantify it?  Knowing these answers with quantified detail can help you better explain your sales history and plan for the future.  Better promotions, better pricing, supply chains that anticipate peaks in demand and make sure the product is on the shelf when it’s needed.  Here are some of the things that could drive your sales:

Cluster Analysis - 101

The current Wikipedia page on Cluster Analysis, excerpted below, is correct, detailed and makes absolute sense.  Then again, if you do not have a background in statistical modeling, I'm guessing these two paragraphs leave you no wiser.

Cluster analysis or clustering is the task of assigning a set of objects into groups (called clusters) so that the objects in the same cluster are more similar (in some sense or another) to each other than to those in other clusters. 

Clustering is a main task of explorative data mining, and a common technique for statistical data analysis used in many fields, including machine learning, pattern recognition, image analysis, information retrieval, and bioinformatics.

Wikipedia 4/2012 

In this post I hope to provide a workable introduction for people that need to be educated consumers of cluster analysis.

Reporting is NOT Analytics

Reporting is about what happened; Analytics is about answering "what if" and "what's best" questions.  Most of the materials that land on a VP/Director’s desk (or inbox) are examples of reporting with no analytical value added.