There has been an increasing trend over the last ten years for marketing research consultants to analyze customer experience survey data through causal modeling or Structural Equation Modeling. Part of the reason for this trend is the availability of easier-to-use modeling languages. Advances in languages have taken the industry from LISREL and LVPLS (two programs that are quite difficult to use) to AMOS, EQS, Mplus, PLSGraph, and SmartPLS, among others. So, with all of this software at our disposal being used by an increasing number of marketing research agencies and consultants, are companies embracing causality? Do managers accept causal modeling output as proof of finding causal relationships in customer experience survey data?
Judea Pearl wrote a book on causality that, although is a landmark, is something I don’t recommend to others. It is “tough sledding” and requires a big investment of time to understand. (Causality Models, Reasoning, and Inference, 2000, Cambridge University Press) He creates a calculus that, taken together, claims to be a formal theory to accurately describe cause and effect using the languages of mathematics and graphical representations. It was written to advance the philosophy and science of causality for the social sciences, economics, business, etc. But company managers don’t read these kinds of esoteric books.
Managers should remember the old saying that the proof is in eating the pudding. In other words they should test the causal relations found in their cusotmer experience models. All it takes is a little controlled experiment. Most customer experience models are rich in causal details. Pearl says that a causal model should enable you to predict an outcome from doing something. For managers, all that is required is for you to do something. Improvements in CSR knowledge will cause an improvement in the call center experience, for example. All it takes is a little focus. With a little more focus, you can prove that improvements in the call center experience can improve overall attitudes about your brand. These, in turn, can influence financial success of the company. It is not rocket science; but, it can be scientific.
Contributed by Bruce Corner, Synovate Customer Experience
So fun article is! I know more from it.
Posted by: moncler jacket outlet | 18 October 2011 at 18:54
Okay I am building a rocket that must consist of one stage of rockets, however with many rockets, I am not sure how many
Posted by: Round and Brown | 02 July 2010 at 03:01
Science has become a sacred cow ... what a crap.
Posted by: generic viagra | 15 March 2010 at 09:35
Okay I am building a rocket that must consist of one stage of rockets, however with many rockets, I am not sure how many....
What the best amount to use in a rocket that will weigh 4 lbs, but have a payload on board, that weights another 6 lbs. So a total of 10 lbs if I can get the cat to remain calm.
Posted by: cialis online | 19 February 2010 at 09:48
Excellent entry! I'm been looking for topics as interesting as this. Looking forward to your next post.
Posted by: Term papers | 03 November 2009 at 23:51