Customers are the lifeblood
of any business. Understanding the unique and rich data that comes from their
core customer base helps in creating services that truly meet the needs of
those customers as well as marketing the most relevant programs to the most
interested parties. It creates a higher level of sales and satisfaction spurred
by the interconnectivity of customers and the organization. The customer’s
needs are better fulfilled with the offering of products and services they are
actually interested in. Precisely how this is done is a process that can be
learned and adapted.
With the advancement of the
Internet and e-commerce the use of social research to understand customer
behavior becomes possible. With the increase in customer data it is possible to
create greater data mining and clustering of customer profiles to understand
buying patterns and behaviors (Prasad & Malik, 2011). It is through the
development of higher levels of data analysis that services can become more
effective and beneficial.
Let us look at an example. Analysis of a
large database finds that customers who bought airplane tickets also purchased
beach related products. Yet what if these customers were also found to purchase
more outdoor gear and spent a greater amount of money on outdoor activities? It
would be possible to build a customer profile based upon their exploratory and
thrill seeking behavior.
In order to understand unique
social purchasing behaviors requires the categorization and analysis of profile
customers. It requires a method of making meaning out of the historical data (i.e.
purchases over time) being presented. Qian et. al. (2006 suggests the
following:
- 1.) Standardize profiles
- 2.) Screen out uninteresting profiles
- 3.) Using basic functions to categorize profiles
- 4.) Apply algorithms to the categorizations
- 5.) Identify unique profiles for further analysis
Once the profiles are
standardized it is possible to categorize their behavior into clusters. These
clusters are used for additional analysis and the determining of patterned
behavior. That patterned behavior indicates that there are latent psychological
functioning occurring and it would be beneficial to use multiple analysis methods to
better highlight their behavioral thought processes.
This process is fairly
accurate and can lead to better marketing techniques based upon profile
attributes and responses to previous marketing (i.e. previous purchases). One simply needs to draw connections between
the different sets of data and tests that were conducted over time. A study by Leung (2009)
found that out of 1,500 profiles analyzed that 91.73% of customer profiles were
segmented correctly.
High levels of accuracy and a
process for separating and analyzing consumer behavior is a benefit that
organizations should not ignore. The use of more pin pointed marketing
techniques further encourages efficient use of company resources by ensuring
that products are actually of interest to the customer. Social research techniques can help identifying latent psychological functions that further enhance organizational
profits.
Leung, C. (2009). An
inductive learning approach to market segmentation based upon customer profile
attributes. Asian Journal of Marketing, 3
(3).
Prasad, P. & Malik, L.
(2011). Generating customer profiles for retail stores using clustering techniques.
International Journal on Computer Science & Engineering, 3 (6).
Qian, Z. et. al. (2006).
Churn detection via customer profile modeling. International Journal of
Production Research, 44 (14).
