Home » Business » Project Management

Design Of Experiments In Marketing

Mar 19, 2008
Design of experiments, (DOE) or conjoint analysis as it is known in marketing context, is known to be the most powerful statistical method for establishing the linkage between a customer's decision-making process and the service or product being offered. After effective application of DOE, companies find it easier to gain an insight into the significant Xs affecting a customer's decision-making ability.

Marketing Problems

Eventually, the primary aim of marketing is to calculate the upcoming market share net sales, or profitability of an offering, thus, allowing a company to:

-Foretell customer buying tendency
-Boost customer retention
-Ascertain trade-off strategies during contract negotiation
-Ascertain competitive pricing
-Predict sales
-Control brand equity
-Devise product elements
-Establish price sensitivity
-Forecast and reduce customer switch rates
-Ascertain best market position for new product introductions
-Forecast and optimal response rates for advertising campaigns regarding content creation and allocation channel mix.

The Unknown X and Y

During decision-making processes, customers usually prefer accepting or rejecting a company's offering. There ought to be a way of understanding the connection between factors customers consider before coming to a final decision, as there are times customers are oblivious of the complicated psychological processes behind their decision-making processes.

A standard contingency plan would be to develop an analytical model that involves evident characteristics of the offering. It is a wrong notion among businesses to assume that customers make decisions based on the cost - a criteria they usually follow. Usually customers make decisions based on a complex interplay of psychological and sociological factors.

Divorcing critical-to-quality elements (CTQs) for the external customer (value) and internal customer (cost) allows a company to ascertain an appropriate offering for optimizing both CTQs simultaneously.

Usually a Six Sigma project concentrates on a single Y for simplifying the analysis. Although there is a complicated mapping of Xs to Ys in the mind of the customer, the primary focus is the "yes-or-no" or the single Y. Since this discrete measurement is not as beneficial as a quantitative measure, hence the data gathered is used for driving into the multiple Ys in a customer's thought process by asking them to decide from the given choices. After this is successful, it allows the derivations of an extrapolative model for the decision-making process.

Rating Factors in Decision Making

Customers are incapable of determining ratings on absolute scales, but are considerably good at forming ratings on the basis of the relative scale. A customer's resolution to reject or accept an offering is done in a similar poorly understood and complex process.

Xs and selecting Factor Levels

Studies carried for selecting Xs use the same brainstorming tools employed in DMAIC's Measure phase. A fishbone illustration could be used for charting all probably factors the marketing department wishes to evaluate. In the end, the fishbone diagram should illustrate "a buying decision" effect.

The project should be expanded for including elements of the form of the presentation or include aspects of the offering. One can limit the factors included to a dozen by spanning the project, or carrying out a short customer survey for determining the essential aspect making buying decisions.

Lastly, studies conducted should depict the advantages of inspecting DOE/conjoint analysis applications for Six Sigma projects after being used in marketing. Quantitative predictive models that capture the customer decision-making process are powerful marketing tools for foretelling customer behavior.
About the Author
Tony Jacowski is a quality analyst for The MBA Journal. Aveta Solution's Six Sigma Online offers online six sigma training and certification classes for lean six sigma, black belts, green belts, and yellow belts.
Please Rate:
(Average: Not rated)
Views: 251
Print Email Share
Article Categories