In total, 10 employees (the number of all employees) of the sales department working with this CRM system in the company took part in the survey. The data was processed through a computer program for statistical data processing SPSS statistics.

Table 2

Survey «Usage of CRM system» by employees of sales department in Orgkhim

An analysis of the survey results showed that CRM functions such as “Execute any necessary task to-be-done throughout the buyer's journey” and “Develop long-term customer value delivery and profitability models” are least used by company employees, while “Access buyer's history and current journey in the buying process in real-time ” employees use to a greater extent. This may be due to a lack of time to learn how to use the tool, or to individual factors, and personal understanding the process of work. Values of variance and standard deviation are minimal, which may indicate the reliability (typical) of the average value. The coefficient of variation (CV) is less than 33%, that is, the sample is considered homogeneous, and the opinion of experts is unambiguous.

Table 3

Descriptive statistics - CRM usage by sales managers in the Orgkhim

After the implementation of CRM in company, a customer satisfaction survey was conducted for the company's customers using the SERVQUAL model in order to look at the dynamics of changes and the mutual influence of the system and positive customer ratings. Clients were also asked to rate the statements characterizing the product or service on a scale of 1 (not at all satisfied) to 5 (completely satisfied).

Table 4

Survey «Customer satisfaction of Orgkhim clients» using SERVQUAL model

In total, 30 companies took part in the survey - the number of regular customers working with Orgkhim. The data was processed through a computer program for statistical data processing SPSS statistics. It should be noted a relatively small sample as a statistical survey, in which the sample was formed from a relatively small number of units. This is due to the fact that the company operates on the B2B market and has a limited number of regular customers. Also, due to the remote work schedule in connection with the current situation, not all companies were able to promptly submit their questionnaires. However, even a limited amount of data allowed to draw conclusions about the relationship between CRM and customer satisfaction and answer a research question, using further analyses. SERVQUAL's goal is to measure the extent of the gap between Customer Expectations and Customer Perception (that is, the actual situation). Despite the fact that almost all companies rated their expectations for the maximum score, the company managed not only to correct past shortcomings in terms of delivery, prompt replies to customer wishes and build competent communication with clients, but also exceed customer expectations. A Multi-Attributes Rating Scale approach was used to measure customer satisfaction of the clients. Overall satisfaction (OS) of every customer was counted as sum of multiplication of gap between expectations score and actual quality rating and weight (the sum of each rating for service attribute and then dividing the sum by the total points of all attributes). Attributes in this particular study equals to statements of SERVQUAL model.

Picture 13 Multi-Attributes Rating Scale approach

OS which is negative would demonstrate overall satisfaction of specific client whereas a positive OS which is > than 1 would propose that client is disappointed with organization's administrations. Based on this analysis, it is possible to classify clients into different blocks like: Block 1(Poor) = customers whose OS are greater than or equal to 1; Block 2 (Fair) = customers whose OS are between 0,5 and 1; Block 3 (Good) = customers whose OS are between 0 and 0,5; Block 4 (Excellent) = customers whose OS are less than or equal -1.

Table 5

Estimation of overall customer satisfaction by the example of two Orgkhim's clients

It is obvious to notice, the OS of two clients is less than 0, which means that the service provided by Orgkhim to its customers exceeds expectations in general. 10 companies by the results of the outcomes were included in the Block 4 (Excellent), 16 companies in the Block 3 (Good) and 4 companies in the Block 2 (Fair), which may indicate that the company has improved all its past performance, and it is likely that this is largely due to the CRM system.

Table 6

Results of survey «Customer satisfaction of Orgkhim clients»

The analysis of the survey results showed that customers found some shortcomings in the quality of the products, in accordance with the technical specifications stated by the client, as well as in the ability to respond to the client's request for dissatisfaction with the quality of the product or service within three days. In addition, a relatively small score was set for the company's ability to change components in the product at the request of the client. This may be due to problems in the process of agreeing on the product specification or the client himself has a faulty laboratory device, which is why the data on the quality standard criteria vary. It is important to note that the minimum score of "3" was set in rare cases, which does not show that there is a systemic problem with the parameters indicated above, but speaks only of some isolated flaws that can be quickly eliminated. However, indicators such as timely delivery and communication with the client, which relate to the Reliability and Responsiveness blocks, have significantly improved compared to the previous results obtained in the 2018 survey, which may indicate a positive impact CRM for customer satisfaction. Values ??of variance and standard deviation are minimum, which may show the reliability (typical) of the average value. The coefficient of variation (CV) is less than 33%, that is, the population of the sample is considered homogeneous, and the opinion of experts is unambiguous.

3.3 Regression analysis

A linear regression model was used to test the hypotheses in this research. Linear regression is used to see whether there are linear relationships between variables and how explanatory variables affect the dependent variable (also called Y variable).

Null Hypothesis (H0)

H0: The independent variables “CRM usage” have no effect on the dependent variable “Customer satisfaction”.

Alternative Hypothesis (H1)

H1: At least one independent variable “CRM usage” has an effect on the dependent variable “Customer satisfaction”.

Regression testing of estimation of clients is used for each dimension on frequency of using CRM by managers who work with these customers with focus on particular CRM functions through Python 3.

Picture 14 Regression analysis

In total, 30 pairs of Client-Manager turned out (all managers collaborated with certain clients). The highlighted values ??have a positive coefficient and are significant, which may indicate that there is a correlation between the effectiveness of the use of CRM tools by managers and the dimensions characterizing customer satisfaction. It will be incorrect to talk about the results of other indicators, since their values ??turned out either insignificant or become negative when the control variable is introduced. The regression analysis was performed 4 times using each factor score as a dependent variable with the CRM usage variables as independent variables. The factor “tangibility” was excluded from performing of regression analysis as CRM does not have any impact on physical facilities, equipment, and appearance of personnel. The functional form of a linear regression looks as follows:

Y_i =б + в₁Q_1i + в₂Q_2i + в₃Q_3i+ в₄Q_4i+ в₅Q_5i+ е_{i where:}

· Q1i - Score from 1 to 5 to the question «Access buyer's history and current journey in the buying process in real-time. »;

· Q2i - Score from 1 to 5 to the question «Anticipate buyer's needs at a given time in order to make next best action recommendations and offers. »;

· Q3i - Score from 1 to 5 to the question «Execute any necessary task to-be-done throughout the buyer's journey. »;

· Q4i - Score from 1 to 5 to the question «Develop long-term customer value delivery and profitability models. Access buyer's history and current journey in the buying process in real-time. »;

· Q5i - Score from 1 to 5 to the question «Assess buyer's journey outcomes and create personalized customer experience. »;

· Yi - Average score of Q1i, Q2i, Q3i, Q4i, Q5i on 4 dimensions

Table 7

Interpretation of relevant coefficients for regression factor score “Responsiveness”

Table 8

Interpretation of relevant coefficients for regression factor score “Assurance”

Table 9

Interpretation of relevant coefficients for regression factor score “Empathy”

Table 10

Interpretation of relevant coefficients for regression factor score “Reliability”

To check consistency and stability of results it was controlled on extra working hours per week because there is possibility that employee works better only because he/she spends more time at work. The functional form of a linear regression with added control variable looks as follows:

Yi =б + в1Q1i + в2Q2i + в3Q3i+ в4Q4i+ в5Q5i+ дWi+ еi where

Wi - extra working hours of manager

The results after control variable was added are following:

Table 11

New interpretation of relevant coefficients for regression factor score “Responsiveness”

Picture 15 Regression analysis with control variable

Table 12

New interpretation of relevant coefficients for regression factor score “Assurance”

Table 13

New interpretation of relevant coefficients for regression factor score “Empathy”

Table 14

New interpretation of relevant coefficients for regression factor score “Reliability”

Conclusions of Chapter 3

The most stable result showed the effect of the CRM function as “Anticipation of buyer's needs at a given time in order to make next best action recommendations and offers” on the empathy factor, since even with the introducing of a control variable in the form of extra working hours, the results remained unchanged, and the value of the variable itself was not significant which might lead to the fact that Hypothesis 1 was proved. Moreover, the stable results demonstrated in the correlation between CRM functions as “Access buyer's history and current journey in the buying process in real-time”, “Anticipate buyer's needs at a given time in order to make next best action recommendations and offers” and “Execute any necessary task to-be-done throughout the buyer's journey” on the responsiveness factor, since even with the introducing of a control variable in the form of extra working hours, the results remained unchanged. There are no doubts that CRM function as attention to buyer's needs at a given time has positive impact on customer satisfaction in general. The rest of variables showed unstable results (decreasing or being insignificant) which means that added control variable “extra working hours” has better impact on almost all factors: the more sales managers work and spend time on some particular tasks regardless to CRM functions - the better results they get integrating with customers.

Conclusion

The globalization, acceleration of the internet, and overflow of information in the twentieth century means for companies that in order for them to remain relevant, they have to become smarter in managing information and customer information. Customer relationship management is a combination of people, processes and technology that work together to collect customer information and use it to bridge more intimate relationships with the company's customers. The ultimate goal of the current study is to discover how CRM usage by the employees in Orgkhim Biochemical Holding affect their customer's satisfaction with the service and product quality level. The evaluation model was built around a regression analysis where the relationship between CRM usage and customer satisfaction is studied, i.e. finding a linear relationship between the variables. The SERVQUAL tool helped in gathering and interpreting customer satisfaction with the service quality of the dealerships while CRM usage was measured based on an inquiry created for employees of sales department in Orgkhim company. he regression analysis was then performed 4 times for each customer satisfaction dimension, except of “tangibility”. The regression's overall results showed weak correlation between customer satisfaction and the CRM actions in the company. In fact, only two customer satisfaction dimensions “Empathy” and “Responsiveness” were affected by the CRM variables to great extent and proved H1: At least one independent variable “CRM usage” has an effect on the dependent variable “Customer satisfaction”.

There are some factors which may have affected the results of this study. Firstly, data collected for this research entails very small sample sizes (10 and 30), whereas there is a general agreement that bigger samples are better (results are more accurate). This is due to the fact that the company operates on the B2B market and has a limited number of regular customers. Also, due to the remote work schedule in connection with the current pandemic situation, not all companies were able to promptly submit their questionnaires. Secondly, the CRM system implemented in Orgkhim company is relatively new and one could say it has not matured long enough to show a significant tangible return on customer satisfaction. It is reasonable to think that actions can be taken faster to change aspects in the dimensions “Empathy” and “Responsiveness” than say dimension “Tangible” since these dimensions explain more the attitude of the company's employees whereas physical objects like equipment and facilities require an up-front investment to change. Perhaps, it takes longer and more data to see a correlation between CRM and the other variables. Descriptive statistics have also shown that the employees have not started using the CRM tools to a great extent yet. Research should accumulate more data over a longer time period to overcome these limitations.

All in all, I believe the technique presented in this study is better for identifying which variables (actions) have an effect on customer satisfaction rather than an evaluation model for a CRM system. The company should employ the technique after a longer period of time after they collected more data. The management should then exploit the actions that show to have a significant effect on customer satisfaction and drop those that have a negative effect. The regression model presented here in the study is useful in identifying which actions are correlated with the customer satisfaction. So, if the strategy of the CRM system would change in the future and/or new tasks were implemented, new regression models could be used to identify if the new variables are linearly related to customer satisfaction. When the company has built a personal link with their customers, it is easier for them to identify the actual needs of customers and serve them in a more effective way. Organizations should consider seriously to invest in CRM tools and systems to implement in their workspace.

In addition, I can recommend the company to work on following actions while continuing CRM implementation:

1. Integrate CRM with other tools as much as possible. Integrations simplify working with a CRM system. The revision will combine all the tools for working with the client in a "single window". Managers will be able to process customers faster. For example, integration with messengers: customers can leave applications and communicate with company employees in messengers convenient for them: Telegram, Facebook, Vkontakte, WhatsApp. Sales staff do not need to open different messengers. They respond to requests from the CRM system. Correspondence is saved in the activity stream. The client can write an appeal where it is more convenient for him, write from different managers - messages are loaded into the deal.

2. Sales managers must monitor employees on a daily basis. It is necessary to gradually train managers to work with analytics: how to track plans, how to make effective decisions based on indicators. They should get used to regularly checking the work with transactions: listen to calls, check the completion of cards, track the reasons for refusing a purchase. Only such a control system can bring results.

Размещено на Allbest.ru