Part 1Alan Green needs to answer the decision problem of whether charging fees for online banking use is more profitable for Pilgrim Bank than offering incentives to promote wider use of the online channel. To begin solving the problem, Mr.
Green first must address the following research issues: how much more/less profit do online users generate; is this difference significant, what are the measures of customer profitability, what are the characteristic of the bank’s online users and profitable customers, what are the costs of operating the online banking channel, and finally what measures does the bank take to retain its most profitable members. The research design is two-part. The first is an informal qualitative meeting with analyst Jane Raines.
The purpose of this research is to obtain any useful general knowledge on measures of profitability, customer behaviour, cost structure, profitability management and their relations with each other. The second part is an in-depth qualitative research based on statistical analyses on a database of customer profits, online usage, demographics (age, income, geographic), and tenure years. Data collection for the meeting is done by simply imputing key points with brief explanations in a word document.
Collecting data for qualitative research is equally uncomplicated with Alan asking Erica Dorstamp to retrieve information from the system database of year-end 1999 and put it on a disk. Demographics of the sample are also given to Mr. Green. With the help of Ms. Dorstamp, Alan randomly selects 30,000 customers out of a total of about 5 million. Key findings in the meeting with Jane Raines are numerous. Alan ascertains that customer profitability is derived from multiplying balance in deposits by the interest spread, plus transaction fees plus interest from loans, and minus the cost of service.
The total cost is further divided into variable costs and fixed costs. Variable costs are lower for online transactions, but it has a higher fixed cost structure. Mr. Green also finds that there is no clear correlation between balance amounts and customer profitability. Lastly, Alan learns about the initiatives Pilgrim Bank take to increase profitability and retain its most profitable customers. Data analysis of the customer database begins with the testing for sample bias. Customers are sorted from descending profitability and they are charted against percent cumulative profitability of the bank.
Alan finds congruency between his findings and the one results presented by Jane Raines, thus finds reassurance that his sample is not biased. The results also confirm that roughly 10% of the customers constitute 70% of Pilgrim Bank’s profits. He then proceeds to summarize the statistics and finds that, on average, online users are more profitable than non users ($116. 36 versus $110. 79. ) The summary of the statistics also include standard deviation. The mean and standard deviation is not calculated for geographic information since it is a nominal scale.
Part 21. Customer revenue is generated from three sources: investment income from deposit balances, transaction fees, and loans. The revenue part of the investment income is the difference between the accrued interest payment on the deposit and the income made from investment activities like mortgage lending. Retail banks deal with deviations in customer profitability by striving to retain its most profitable customers and giving opportunities for lower profit customers to become more lucrative.
One way of achieving this goal is diverting more resources to more valuable customers. Banks also give incentive to customers to use lower cost channels to increase their profitability. Moreover, customers are educated on the cost benefits, availability and convenience of the low cost channels. As more and more customers rely on ATM’s, telephone and online services, there is a decreased need for physical infrastructure, thus further reducing costs. Finally, profit tiers can also include long term factors to produce more accurate customer lifetime profitability evaluations. . 1. There are numerous conclusions about the profitability of the entire customer population. The mean is $111. 50 with a standard deviation of 272. 84 meaning that approximately 64% of the population have a profit between -$161. 31 and $384. 34. Plotting a frequency chart (Exhibit 1) yields results where the majority of profits lie in the -$100 to $200 range. However, the frequency chart seems to indicate that the mean should be around $-100 to $0, the interval with the most frequency counts. The median of $9 and the mode of $-2 further contributes to this notion.
When examining the cumulative profit chart (Exhibit 2), Green concludes that only a little more than half of the customers are profitable, and more importantly, a little over 20% of Pilgrim’s Bank customers contribute to 100% of total profits. The cumulative profit chart not only stresses the importance of retaining existing highly profitable customers, but it also shows that there is plenty of room to make non-profitable customers valuable. Profit reaches as high as $2071 in the sample yet the average is only 5% of that value. 2. 2.
The regression analysis with profit as the dependant variable and online usage as the independent variable is summarized in Exhibit 3. Special attention should be given to the adjusted r-squared value, the co-efficient of the online variable, and the p-value. The adjusted r-squared value is very close to 0, meaning that the best fit line does not accurately estimate the relationship between profit and online usage. It also translates to a possibly poor regression model where important variables are left out. The co-efficient for the online variable is 5.
This value is important when we compare it to the other regression model with added demographic variables. The most significant information derived from the regression is the p-value. The associated Ho is ? 2 = 0 and Ha is ? 2 ? 0 in the model y = ? 1 + ? 2x+ ? where y is profit, ? 1 is the intercept, ? 2 is the co-efficient of online usage and ? is the standard error. The 36. 98% p-value is much greater than the significance level of 5%, thus we do not have enough evidence to reject Ho. In other words, we do not know for sure if online usage significantly affects profit.
The regression analysis with profit as the dependant variable and online usage with demographics as independent variables is summarized in Exhibit 4. At 0. 0452, the adjusted r-squared value is larger than the previous one, so the goodness-of-fit is better. However, the value remains small. Since most variables from the database are exhausted, the only conclusion is that the profit relationship is not exactly linear. The online variable co-efficient is 16. 68 meaning that there is a downward bias when demographic data are excluded. The co-efficient also reveals that online users generate 6. 68 more revenue. Interestingly, the online p-value is 0. 0026 with the added variables, thus we now have enough evidence to reject Ho. Similar to the last regression Ho is ? 2 = 0 with ? 2 being the co-efficient of online usage in the regression model y = ? 1 + ? 2(online usage)+ ? 3(Age)+ ? 4(Income)+ ? 5(District)+ ?. We accept Ha (? 2 ? 0), and conclude online usage significantly affects profit. Looking at the other p-values, it should also be noted that there is very strong evidence that age and income are both related to profit.
Not surprisingly, there does not seem to be any proof supporting geographic region as a significant estimator of profit. In trying to explain why the p-value becomes significant in the latter regression, we need to first look at correlation values of online usage and the demographic variable. The correlation matrix in Exhibit 5 reveals that age is slightly negatively correlated with online usage and income has an extremely small positive correlation. This information seems reasonable since younger generations are more computer savvy, and some income is required to have computer and internet access, plus the education to be computer literate.
Thus, when these variables are factored in the regression, we obtain a truer effect of online usage on profit, and consequently, enough evidence to reject the null hypothesis. 2. 3. To investigate if there is any systematic difference between consumers with complete demographic records and those who don’t, basic descriptive statistics should first be reviewed. Profit and online usage averages are 127. 17 and 0. 1295 respectively for data with demographic records, and 70. 99 and 0. 1020 for incomplete records.
There is a big difference of more than 50 in the two means. Nonetheless, a hypothesis testing is required to see if the difference is significant. This is shown in Exhibit 6. The results conclude that the differences for both profit and online usage are material. This raises the question of external validity. The regression analysis shows that there is a positive effect on profit from online users, but the analysis omits the data without complete demographic records, thus we need to excise caution when generalizing conclusions to the entire population.
More specifically, in order to avoid the issue of external validity, we may only apply our conclusion to customers with complete demographic records. 3. Firstly, to maintain Pilgrim Bank’s profits, resources should be continually invested to retain the top 20% most profitable customers since they considerably contribute to the bank’s entire revenue. We have proven through regression analysis that online usage does significantly affect profit, but a few more analyses still need to be performed before any strategizing begins.
There are two factors working against the decision to promote wider usage of the online channel. Firstly, Alan notes that the profit averages of online users and nonusers are different with the mean of internet bankers being slightly higher. The significance of this difference is tested in Exhibit 7, and it shows there is not enough evidence to reject the null hypothesis at 95% confidence level. Additionally, there is the issue of external validity. Nonetheless, it is still recommended that Pilgrim Bank should provide incentives to use the online channel as opposed to charging online fees for existing users.
While the difference in profit averages did not pass the 95% level, it does exceed the 70% level meaning that the difference is more likely to be significant than not. As for the external validity issue, it should be noted that only about a quarter of the sample did not have demographic records. Thus, the regression analysis is based on the bigger portion of the sample. Management at Pilgrim Bank should, nevertheless, be notified of this risk. Another point supporting the decision to offer incentives for online banking is that charging online fees may hurt profitability in the long run.
The introduction of these fees may boost profit from those who wish to remain with Pilgrim Bank in the short term, but at the same time, there will be customers who will switch banks because of high transaction costs. A regression analysis of profit with an additional variable of tenure years, shown in Exhibit 8, indicates that there is significant evidence to conclude that the longer a customer stays, the more profitable they become. Consequently, charging online fees will lead to the risk of losing customers who may become very profitable in the future.
To effectively implement the online banking promotion strategy, we need to determine any significant characteristics of online users. We run a regression analysis with online usage as the dependant variable and age, income, and region as the independent variables. The results, shown in Exhibit 9, indicate that age and income are significant variables, while geographic region is not. Since the age co-efficient is negative, Pilgrim Bank should focus more efforts to younger customers to migrate them to the online channel. The same should be done with customers in higher income brackets since the income co-efficient is positive.
Although we have determined the statistical significance of online users, the economical significance should also be reviewed. Offering incentives to do online banking will increase the load of the online channel. The management of Pilgrim Bank must carefully assess the estimated increase in online usage after the promotions to see if existing infrastructure can support the extra load. All costs associated with the increase of online bankers, including any new infrastructure needed to be built, will need to be compared with the expected increase in profit to determine the net value.
Only then will the economic value of this strategy be entirely addressed. Exhibit 1 Exhibit 2 Exhibit 3 |SUMMARY OUTPUT ONLINE VARIABLE | | | | | | | | | |Regression Statistics | | | | | | |Multiple R |0. 059 | | | | | | |R Square |0. 0000 | | | | | | |Adjusted R Square |0. 0000 | | | | | | |Standard Error |282. 8572 | | | | | | |Observations |22812. 000 | | | | | | | | | | | | | | |ANOVA | | | | | | | | |df |SS |MS |F |Significance F | | |Regression |1. 000 |64359. 4777 |64359. 4777 |0. 8044 |0. 3698 | | |Residual |22810. 0000 |1824986620. 0204 |80008. 1815 | | | | |Total |22811. 0000 |1825050979. 982 | | | | | | | | | | | | | | |Coefficients |Standard Error |t Stat |P-value |Lower 95% |Upper 95% | |Intercept |126. 5216 |2. 0072 |63. 0326 |0. 0000 |122. 5872 |130. 4559 | |Online |5. 0028 |5. 780 |0. 8969 |0. 3698 |-5. 9304 |15. 9360 | Exhibit 4 |SUMMARY OUTPUT ONLINE AND DEMOGRAPHICS VARIABLE | | | | | | | | | |Regression Statistics | | | | | | |Multiple R |0. 130 | | | | | | |R Square |0. 0454 | | | | | | |Adjusted R Square |0. 0452 | | | | | | |Standard Error |276. 3899 | | | | | | |Observations |22812. 000 | | | | | | | | | | | | | | |ANOVA | | | | | | | | |df |SS |MS |F |Significance F | | |Regression |4. 000 |82792772. 1376 |20698193. 0344 |270. 9493 |0. 0000 | | |Residual |22807. 0000 |1742258207. 3606 |76391. 3802 | | | | |Total |22811. 0000 |1825050979. 982 | | | | | | | | | | | | | | |Coefficients |Standard Error |t Stat |P-value |Lower 95% |Upper 95% | |Intercept |-104. 6908 |47. 1288 |-2. 2214 |0. 263 |-197. 0664 |-12. 3151 | |Online |16. 6854 |5. 5432 |3. 0101 |0. 0026 |5. 8204 |27. 5505 | |Age |27. 1774 |1. 1390 |23. 8617 |0. 0000 |24. 9450 |29. 4098 | |Income |18. 8810 |0. 7874 |23. 9784 |0. 0000 |17. 3376 |20. 244 | |District |0. 0130 |0. 0387 |0. 3350 |0. 7376 |-0. 0629 |0. 0889 | Exhibit 5 |Correlation Matrix of Different Variables | | |Profit |Online |Age |Income |District |Tenure | |Profit |1. 0000 | | | | | | |Online |0. 0059 |1. 000 | | | | | |Age |0. 1426 |-0. 1686 |1. 0000 | | | | |Income |0. 1466 |0. 0807 |-0. 0700 |1. 0000 | | | |District |0. 0015 |0. 0056 |-0. 0305 |0. 0257 |1. 0000 | | |Tenure |0. 1699 |-0. 0808 |0. 4203 |0. 0400 |-0. 0072 |1. 0000 | Exhibit 6 Data With Demographics | |Profit | |Online | | | | | | | |Mean |127. 1694 |Mean |0. 1295 | |Standard Deviation |282. 8560 |Standard Deviation |0. 3358 | |Count |22812. 000 |Count |22812. 0000 | | | | | | |Data Without Demographics | |Profit | |Online | | | | | | | |Mean |70. 9916 |Mean |0. 020 | |Standard Deviation |240. 3740 |Standard Deviation |0. 3027 | |Count |8822. 0000 |Count |8822. 0000 | | | | | | | | | | | |Testing if profit mean is the same | | | |Ho : ? = ? 2 |Ha: ? 1 ? ?2 | | | |s = 271. 68 | | | | |t = 16. 49 | | | | |d. f. = 31632 |Critical Value = 1. 960 | | | |Reject Ho, 16. 49>1. 60 | | | | | | | | |Testing if online usage is the same | | | |Ho : ? 1 = ? 2 |Ha: ? 1 ? ?2 | | | |s = 0. 10684 | | | | |t = 20. 097 | | | | |d. f. = 31632 |Critical Value = 1. 960 | | | |Reject Ho, 20. 5097>1. 960 | | | Exhibit 7 |Profit non online users | | | | |Mean |110. 862 | |Standard Deviation |271. 3010 | |Count |27780. 0000 | | | | |Profit Online Users | | | | |Mean |116. 6668 | |Standard Deviation |283. 6646 | |Count |3854. 000 | | | | |Testing if Profit is the same | |Ho : ? 1 = ? 2 |Ha: ? 1 ? ?2 | |s = 272. 836 | | |t = 1. 254 | | |d. f. = 31632 | | |Critical Value = 1. 960 (95% confidence level) | |Cannot Reject Ho, 1. 541. 036 | Exhibit 8 |SUMMARY OUTPUT WITH TENURE YEARS VARIABLE | | | | | | | | | |Regression Statistics | | | | | | |Multiple R |0. 396 | | | | | | |R Square |0. 0574 | | | | | | |Adjusted R Square |0. 0572 | | | | | | |Standard Error |274. 6444 | | | | | | |Observations |22812. 000 | | | | | | | | | | | | | | |ANOVA | | | | | | | | |df |SS |MS |F |Significance F | | |Regression |5. 0000 |104804631. 891 |20960926. 3778 |277. 8875 |0. 0000 | | |Residual |22806. 0000 |1720246347. 6090 |75429. 5513 | | | | |Total |22811. 0000 |1825050979. 4982 | | | | | | | | | | | | | | |Coefficients |Standard Error t Stat |P-value |Lower 95% |Upper 95% | |Intercept |-103. 9244 |46. 8312 |-2. 2191 |0. 0265 |-195. 7169 |-12. 1320 | |Online |18. 2421 |5. 5089 |3. 3114 |0. 0009 |7. 4442 |29. 0400 | |Age |18. 2882 |1. 2457 |14. 6816 |0. 000 |15. 8467 |20. 7298 | |Income |17. 8416 |0. 7848 |22. 7337 |0. 0000 |16. 3033 |19. 3799 | |District |0. 0101 |0. 0385 |0. 2631 |0. 7925 |-0. 0653 |0. 0856 | |Tenure |4. 0283 |0. 2358 |17. 0827 |0. 0000 |3. 661 |4. 4905 | Exhibit 9 |SUMMARY OUTPUT ONLINE AS DEPENDANT VARIABLE | | | | | | | | | |Regression Statistics | | | | | | |Multiple R |0. 822 | | | | | | |R Square |0. 0332 | | | | | | |Adjusted R Square |0. 0331 | | | | | | |Standard Error |0. 3302 | | | | | | |Observations |22812. 000 | | | | | | | | | | | | | | |ANOVA | | | | | | | | |df |SS |MS |F |Significance F | | |Regression |3. 0000 |85. 3369 |28. 4456 |260. 9619 |0. 000 | | |Residual |22808. 0000 |2486. 1401 |0. 1090 | | | | |Total |22811. 0000 |2571. 4769 | | | | | | | | | | | | | | |Coefficients |Standard Error |t Stat |P-value |Lower 95% |Upper 95% | |Intercept |0. 221 |0. 0563 |3. 9464 |0. 0001 |0. 1118 |0. 3324 | |Age |-0. 0337 |0. 0013 |-25. 0820 |0. 0000 |-0. 0363 |-0. 0310 | |Income |0. 0099 |0. 0009 |10. 6030 |0. 0000 |0. 0081 |0. 0118 | |District |0. 0000 |0. 0000 |-0. 1863 |0. 8522 |-0. 0001 |0. 0001 |