Interest and reaction time analysis of credit card offers: Create individual models for interest

Create individual models for interest and for response time and estimate the impact values for untested elements

One dependent variable is the direct rating of interest (eg 1—9 or its binary transform so that ratings 1—6 transform to 0, ratings 7—9 transform to 100). The other dependent variable is the number of tenths of seconds ensuing between the presentation of the concept on the computer screen and the respondent’s rating. Regression analysis uses these two dependent variables along with the presence or absence of the elements to estimate the contribution of each element to the dependent variable. Each of the two variables generates its own respective equation. Since each individual evaluated concepts developed by an experimental design, the dummy variable regression modelling can be done at the individual level.

Identify key subgroups or key new segments of individuals

The individual-level models for impact (based on interest ratings) can be processed further in order to identify segments. Some segments may be those individuals who are similar in their banking behaviour. These are traditional and conventional methods to create subgroups. There are more psychological approaches to segmentation, which try to find like-minded individuals. However, it is not clear from these psychographic segments whether or not individuals in the same segment respond similarly to messages for a specific type of product offering. At a more profound level, do there exist endogenous segments in the population, transcending conventional demographics and usage patterns, but similar in the way that they respond to the concept elements? This approach to concept segmentation was described previously, and has been shown to work internationally, so that individuals in different countries can be classified as members of the same segment. This segmentation was used here to identify new, hitherto unexpected clusters of individuals. In addition to concept segmentation, the respondents were divided into pre-defined groups of individuals, based upon geo-demographic and bank-supplied criteria.

Create new concepts

The results were used to create new offers for the market, based upon the specific elements that the bank could use. These new concepts, underlying offerings, were deemed by the bank to be financially conservative to the bank, yet attractive to the potential customer.


Representative APR 391%. Average APR for this type of loans is 391%. Let's say you want to borrow $100 for two week. Lender can charge you $15 for borrowing $100 for two weeks. You will need to return $115 to the lender at the end of 2 weeks. The cost of the $100 loan is a $15 finance charge and an annual percentage rate of 391 percent. If you decide to roll over the loan for another two weeks, lender can charge you another $15. If you roll-over the loan three times, the finance charge would climb to $60 to borrow the $100.

Implications of Non-payment: Some lenders in our network may automatically roll over your existing loan for another two weeks if you don't pay back the loan on time. Fees for renewing the loan range from lender to lender. Most of the time these fees equal the fees you paid to get the initial payday loan. We ask lenders in our network to follow legal and ethical collection practices set by industry associations and government agencies. Non-payment of a payday loan might negatively effect your credit history.

Calculate APR