Professional Documents
Culture Documents
Business Analytics Paper CPS Products
Business Analytics Paper CPS Products
Products
Understanding Customers
1|Page
Introduction/Assessment/Recommendation
Corrugated Paper Sales (CPS Paper Products), an independent United Kingdom-based
paper manufacturer, operates in the paper and packaging markets. Its competitive edge in the
products. While the core business of CPS entails rewinding papers for corrugators, the book
publishing industry and the magazines industry mark the two main market segments of this
company. This firm makes sales either through brokers or directly to the customers.
CPS aims at formulating its marketing plan for the subsequent year. Establishing a better
understanding of its customers' attributes and the links between their CPS viewpoints and their
behaviour towards the organization underlies the development of an effective marketing scheme.
likelihood to recommend, likelihood to purchase, purchase level, and consideration for a strategic
customer size, customer type, distribution system, and customer region, to establish the existence
of any significant differences in these measures. Furthermore, it designs models to determine the
Salesforce image, product line, warranty and claims, competitive pricing, ordering and billing,
new products, delivery speed, and price flexibility. Consequently, the analysis would allow
identifying the most influential performance variables and enhancing the understanding of CPS
regarding their customers, thereby facilitating the formulation of the firm’s marketing plan for
the next year. The participants of this study included 200 purchasing managers from CPS’s
customer firms.
2|Page
From the situational analysis, significant distinctions in the outcome/relationship
variables subject to any of the customer characteristics variables exist. Moreover, different
outcome implies that the impact of the predictors on the dependent variables varies across the
study subjects. Therefore, CPS should address each measure of customer characteristics
distinctively when seeking to improve the purchase outcomes and relationship. Consequently,
the inputted effort would sufficiently match the required attention for each dimension of
Increases in customer duration and customer size are associated with increased customer
satisfaction. Therefore, CPS should institute strategies that favour customer retention. Besides, it
should target customers with a high number of employees since a positive relationship exists
between customer size and customer satisfaction. Moreover, since the distribution system
positively influences customer satisfaction, the firm should establish reliable and effective
product distribution.
satisfaction. Hence, CPS should ensure the production of high-quality paper products that satisfy
the needs of the clients and uphold a likable overall impression of its Salesforce to attract sales.
customer satisfaction, the company should cease focusing on improving its website since this
strategy may not pay-off the costs incurred. Remarkably, customers may fail to rely on the
3|Page
Improved product quality, technical support, and Salesforce image lead to an increased
likelihood of recommendation and likelihood to purchase. Therefore, CPS should ensure high-
quality paper products to increase its sales and customer base through referral clients.
Furthermore, the firm should intensify technical support for solving product or service issues to
raise customer satisfaction, and thereby trigger sales. Besides, the Salesforce image affects the
customer likelihood to recommendation and likelihood to purchase; hence, CPS should invest in
improving the appearance of its Salesforce. Finally, the warranty and claims services are
negatively related to the likelihood of recommendation; thus, this firm should reduce its input to
satisfaction and Likelihood to recommend) and any of the customer characteristics variables (the
type of customer, customer size, customer region, distribution system, and customer duration)
4|Page
not
assumed
As shown in table 1 above, the analysis revealed that customer satisfaction was no
significant difference between magazine and book customers at 0.05 level of significance (t
(198) = 1.049, p = 0.631) assuming equal variances. The analysis findings suggested that the
satisfaction rate between magazine customers was not different from that of book customers.
Conversely, the analysis discovered that customer satisfaction was a significant difference
between small customers (below 500) and large customers (500 and above) at a 0.05 level of
significance (t (198) = 2.781, p = 0.006) assuming equal variances (See table 2, Appendix).
Based on the analysis, large customers were more satisfied than small customers. Moreover, the
analysis identified a statistically significant difference between customers within Europe and
those outside Europe at a 5% level of significance (t (198) = 2.467, p = 0.014) assuming equal
variances. See table 3 in the appendix. The analysis surmised that European customers were
more satisfied than customers outside of Europe. Also, when the equal variance was assumed
between the two groups, the analysis revealed a statistically significant difference between the
distribution of products directly through a salesforce and indirectly through a broker (t (198) =
9.236, p < 0.005) (See table 4 in the appendix). This suggested that customers directly through
magazine and book customers at a 0.05 level of significance (t (198) = 0.072, p = 0.0.863) (see
table 5 in the appendix). This postulated that magazine customers recommended the same as
books customer. The analysis revealed that the Likelihood to recommend among small
customers (below 500) is significantly different from the Likelihood to recommend among large
customers (500 and above) at a 0.05 level of significance (t (198) = -4372, p < 0.05). Based on
5|Page
the analysis, large customers are likely to recommend more than small customers (see table 6 in
appendix). Nevertheless, according to the analysis displayed in table 7 (see the appendix), the
Likelihood to recommend among customers within Europe was not significantly different from
that of customers from outside Europe (t(198) = 0.937, p = 0.350). This postulated that the
Likelihood to recommend among European customers was the same as those of customers
outside Europe. It was evident from the analysis displayed in table 8 (see appendix) that the
salesforce and indirectly through a broker (t(198)= -7.41, p < 0.05). Based on the analysis,
One-Way ANOVA
Table 2: One-Way ANOVA for each of the Response variable and Customer Duration
Source Dependent Variable Type III Sum of d Mean F Sig.
Squares f Square
Corrected Q19 - Satisfaction 164.311 a
2 82.156 113.794 .00
Model 0
Q20 - Likely to 71.043 b
2 35.521 43.112 .00
Recommend 0
Intercept Q19 - Satisfaction 9676.798 1 9676.798 13403.324 .00
0
Q20 - Likely to 9673.560 1 9673.560 11740.626 .00
Recommend 0
Q1 Q19 - Satisfaction 164.311 2 82.156 113.794 .00
0
Q20 - Likely to 71.043 2 35.521 43.112 .00
Recommend 0
According to one-way ANOVA the analysis results (Table 9) for each outcome variable
with customer duration, satisfaction was significantly different across customer duration
(F=113.79, p < 0.05). Besides, likely to recommend was also found to be significantly different
across customer duration (F=43.11, p < 0.05). Further analysis (presented in table 10) was
6|Page
needed to determine the pairwise comparison of satisfaction and Likelihood to recommend
The multiple comparisons using Tukey HSD portrayed in table 10 above, indicated that
customer satisfaction was significantly different between customers with less than one year and
those with less or more than five years (p < 0.05). Moreover, it was also identified that customers
with one to 5 years had significantly different satisfaction from those with longer than five years.
7|Page
Similarly, customers with less than one year had a significantly different likelihood to
recommend than customers with less or more than five years (p < 0.05). Conversely, customers
with one to five years had no significant difference from those with longer than five years (p =
0.12).
Multiple regression analysis was used to develop two models used to predict customer
satisfaction and the likelihood to recommendation from the perceptions of the performance
product line, salesforce image, competitive pricing, warranty & claims, new products, order &
Satisfaction Model
Satisfaction model was developed using customer satisfaction as the response variable
and perceptions of the performance as the predictor variables. The R-Square of 0.804 suggested
that perceptions of the performance variables can explain 80.4% of the customer satisfaction
variation. Only 19.6% of the variation is explained by other variables not included in the model
8|Page
B Std. Error Beta
1 (Constant) -2.255 .789 -2.859 .00
5
Q6 - Product Quality .412 .038 .459 10.87 .00
9 0
Q7 - E-Commerce -.264 .087 -.163 -3.044 .00
3
Q8 - Technical Support .031 .046 .041 .669 .50
5
Q9 - Complaint Resolution .099 .073 .097 1.366 .17
3
Q10 - Advertising -.034 .047 -.031 -.714 .47
6
Q11 - Product Line .347 .186 .368 1.862 .06
4
Q12 - Salesforce Image .629 .068 .572 9.194 .00
0
Q13 - Competitive Pricing -.064 .033 -.082 -1.965 .05
1
Q14 - Warranty & Claims .002 .089 .002 .026 .97
9
Q15 - New Products .033 .028 .040 1.185 .23
7
Q16 - Order & Billing .030 .073 .022 .413 .68
0
Q17 - Price Flexibility .263 .193 .252 1.358 .17
6
Q18 - Delivery Speed .008 .375 .005 .022 .98
2
a. Dependent Variable: Q19 - Satisfaction
Based on the regression analysis displayed in table 12 above, only product quality, E-
commerce, and salesforce image had a significant contribution to the predicting power of the
model at 0.05 level of significance (p < 0.05). All the other perceptions of the performance
variables did not significantly contribute to the predicting power of the model (p>0.05). The
analysis revealed that customer satisfaction when all the perceptions of the performance
9|Page
variables are zero was a negative value (-2.255), suggesting that customers have a negative
attitude towards the products without perceptions of performance. The coefficients of the
significant predictor variables suggest that controlling for all the other variables, increasing
product quality by one unit, increases customer satisfaction by 0.412. similarly, increasing
salesforce image by one unit increases customer satisfaction by 0.629. Conversely, controlling
for all the other variables, increasing E-commerce by one unit decreases customer satisfaction by
0.264. Although most of the performance variables' perceptions have no significant contribution
in predicting customer satisfaction, they were used as confounding variables in the model.
pricing; Q14=Warranty & claims; Q15=New products; Q16=Order & billing; Q17=Price
products to potential clients (Khadka and Maharjan, 2017; Jumawan 2018). The ability of a
business organization to grow without regarding or acknowledging the needs of customers seems
impossible. Seemingly, increasing product quality and salesforce image and decreasing E-
likelihood to recommend as the response variable and perceptions of the performance as the
predictor variables. The R-Square of 0.536 suggested that performance variables' perceptions can
10 | P a g e
explain 53.6% of the likelihood to recommend variation. And 46.4% of the variation is explained
by other variables not included in the model (see table 13 in the appendix).
11 | P a g e
The multiple regression analysis displayed in table 14 revealed that only four perceptions of
performance variables (product quality, technical support, salesforce image, and warranty & claims) had a
significant contribution to the model's predicting power at a 0.05 level of significance (p <0.05). All the
other variables had no significant contribution in predicting the likelihood to recommend (p>0.05).
When all the perception of performance variables is set to zero, the likelihood to recommend would be
0.808. Holding all the other variables constant, a unit increase in product quality and technical support
would increase the likelihood to recommend by 0.344 and 0.214, respectively. Similarly, increasing the
salesforce image by one unit would increase the likelihood to recommend customers by 0.463. However,
increasing warranty & claims by one unit would decrease the likelihood to recommend by 0.302. Since all
the other performance variables' perceptions had no significant contribution in predicting likelihood to
recommend, they were considered controlling variables. Therefore, the likely to recommend model was
as shown below:
pricing; Q14=Warranty & claims; Q15=New products; Q16=Order & billing; Q17=Price
Therefore, it can be summarized that increasing product quality, technical support, and
salesforce image and decreasing warranty & claims would facilitate the likelihood to recommend
other customers.
The analysis revealed to increase customer satisfaction with CPS would be improved by
increasing the quality of the product and salesforce image and decreasing E-commerce.
12 | P a g e
Moreover, to enhance the likelihood of recommending other customers for CPS, the firm needs
to increase its quality, technical support, and salesforce image and decrease warranty & claims.
Therefore, CPS needs to devise strategies on lowering E-commerce and decreasing warranty and
claims to facilitate customer satisfaction and likelihood to recommend, respectively. CPS should
ensure the production of high-quality paper products that satisfy the needs of the clients and
uphold a likable overall impression of its Salesforce to attract sales. Customers prefer high-
quality products and services due to the possibility of fewer losses (Hoe and Mansori 2018;
Waluya, Iqbal, and Indradewa, 2019); hence, such products remain instrumental in promoting
social change. Additionally, the salesforce image substantially affects customer satisfaction. The
sales team should maintain a long-term customer relationship and possess customer orientation
for the customer to perceive high Salesforce cooperation (Lussier and Hall 2018; Xu, Yu, and
Zhang 2018). The staff should present themselves as experts to enhance relationship outcomes.
Moreover, CPS should focus on improving the quality of the products, salesforce image, and
technical support when developing its business plan for the next year. They should also examine
other influencing factors not included in the model that would affect the likelihood to
13 | P a g e
References
Casielles, R.V., and Alvarez, B.A., 2007. Consumers' characteristics and brand choice behaviour:
Dendena, B., 2019. An assessment of the draft beer distribution system in light of customer
satisfaction: The Case of Heineken Brewery Share Company in Addis Ababa (Doctoral
Hoe, L.C. and Mansori, S., 2018. The effects of product quality on customer satisfaction and
Imran, M., Hamid, S., Aziz, A. and Hameed, W., 2019. The contributing factors towards e-
Javed, F. and Cheema, S., 2017. Customer satisfaction and customer perceived value and its
Jumawan, J., 2018. The Effect of Service Quality on Loyalty using Satisfaction as an Intervening
Khadka, K. and Maharjan, S., 2017. Customer satisfaction and customer loyalty: Case
Lussier, B. and Hall, Z.R., 2018. Cooperation in B2B relationships: Factors that influence
69, pp.209-220.
14 | P a g e
Park, C., Jun, J., Lee, T. and Lee, H., 2018. Customer orientation or employee orientation:
Which matters more? The moderating role of firm size. Journal of Business & Industrial
Marketing.
Waluya, A.I., Iqbal, M.A. and Indradewa, R., 2019. How product quality, brand image, and
Xu, W., Yu, Y. and Zhang, Q., 2018. An evaluation method of comprehensive product quality
15 | P a g e
Appendix
16 | P a g e
Figures and Charts: Customer Region and Distribution System
17 | P a g e
Table 7: Independent Samples Test for satisfaction and size of customer
Equal variances -
192.629 .006 -.4801 .1719
not assumed 2.792
Equal variances
2.617 197.114 .010 .4355 .1664
not assumed
Equal variances -
192.012 .000 -1.3630 .1478
not assumed 9.225
18 | P a g e
Table 10: Independent Samples Test for Likelihood to recommend and type of customer
Equal variances
.072 197.792 .943 .0110 .1535
not assumed
Table 11: Independent Samples Test for Likelihood to recommend and customer size
Equal variances -
197.916 .000 -.6412 .1465
not assumed 4.377
Table 12: Independent Samples Test for Likelihood to recommend and customer region
Equal variances
.995 197.202 .321 .1462 .1470
not assumed
19 | P a g e
Table 13: Independent Samples Test for Likely to recommend and distribution system
Equal variances -
195.924 .000 -1.0099 .1356
not assumed 7.445
Change Statistics
1 .732
a
.536 .503 .7632 .536 16.506 13 186 .000
20 | P a g e