Thursday, November 10, 2016

Getting Un-Stuck

Getting Un-Stuck by TR Dorfman PhD, ...dorfstats.com



Can’t seem to get things going?  Nothing necessarily wrong, but also nothing positive in the way of change.  Try this.

  • 1.       Acquire and store important quantitative data, especially by date.  This includes sales, marketing, pricing, commodity prices,  and competitors price.  Don’t be restricted by this list-save the key data you think is important.
  • 2.       Decide specifically the issues you would like to solve.  Make a written statement describing the problem.
  • 3.       Have this data analyzed. 
  • 4.       A skilled data professional can help determine the important components of your data in relation to your problem.  Additionally, accurate forecasting is quite possible in terms of predictive analysis of the data.













Thursday, November 3, 2016

WHY USE PREDICTIVE ANALYTICS?


Companies of all sizes that utilize predictive analytics:

2x more likely to be in the upper quartile in their financial performance

5x more likely to make decisions faster and more timely than their competition

3x more likely to execute decisions as intended

2x more likely to use their data more frequently when making important decisions.


There you have it.  Increased profitability from faster, better decisions.

TR Dorfman PhD, dorfstats.com

Tuesday, November 1, 2016

ANALYTICS WORKING HAND IN HAND WITH MANAGEMENT DECISIONS:

In corporation Alpha, there’s been a slowdown in orders from major customers for Product A, but not for a similar Product B.  Market intelligence shows an increase in sales for a similar Product A at a competitor.  In the last year Alpha raised the prices twice for both products, but kept the prices in line with the competition.  However Product A sales declined following both price hikes, but not     Product B.

Identify the Problem: 

Why are sales declining for Product A but stable for Product B?

The Solution:

Review of the pricing decision showed management anticipated a rise in the price of a key commodity for both products.  Although these increased costs did not occur, their imminent price rise continued to weight on managements pricing decisions of their products.

Management meeting about the slowdown in Product A revealed data concerning technical support calls for both products.  Product A showed a sharper slowdown in support calls than Product B.

A contract was issued to a predictive analytic company to help figure out the dilemma.  Within a week a forecast time series analysis confirmed the coming price increase in the important component commodity.  Additionally a statistical comparison between the support calls for both products with sales data revealed a sweet spot for the number of support calls that correlated with increased sales.  Also, support calls calculated to account for 26% of repeat sales.


Management maintained their price improvements and increased their targeted support calls for Product A.  Within the quarter the sales declines ended and started their rebound.