Digital Transformation and Automation: Application and

Effectiveness of Chatbots in Bulgarian Entrepreneurship

Ana Todorova

and Irina Kostadinova

University of Ruse “Angel Kanchev”, 8 Studentska Street, Ruse, Bulgaria

Keywords: Chatbots, Efficiency, Bulgarian Entrepreneurs, Chatbot Solutions, Artificial Intelligence.

Abstract: In view of the growing importance of digitalisation, this study examines the adoption and use of chatbot

solutions based on artificial intelligence among Bulgarian entrepreneurs. The aim is to identify the main

reasons for their implementation, the perceived benefits and challenges encountered, and the main areas of

application. A survey was conducted among 401 business owners in Bulgaria. The results show that 38%

(154) of the surveyed entrepreneurs use chatbots and the analysis focuses on their data. The main factors

driving implementation are increased efficiency, improved customer service and reduced costs. Improved

operational efficiency and customer satisfaction were identified as the leading benefits. However,

entrepreneurs face significant challenges, mainly related to the ability of chatbots to handle complex queries,

the need for staff training and customer acceptance. Chatbots are most often used in marketing and sales, but

they also have applications in customer service, internal processes and data analysis. In conclusion, despite

the still relatively low adoption rate among all respondents, Bulgarian entrepreneurs with experience using

chatbots see clear benefits related to efficiency and customer service. Successful expansion of this technology

requires addressing current challenges, especially in terms of functionality and the human factor.

1 INTRODUCTION

The advancements in artificial intelligence (AI) and

the global proliferation of chatbots (CBs) have led to

their significant rise across various industries,

highlighting their potential to improve customer

service (CS), streamline operations, and drive

business growth. CBs offer 24/7 availability, efficient

processing of routine tasks, cost-effective solutions,

and fast response times (McClune, 2024). In addition,

modern AI CBs can continuously learn from

interactions, provide personalised experiences, and

automate multiple processes (SAP, 2024).

In a broad sense, CBs can be defined as software

that accepts natural language as input and generates

natural language as output by engaging in a

conversation. Another definition emphasises their

attempt to resemble a human-like character and

defines them as interactive virtual characters whose

mission is to assist people in high-profile

environments. In addition to engaging in written

conversations (text-based CBs), CBs can also imitate

https://orcid.org/0009-0007-2993-077X

https://orcid.org/0000-0001-8845-7598

human speech (voice-based CBs) to improve the user

experience and cultivate customer loyalty. CBs can

be found on websites, social media, or instant

messaging applications. They can be deployed within

an organisation to support various services and

processes, such as internal support systems, IT

service management, training, or human resource

management (Miklosik et al., 2021). According to

one of the most established developers of

technological innovations, integrating AI CBs into

business processes can lead to a 45% increase in

customer trust and loyalty through transparent and

responsible AI practices (ESG with AI, n.d.).

At the same time, the Bulgarian entrepreneurial

environment is characterised by a significant role of

small and medium-sized enterprises (SMEs) as key

drivers of economic growth (Ministry of Transport,

2020). The country’s digital transformation efforts

and the growing importance of technology adoption

for competitiveness are evident (Digital Watch

Observatory, 2020). However, a number of analyses

highlight the relatively lower levels of digital skills

Todorova, A. and Kostadinova, I.

Digital Transformation and Automation: Application and Effectiveness of Chatbots in Bulgarian Entrepreneurship.

DOI: 10.5220/0013684600004000

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 17th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2025) - Volume 2: KEOD and KMIS, pages

225-236

ISBN: 978-989-758-769-6; ISSN: 2184-3228

225

and technological adoption in Bulgarian SMEs

compared to the European Union (EU) average and

identify them as both challenges and opportunities

(Ministry of Transport, 2020). While CBs adoption is

growing, the level of sophistication and degree of

integration can vary significantly among Bulgarian

SMEs. The emergence of local CBs platforms and

language-specific AI models indicates an adapted

approach to the Bulgarian market. This creates an

opportunity to explore the specifics of CBs adoption

and effectiveness in this context.

This study focuses on the need to understand the

extent and effectiveness of CBs implementation

among Bulgarian entrepreneurs. The main objectives

include collecting data from business owners and

managers through a questionnaire to assess their

perspective on the use of CBs and their effect on

various aspects of their business. The report is

structured by first analysing the status and benefits of

CBs implementation, as well as analysing sample

indicators to measure their effectiveness. The

research methodology and the design of a

questionnaire that examines the benefits and

challenges of and against the implementation of CBs

in the Bulgarian context are presented.

2 THEORETICAL

BACKGROUND

AI-powered systems have many potential

applications in decision support, manufacturing

automation, learning, communication, etc.

Communication between online users and

organisations is shifting towards interactions with AI-

powered systems. CBs are just such an example of a

technology being used in computer-mediated

communication, where AI agents are increasingly

taking on roles that humans once performed. The

advantage of deploying AI-powered CBs is that they

create the impression of intelligence as they become

smarter with increased data and user interactions

(Miklosik et al., 2021).

CBs powered by AI are used in a variety of

business areas, such as customer communication,

marketing activities and sales processes, which leads

to the optimisation of time and resources. In a study

conducted by CBInsights and analysing the 12 most

significant CS technologies, AI CBs are identified as

a top priority for businesses. Since the challenges and

current trends observed internationally are

comparable to those in the Bulgarian business

environment, local companies are expected to invest

in this technology. The goal is to improve customer

interaction and other operational aspects. Estimates

show that most AI CBs can independently process

between 60% and 70% of incoming inquiries (Umni,

n.d).

Azmi et al. (2023) also define a chatbot as an AI

application designed to simulate human conversation

through text or voice interactions. It is typically used

to provide automated customer support, answer

frequently asked questions, assist with tasks, and

engage in conversation with users naturally and

conversationally. According to the authors, CBs can

operate in two main types: rule-based and AI-driven.

Rule-based CBs follow predefined rules and patterns

to respond to user inputs; however, AI-driven CBs

use machine learning algorithms to better understand

and generate responses based on vast amounts of

training data. CBs can complement business

intelligence systems in several ways. In essence, CBs

provide a natural language interface that allows users

to interact with the business intelligence system using

natural language. They can help users formulate

queries, select appropriate visualisations, and

interpret results, allowing users to extract information

on their own.

This makes it easier for non-technical users to

access and analyse data about the company or

organisation. CBs can offer contextual information

and suggestions during conversations or decision-

making processes. Therefore, CBs could provide

personalised analysis and recommendations

according to user preferences, patterns, and

behaviour. Routine tasks in a business intelligence

system, such as generating reports, scheduling data

refreshes, or monitoring key performance indicators

(KPIs), can be easily automated by CBs, which allow

for continuous learning from user interactions and

improving responses using machine learning

techniques (Interoperable Europe, 2024; Microsoft,

2025).

The pandemic has stimulated the adoption of

digital services among Bulgarian businesses, which

had not widely adopted CBs until then. This has

become a necessity due to challenges such as the lack

of employees and the demand for self-service options

by customers. In the context of the EU digital agenda,

digitalisation is becoming a top priority for Bulgaria

and its business sector (Interoperable Europe, 2024).

A survey shows that a significant part of companies

(47%) intend to invest more in digital technologies

than in the past year. About a third will maintain the

same volume of investments, and a tiny percentage

(2%) foresee a decrease. These data clearly

KMIS 2025 - 17th International Conference on Knowledge Management and Information Systems

226

demonstrate increasing digital activity in the majority

of industries (Umni, n.d.).

Available information shows a growing interest

and adoption of CBs among Bulgarian enterprises,

especially SMEs, driven by the need to optimise

customer communication and improve business

processes. There are concrete examples of Bulgarian

companies using CBs in various sectors such as

tourism, beauty services and e-commerce (Umni,

n.d). In addition, the development of Bulgarian-

specific AI CBs such as BgGPT shows a growing

local capacity and focus (AI4Green, 2024). Also, the

cost reduction and labour savings are significant

motivators for the adoption of CBs among Bulgarian

entrepreneurs. The need for 24/7 customer support

and improved response times plays an equally

important role (Umni, n.d). The automation of routine

tasks and the release of human resources for more

complex issues are other key factors. At the same

time, the desire to improve customer engagement and

provide personalised experiences is also driving the

adoption of CBs (Maderis, 2024).

Many sources point to a growing adoption of CBs

in Bulgaria (Ministry of Transport, 2020; SAP, 2024;

AI4Green, 2024), including basic, rule-based CBs for

handling frequently asked questions and simple

queries. Alongside these, AI-based CBs with natural

language processing are being implemented to

facilitate more complex interactions. These CBs are

finding applications in CS, sales, marketing and

internal communication in Bulgarian SMEs (SAP,

2024).

However, the study shows that there is a gap

between the global trend of CBs use and the specific

context of Bulgarian SMEs, suggesting a potentially

lower level of understanding. The global increase in

CBs adoption is undoubtedly due to their numerous

benefits (McClune, 2024), but lower digital skills in

Bulgaria (Ministry of Transport, 2020) may affect the

ease of CBs adoption and perceptions of their

effectiveness.

The effectiveness of CBs in a business context can

be assessed using several key dimensions. One of

them is efficiency, which includes aspects such as

response time, problem resolution rate, and capacity

to handle a large number of queries. Another critical

dimension is cost reduction, which can manifest itself

as savings in CS and reduced employee workload.

Customer satisfaction is also essential and can be

measured through metrics such as user satisfaction

and positive feedback. CBs can also influence

employee efficiency through the automation of

routine tasks and the ability to focus on more complex

issues. Finally, the percentage of completed goals,

such as customer inquiries successfully handled,

leads generated, or sales made, is an essential

indicator of CBs effectiveness.

There are specific quantitative KPIs that can be

used to quantify the performance of CBs. These

include the total number of interactions, the average

chat duration, and the percentage of goals completed.

Other important metrics include the number of

misunderstood lines and the rate of escalation to a

human (Behl, 2024). Conversion rate, retention rate,

error rate, and return on investment (ROI) are also

critical metrics. In addition, metrics such as bot

experience score, bot automation score, and

automated conversation cost can be tracked.

Interaction volume, bounce rate, conversation length,

and handling time also provide valuable information

(Calabrio, n.d.).

3 METHODOLOGY

The research methodology is based on a survey

conducted between April 20 and May 10, 2025, and

aimed at owners of micro, small and medium-sized

enterprises, as well as large companies with over 250

employees. The study sample is formed based on the

number of owners of Bulgarian companies.

According to data from the National Statistical

Institute of Bulgaria (NSI, 2024), the total number of

non-financial enterprises in Bulgaria for 2023 is

462,752. Therefore, the minimum recommended size

of the study is 383 respondents, with a confidence

level of 95% and a margin of error of 5%. It is

important to emphasise that the minimum sample size

does not guarantee a number of respondents that fully

meet the scope of the study, such as companies that

have implemented CBs solutions. At the moment, the

latter is impossible to achieve, since there is no data

on how many enterprises in Bulgaria have

implemented CBs solutions.

The questionnaire was developed based on the

main challenges, benefits and indicators for

measuring the effects of implementing CBs identified

in the theoretical background. This also determines

the content of the main statements in the

questionnaire, including the following topics: reasons

for implementation, implementation process,

perceived benefits and challenges, impact on business

goals, and satisfaction with the work of the CBs.

Respondents are invited to respond to the formulated

statements on a scale from 1 to 5, where 1 is the

lowest and 5 is the highest possible score.

Regarding the question exploring the areas of

application of CBs – "Where exactly in your company

Digital Transformation and Automation: Application and Effectiveness of Chatbots in Bulgarian Entrepreneurship

227

have you implemented CBs solutions?", the following

options are included, with the possibility of selecting

more than one:

• Customer Service / Support:

On the company website (to answer frequently

asked questions, direct customers, collect information

for requests); On social networks (e.g. Facebook

Messenger, Instagram DMs – to answer inquiries,

provide information); On a mobile application (if

available – for support and navigation); For initial

processing of customer requests before handing them

over to an employee;

• Sales / Lead Generation:

On the website to qualify visitors and collect

contacts; To recommend products/services based on

customer needs; To schedule demonstrations or

consultations; To process orders or reservations (less

common for complex processes, but possible for

standard ones);

• Marketing:

To engage website visitors (e.g. through

interactive quizzes, collecting opinions); To promote

new products/services or special offers; To segment

the audience and personalise messages; To conduct

surveys and collect feedback;

• Internal processes / Employee support:

IT support (to answer common technical

questions, collect information about problems);

Human Resources (HR) (to answer questions about

leave, benefits, internal procedures; to onboard new

employees); To access internal information and

databases (e.g. to check availability, project statuses);

• Data collection and analytics:

For automated collection of opinions and

feedback from customers; For analysis of the most

common queries and problems;

• Other (please specify):

This option is very important to capture specific

or innovative applications that you did not foresee;

• We have not implemented chatbot solutions.

The survey was distributed through email

newsletters from various professional networks,

social networks, communication channels, and

personal contacts of the study authors. The main

research questions (RQ) that the study seeks to

answer are:

RQ1: What are the main reasons for

implementing chatbot solutions?

RQ2: What benefits and challenges do Bulgarian

entrepreneurs identify?

4 RESULTS

The survey included 401 respondents. According to

the size of the organisation, 122 (30.4%) represent

companies with less than 10 employees, 187 (46.6%)

– with between 11 and 50 employees, 69 (17.2%) –

with between 51 and 250 employees, and 23 (5.7%) –

with over 250 employees. Therefore, the main profile

of the survey participants is that of representatives of

micro and SMEs in Bulgaria.

The organisations that have not yet implemented

CBs solutions are mainly representatives of micro-

companies – 82 (33%), small organisations – 97

(39%), medium-sized enterprises – 49 (20%), and 19

large companies (8%). Probably the micro and small

enterprises cannot secure access, underestimate or at

the highest level ignore CBs solutions as

opportunities for improved performance. It is also

possible that, according to the specifics of their

activities, as well as the current level of success

achieved, these organisations do not currently need

CBs solutions in their activities.

Of those who have implemented CBs solutions,

40 (26%) are representatives of micro companies, 90

(58%) are representatives of small organisations, 20

(13%) are medium-sized enterprises, and 4 (3%) fall

within the definition of large companies.

No respondent chose the opportunity to indicate

an option other than those already suggested for the

question "Where exactly in your company have you

implemented CBs solutions?". Of the total number of

respondents, 247 (62%) stated that they have not yet

implemented CBs solutions in their activities. With

the possibility of selecting more than one answer, the

remaining 154 (38%) respondents indicated different

areas of application (Figure 1). The data shows that

Bulgarian entrepreneurs perceive CBs primarily as

tools for improving external communication and

operations, with marketing and lead generation being

the leading application areas. While CS remains an

important aspect, it is not the dominant function.

Figure 1: Areas of application of

CBs

according to

Bulgarian entrepreneurs. Source: Own development.

KMIS 2025 - 17th International Conference on Knowledge Management and Information Systems

228

Table 1: Descriptive statistical analysis of the data generated in the study. Source: Own development.

Descriptive statistical analysis

–

number of responses* (N), minimum value (Min), maximum value (Max), average value (Mean), standard deviation (Std. Deviation)

N Min Max Mean Std. Deviation

To what extent did the following factors influence your decision to implement a CBs?

Cost Reduction

154 2 5 3,98 ,718

Improving CS

153 2 5 4,09 ,729

Increasing Efficiency

154 2 5 4,10 ,697

Task Automation

154 2 5 4,12 ,660

Competitive Pressure

154 1 5 3,96 ,740

To what extent do you agree with the following statements about the benefits of implementing a CBs?

Reduce CS costs

154 1 5 3,56 ,758

Increase the efficiency of business operations

154 2 5 3,75 ,811

Automate routine tasks efficiently

154 2 5 3,63 ,741

Improve customer satisfaction

153 2 5 3,69 ,719

To what extent do you agree with the following statements about the challenges of implementing a CBs?

We encountered difficulties integrating it with existing systems

154 1 5 3,44 ,870

The CBs struggled with complex queries

154 1 5 3,80 ,820

Maintaining the CBs’ knowledge base required significant effort

154 1 5 3,54 ,864

Our employees needed significant training to use and manage the CBs

154 1 5 3,63 ,885

Customer adoption of the CBs was lower than expected

154 1 5 3,61 ,843

To what extent did the CBs contribute to achieving the following business goals?

Increased Sales

154 1 5 3,32 ,830

Improved Lead Generation

154 1 5 3,55 ,768

Improved Customer Retention

153 1 5 3,46 ,835

Reduced Operating Costs

153 1 5 3,48 ,828

Improved Employee Productivity

153 1 5 3,43 ,809

Please answer to what extent...:

…are you satisfied with the level of integration of the CBs with your existing systems?

152 1 5 3,31 ,893

…were the available resources (e.g. budget, technical expertise) sufficient for the

implementation of the CB?

154 1 5 3,50 ,865

…was the process of implementing the CB in your organisation easy?

154 1 5 3,37 ,893

…are you satisfied with the overall performance of the CB?

152 1 5 3,42 ,917

…are you satisfied with the ability of the CB to understand and accurately respond to

customer inquiries in Bulgarian?

154 1 5 3,46 ,894

…are you satisfied with the information and data provided by the CB 's analytics?

153 1 5 3,38 ,874

*The number of responses (N) in Table 1 may differ from the actual number of respondents (N=154), as respondents were given the option to answer or

not to the relevant question/statement, according to their business environment.

The use of CBs applications for internal processes

and analytics is still in its infancy, suggesting

potential for future growth and diversification in the

Bulgarian business environment. This distribution

reflects the general trends in global CBs adoption,

where the initial focus is often on marketing and sales

funnel automation.

The descriptive analysis of the survey data,

presented in Table 1, shows that Bulgarian

entrepreneurs are implementing CBs primarily to

improve task automation, increase efficiency, and

improve CS. Competitive pressure and the desire to

reduce costs are also strong drivers, but there is also

greater diversity in the responses of Bulgarian

entrepreneurs.

Although there is even more dispersion in the

answers, the next section of statements/questions

supports this data to some extent. Respondents do not

clearly identify one of the benefits mentioned. With

the lowest support, but still close to the others, is the

reduction of costs. The fluctuations here may be the

result of the fact that technological innovations,

which in themselves require significant investment,

are unlikely to lead to cost reductions generally. Or at

least not as quickly as, for example, improving the

efficiency of business operations, which stands out

with the highest degree of support. Based on the data

generated by the study, it is possible to search for and

discover interesting correlations (Table 2) that could

reveal important insights into the impact and

outcomes of CBs' implementation.

Digital Transformation and Automation: Application and Effectiveness of Chatbots in Bulgarian Entrepreneurship

229

Table 2: Relationship between reasons for implementation and achieved business goals. Source: Own development.

Improved Customer

Retention

Improved Lead

Generation

Increased

Sales

Improve customer

satisfaction

Improving CS

Pearson

Correlation

,805** ,768** ,850** ,770**

Sig. (2-tailed) ,000 ,000 ,000 ,000

N 153 153 153 153

** Correlation is significant at the 0.01 level (2-tailed).

Reduce CS

costs

Increase the

efficiency of

business operations

Automate

routine tasks

efficiently

Increased

Sales

Improved

Employee

Productivity

Task

Automation

Pearson

Correlation

,762** ,802** ,749** ,846** ,788**

Sig. (2-tailed) ,000 ,000 ,000 ,000 ,000

N 154 154 154 154 153

** Correlation is significant at the 0.01 level (2-tailed).

Table 3: Regression Model Summary: Impact of Achieved Business Goals on Chatbot Integration Satisfaction. Source: Own

development.

Model Summary

Model R R Square Adjusted R Square Std. Error of the Estimate Durbin-Watson

1 ,940

,883 ,879 ,310 ,570

a. Predictors: (Constant), [Improved Employee Productivity], [Improved Lead Generation], [Increased Sales], [Improved Customer

Retention], [Reduced Operating Costs]

b. Dependent Variable: [To what extent are you satisfied with the level of integration of the CBs with your existing systems?]

Table 4: Analysis of variance (ANOVA) of the regression model: Impact of achieved business goals on satisfaction with

chatbot integration. Source: Own development.

ANOVA

Model

Sum of Squares df Mean Square F Sig.

1 Regression 106,392 5 21,278 220,724 ,000

Residual 14,075 146 ,096

Total 120,467 151

a. Dependent Variable: [To what extent are you satisfied with the level of integration of the CBs with your existing systems?]

b. Predictors: (Constant), [Improved Employee Productivity], [Improved Lead Generation], [Increased Sales], [Improved Customer

Retention], [Reduced Operating Costs]

The correlation analysis presented in Table 2

examines the relationships between two main reasons

for implementing CBs – "Improving CS" and "Task

automation" – and various achieved business goals.

For the analysis, Pearson correlation coefficients (r)

were calculated, with the statistical significance of all

correlations assessed at the p<.01 level (two-tailed).

The results reveal that the drive to improve CS is

strongly and statistically significantly associated with

multiple positive business outcomes. A robust

positive correlation was found between "Improving

CS" and "Increased sales" (r=.850, N=153, p<.001),

suggesting that efforts in this direction contribute

significantly to growth in commercial activity.

Similarly, substantial positive relationships were

reported with "Improved Customer Retention"

(r=.805, N=153, p<.001), "Improved Lead

Generation" (r=.768, N=153, p<.001), and "Improved

Customer Satisfaction" (r=.770, N=153, p < .001).

These findings underscore the pivotal role of focusing

on customer experience in implementing CBs to

achieve a broad range of desired business objectives

directly related to the customer base.

In parallel, the analysis demonstrates that task

automation, as a motive for implementing CBs, is

also strongly and statistically significantly associated

with achieving key operational and financial goals. A

robust positive correlation was found between "Task

Automation" and "Increase in the efficiency of

business operations" (r=.802, N=154, p<.001), which

is a direct reflection of the expected benefits of

automation. Significant positive relationships were

also found with "Reduce CS costs" (r=.762, N=154,

p<.001), "Effective automation of routine tasks"

(r=.749, N=154, p<.001), and "Improved employee

productivity" (r=.788, N=153, p<.001). It is also

KMIS 2025 - 17th International Conference on Knowledge Management and Information Systems

230

noteworthy to observe the strong positive correlation

with "Increased sales" (r=.846, N=154, p<.001),

which suggests that optimising internal processes

through automation indirectly contributes to revenue

growth.

To examine the impact of achieved business goals

on satisfaction with the level of chatbot integration, a

multiple linear regression analysis was conducted. In

this model, the dependent variable is " To what extent

are you satisfied with the level of integration of the

CBs with your existing systems?". Five achieved

business goals were included as independent

variables (predictors): [Improved Employee

Productivity], [Improved Lead Generation],

[Increased Sales], [Improved Customer Retention],

and [Reduced Operating Costs].

The results of the model summary are presented

in Table 3. The model demonstrates a strong multiple

correlation between the predictors and the dependent

variable, with a correlation coefficient of R=.940. The

most important indicator, the coefficient of

determination (R

), is .883. This value indicates that

approximately 88.3% of the total variance in

satisfaction with the level of chatbot integration is

explained by the five achieved business goals

included in the model. The adjusted coefficient of

determination (Adjusted R

) is .879, which is a more

conservative estimate and confirms the high

explanatory power of the model. The standard error

of the forecast (Std. Error of the Estimate) is .310,

which indicates the average magnitude of the error in

the model's predictions. The value of the Durbin-

Watson statistic is 0.570, which may indicate the

presence of positive autocorrelation in the residuals,

an aspect that warrants further study.

To assess the overall statistical significance of the

regression model, an analysis of variance (ANOVA)

was conducted, the results of which are presented in

Table 4. The observed F-statistic is 220.724 at 5 and

146 degrees of freedom (F(5,146)=220.724). The

corresponding significance level is p<.001. Since this

p-value is significantly less than the accepted

significance threshold of α=0.05, the null hypothesis

that all regression coefficients (except the constant)

are equal to zero is rejected.

This conclusion confirms that the regression

model as a whole is statistically significant.

Collectively, the achieved business goals included in

the model make a statistically significant contribution

to explaining the variance in satisfaction with the

level of integration of the chatbot with existing

systems. The high explanatory power of the model

(R²=.883) combined with its overall statistical

significance (p<.001) strongly suggests that

achieving these business goals is a powerful predictor

of higher satisfaction with the integration aspects of

chatbot solutions.

In summary, the results of the correlation analysis

between the variables in Table 2 strongly indicate that

both the orientation towards improving CS and

towards task automation in implementing CBs are

associated with significant achievement of the

respective business goals. All the correlations

examined are strong, with coefficients ranging from

.749 to .850, and are highly statistically significant. It

is essential to note that although these correlations

suggest strong associations between the variables,

they do not establish causal relationships, but rather

indicate that these phenomena are observed

simultaneously and in a similar direction within the

studied population.

5 DISCUSSION

Bulgarian entrepreneurs identify the ability of CBs to

process complex queries as the biggest challenge.

This is particularly important in the context of the

Bulgarian language, as satisfaction with the ability of

CBs to understand and respond in Bulgarian has until

recently been rated as relatively low. Other

significant challenges include maintaining the

chatbot’s knowledge base, staff training, and

customer acceptance. Integration with existing

systems is a challenge, but less pronounced than the

others, which indirectly indicates that CBs

development most likely follows relatively universal

and easily applicable implementation principles.

Of interest here is the identified problem with user

acceptance of CBs. The answers vary significantly,

and the predominant rating is between 3 and 4. This

reported non-acceptance by users may be due to high

expectations on the part of entrepreneurs, problems

with the language capabilities of the respective

chatbot, or overly complex requests (queries) from

customers. In all three cases, entrepreneurs should

improve the information security of their digital

platforms. For example, suppose users are forced to

send overly complex queries through chatbot

systems. In that case, the corresponding online

application (website, social network, etc.) does not

offer the necessary quantity and quality of

information to satisfy the needs of users without the

need for automated queries. In any case, it is essential

to identify not only the rejection itself (which is

obviously already done by some respondents), but

also the reasons that led to the rejection.

Digital Transformation and Automation: Application and Effectiveness of Chatbots in Bulgarian Entrepreneurship

231

The overall satisfaction with CBs is moderate to

high in terms of their contribution to business goals

(lead generation, customer retention, sales, costs,

productivity). The diversity in the answers is also

understandable, given that each business unit pursues

different business goals. The majority report

improved customer acquisition and retention, which

is the primary mission of chatbot solutions. Cost

reduction and improved employee efficiency are also

reported as significant results of the implementation

of CBs solutions. Accordingly, the lowest support is

for increased sales, and the reasons for this can again

be sought in various directions. It is possible that the

company in question does not have a correct sales

tracking system in place or, by default (or due to

insufficient understanding), the capabilities of CBs

are limited solely to a communication tool.

On the other hand, satisfaction here with regard to

sales can be defined as low and even unsatisfactory,

given that Bulgarian entrepreneurs indicate that the

main area of application is precisely sales, not only

attracting potential customers (Figure 1). Therefore,

although chatbot solutions generally perform the

functions expected of them, there is a discrepancy

between the goals set and achieved with the help of

CBs.

The "Please answer to what extent..." section

contains more general questions related to the

experience with CBs. In terms of satisfaction with the

level of integration of the chatbot with the existing

systems, an average value of 3.31 is reported. This

indicates moderate satisfaction, especially

considering that the integration itself is shown as not

that challenging. Satisfaction with the information

and data provided by the chatbot's analytics and with

the chatbot's ability to understand and accurately

respond to customer inquiries in Bulgarian are also

rated with average values. Satisfaction with the

analytical data is moderate. The assessment of

language capabilities confirms that there are

significant problems with understanding and

responding in Bulgarian, especially with complex

inquiries, as reported in the previous questions.

An interesting fact is that only satisfaction (with

an average score of 3.5 / Table 1) that the available

resources were sufficient for the implementation of a

chatbot solution, theoretically exceeded the average

values and tends to a good rating. Despite significant

variance in respondents' answers, this response shows

that entrepreneurs approach the implementation of

technological solutions relatively well-off

financially.

The study also shows that the implementation of

CB solutions in the modern business landscape is

motivated by a wide range of strategic goals aimed at

optimising processes, improving customer

experience and achieving competitive advantages.

Respondents from Bulgaria identify "Cost

Reduction", "Improving CS", "Increasing Efficiency"

and "Task Automation" as the leading motives for

implementing CBs. These data correspond to general

global trends, where the automation of routine

processes and cost optimisations in the field of CS are

often the primary drivers for investments in AI and

chatbot technologies. These findings also reveal

significant similarities with global trend results,

providing valuable context for understanding

regional trends.

According to Umni (n.d.), a leading developer of

chatbot solutions for the Bulgarian market, as of

2020, only 30% of companies worldwide consider

themselves to be lagging in implementing innovative

technologies. Notably, this percentage was even

lower for Bulgaria, with only 25% of companies

feeling the need to invest in digital tools. The

company concludes that this relative digital

conservatism is one of the reasons for the accelerated

rise of AI CBS outside Bulgaria. They also emphasise

that, against the backdrop of general market

conservatism and still limited knowledge of the

technology in the country, the use of AI CBs provides

a significant opportunity for Bulgarian companies to

differentiate themselves from their competitors.

Supporting this growing trend, the current study finds

that 38% of respondents have already implemented

chatbot solutions. These solutions are mainly used in

the field of marketing and sales, but they also play a

significant role in CS, internal processes, and data

collection and analysis.

Research suggests that CBs can bring significant

benefits to companies, especially in the area of CS.

According to a Gartner study, approximately 30% of

the companies surveyed have already implemented or

have plans to use CBs. A report by CapGemini also

confirms this trend, indicating that almost half of the

top hundred companies in the banking and insurance

sectors have already implemented CBs. In Norway,

where this study is conducted, several large

corporations and government organisations have

implemented CBs in recent years as part of their

digitalisation strategies. As a result of this process,

there has also been an increase in the number of local

chatbot platform providers in the country (Zhang et

al., 2021).

In the Romanian context, academic studies

(Anton et al., 2024; Clim, 2025) also emphasise the

leading role of efficiency and optimisation in

customer interactions as key motivators for the

KMIS 2025 - 17th International Conference on Knowledge Management and Information Systems

232

adoption of chatbot solutions. Research in the

Romanian banking sector shows that banks are

adopting CBs "to improve CS by offering fast and

accurate answers to their queries", aiming for a

"positive impact… on customer satisfaction". This

goal has been successfully achieved, demonstrating

that improving CS is a leading motive and a notable

result in Romania. Furthermore, the implementation

of CBs "increases the efficiency of operations" by

"automating routine tasks" and "reducing waiting

times and personnel costs". More broadly, a survey of

Romanian SMEs confirms that "increasing the

efficiency of business operations" and "reducing

costs" are the primary drivers for implementing AI

technologies, including CBs, which also contribute to

enhancing "customer interactions" and "customer

satisfaction". These results demonstrate a strong

correlation between the strategic motives for

implementing CBs in Bulgaria and Romania,

explicitly focusing on operational efficiency, cost

reduction, and customer experience optimisation.

The use of chatbots in Croatia is expanding

dynamically in both the public and private sectors.

During the COVID-19 pandemic, the healthcare

chatbot “Andrija” was used by over 30,000 citizens,

with over 75% expressing satisfaction (Petričević &

Mustić, 2022). In the corporate sector, 65% of

Croatian companies implement chatbots for customer

service, automating around half of queries (EOS

Group, 2021). Students at the University of North

actively use AI chatbots such as ChatGPT and

Copilot for training, while a survey of employees in

IT companies showed that 95.7% have used

ChatGPT, and over 30% use it regularly

(Szombathelyi et al., 2023; Horvat et al., 2025).

Despite the lack of nationally representative data,

these studies point to a growing use of and trust in

chatbots in Croatia, in line with global trends in AI.

Users appreciate their speed, efficiency, and

accessibility, although concerns about security and

data privacy remain.

At the same time, the businesses in Croatia are

also actively researching and investing in AI and

chatbot solutions. One notable example is the

Croatian startup Splix.ai, which successfully raised

almost €2 million in a pre-seed funding round. This

company is developing a chatbot security platform

based on AI, which highlights the growing emphasis

on security and trust challenges in implementing

these technologies (Kostanic, 2024). This

demonstrates not only the adoption of CBs but also

the development of the ecosystem to support their

secure and efficient operation in the country.

In the context of user engagement, Croatian

research highlights that the primary reasons users

choose to use CBs are their convenience, speed, and

24/7 availability. The use of CBs can lead to higher

user satisfaction. These results suggest that

companies implementing CBs are likely seeking to

enhance CS and reduce costs by automating routine

tasks, aligning with trends observed in other countries

in the region (Dobrinić et al., 2021; Horvat et al.,

2025).

A comparative analysis of results between

Bulgaria and other Balkan countries, such as

Romania and Croatia, is essential due to the shared

economic and socio-cultural context in the region.

This commonality of factors ensures higher relevance

of comparisons to markets with different maturity and

stages of digital transformation, such as those in

Western Europe.

Such cross-country comparisons enable the

identification of specific regional trends in the

adoption of technologies, such as CBs. They can be

used to establish whether the motivations for

implementation and the achieved results are universal

for the Balkans or whether there are unique factors

influencing the success of chatbot solutions in

specific countries.

Furthermore, such a comparison provides

valuable benchmarking for businesses in the region,

enabling them to compare their achievements with

those of competitors or similar markets. This can

contribute to better strategic positioning and

adaptation of chatbot solutions to local needs and

challenges. Finally, given the often limited

specialised scientific literature for smaller markets in

the region, comparative analysis helps to build a more

comprehensive picture and compensate for this lack

of local data.

The analysis performed allows answering the

formulated research questions. Regarding the first

RQ1: What are the main reasons for implementing

chatbot solutions? the study shows that the main

reasons for implementing chatbot solutions by

Bulgarian entrepreneurs are related to operational

optimisation (efficiency, automation, cost reduction)

and improving market position (improving CS,

competitive advantage), with risk management also

appearing to play a significant role. Bulgarian

entrepreneurs identify key benefits such as increased

operational efficiency, improved customer

satisfaction and automation of routine tasks. At the

same time, they face significant challenges, mostly

related to the processing of complex queries by the

chatbot, the need for staff training, customer

acceptance and maintaining the knowledge base. This

Digital Transformation and Automation: Application and Effectiveness of Chatbots in Bulgarian Entrepreneurship

233

also provides a satisfactory answer to the second

research question – RQ2: What benefits and

challenges do Bulgarian entrepreneurs identify?

Based on this data, the authors conclude that the

Bulgarian entrepreneurs and CBs developers should

focus on:

• Improve the understanding and processing of

complex queries, primarily in Bulgarian. This may

include investments in more advanced AI models

specifically trained for Bulgarian, or hybrid systems

that route complex queries to human operators;

• Optimise the processes for maintaining the

chatbot knowledge base to reduce the effort required

to update it.

• Develop more effective employee training

programs on the use and management of CBs.

• Strategies to improve customer acceptance of

CBs by emphasising the benefits and convenience

they offer and ensuring a seamless user experience.

Limitations of the study

This study has certain limitations that should be taken

into account when interpreting the results. First, the

lack of detailed information about the field of activity

of the surveyed companies and the specifics of their

operation (e.g., predominantly online or offline

presence, geographical location – city/rural) limits the

possibility of a more nuanced analysis of the factors

influencing the adoption and effectiveness of CBs

solutions in different business contexts. Similarly, the

study does not collect data on the specific type of CBs

used (e.g., rule-based, AI, hybrid) and the duration of

their implementation, which could affect the reported

effectiveness.

Secondly, a limitation arising from the method of

distribution of the questionnaire survey should be

noted. The approach used does not guarantee

complete coverage of the entire population of

Bulgarian entrepreneurs using CBs technologies,

which may introduce a certain degree of selective

under-exhaustion and affect the representativeness of

the sample. Therefore, the results should be

considered indicative of the surveyed group, and

generalisations for the entire entrepreneurial

community in the country should be made with due

caution.

Additionally, during the statistical analysis, a

variation in the number of respondents included (N)

was observed across the different correlation

calculations. In the majority of the analyses, the

number of cases was N = 154, but in specific

correlation pairs, it decreased to N = 153. This

difference is a result of applying a method for

handling missing data known as pairwise deletion.

The pairwise deletion method assumes that each

correlation coefficient is calculated by including only

those respondents who have valid data for both

specific variables involved in the given calculation.

Unlike "listwise deletion", which excludes the entire

respondent if they are missing even one response in

the whole set of variables, pairwise deletion allows

the use of a larger amount of available information.

In this case, the variation of one respondent (154 vs.

153) indicates that one respondent provided a

response for one of the variables but omitted a

reaction for the other variable included in the

correlation. Adopting pairwise deletion has its

advantages, as it maximises the use of available data,

especially when the number of missing values is

small and scattered. However, it is essential to note

that, although insignificant in this case, it can

potentially lead to the calculation of correlations

based on slightly different samples, which requires

caution in interpretation, especially if the number of

missing data points is larger or not random. In the

present study, given the minimal difference in N and

the high statistical significance of the correlations,

this approach is not expected to affect the validity and

reliability of the conclusions drawn significantly.

Applicability

The present study contributes to expanding the

scientific knowledge on digital transformation in the

Bulgarian entrepreneurial sector by providing initial

empirical data on the level of adoption and

subjectively assessed effectiveness of chatbot (CBs)

technologies. It identifies the main areas of

application of these tools and, despite its

methodological limitations, lays the foundation for

future, more in-depth research on the factors

determining the successful integration and return on

investment in automated communication solutions in

the specific national context.

In practical terms, the study's results offer

valuable guidance for Bulgarian entrepreneurs. These

data can help them make informed decisions about

investments in chatbot technologies, focusing on the

proven benefits that arise from improved CS, task

automation, and increased efficiency. The identified

implementation challenges, such as integration

difficulties and the need for knowledge base

maintenance, allow entrepreneurs to anticipate

potential obstacles and proactively plan resources and

strategies to overcome them. Understanding user

expectations and perceptions of CBs, including

factors that drive their satisfaction, can help create

more effective and user-centric solutions, ensuring a

higher level of adoption.

KMIS 2025 - 17th International Conference on Knowledge Management and Information Systems

234

For chatbot solution providers, the study provides

clear guidance on the specific needs and areas of

interest in the Bulgarian market. The high correlation

between addressing challenges and overall chatbot

satisfaction highlights that developing solutions that

are easy to integrate, effective in handling complex

queries, and supported by an intuitive knowledge

base is key to market success.

Providers can use these results to adapt their

products and services, offering more targeted

functionalities, improved implementation services,

and tailored training that meets the current needs of

Bulgarian companies. This will enable them to

establish stronger partnerships and drive broader

chatbot adoption.

For policymakers, this report provides an

empirical basis for understanding the dynamics of

digital transformation in the Bulgarian

entrepreneurial sector. The findings can inform the

development of government strategies and programs

to support businesses in implementing innovative

technologies. Policies can be aimed at stimulating

investments in automated communication solutions,

ensuring access to training programs to address

technical challenges, and increasing digital literacy.

Supporting the research and development of solutions

tailored to the local context and language-specificities

(such as the Bulgarian language) can also be a priority

to ensure the maximum applicability and benefit of

chatbot technologies for the Bulgarian economy.

Furthermore, the study highlights the need for

further research into good practices and challenges in

the effective use of CBs, which is beneficial for

consultants and training organisations.

6 CONCLUSION

The growing global importance of CBs contrasts with

the slower adoption rate and potentially lower level

of understanding in the Bulgarian context, which is

influenced by factors such as digital skills and

specific market conditions. However, there are clear

signs of growing interest and adoption of this

technology among Bulgarian enterprises, driven by

the potential to reduce costs, improve customer

service and increase efficiency.

The proposed questionnaire aims to collect

valuable data to understand the reasons for

implementation, the process, perceived benefits and

challenges, as well as the impact on business

objectives and daily work. The analysis of the

benefits and challenges of implementing CBs in

Bulgaria reveals that despite significant potential

advantages, there are also country-specific barriers

related to technical expertise, language and cultural

characteristics. The consideration of the human factor

underscores the importance of adopting CBs by both

employees and customers. Ensuring adequate

employee training and effective customer

communications is crucial for the success of these

technologies.

Statistical analysis further supports these findings,

demonstrating strong and significant correlations

between successfully overcoming challenges and

high satisfaction with various aspects of chatbot

solutions. It was found that companies that coped

better with integration difficulties, managed complex

queries, and maintained a knowledge base

experienced greater satisfaction with the chatbot's

implementation and overall performance. Regression

analysis revealed that achieved business goals, such

as improved employee productivity and reduced

operating costs, collectively explain a significant

portion of the variance in satisfaction with chatbot

integration, highlighting the direct relationship

between achieved results and a positive perception of

the technology.

In conclusion, CBs represent an essential tool for

the digital transformation of Bulgarian entrepreneurs.

To realise their full potential, careful consideration of

the specific conditions of the Bulgarian market is

necessary, along with continuous monitoring and

evaluation of their effectiveness.

Future research should focus on gaining a deeper

understanding of the specific needs and impact of

CBs in various sectors of the Bulgarian economy, as

well as on conducting longitudinal studies to track the

long-term effectiveness and return on investment.

The proposed recommendations and overcoming the

study's limitations will enable, on the one hand, more

in-depth analyses of the topic, and on the other hand,

informed business decisions in this rapidly evolving

field.

ACKNOWLEDGMENTS

This study was funded by the Scientific Research

Fund of the University of Ruse "Angel Kanchev",

Bulgaria.

REFERENCES

AI4Green. (2024). BgGPT: Bulgaria's AI chatbot benefits

business, education, and government. Retrieved May 6,

2025, from https://ai4greensme.eu/news/bulgarian-ai-

Digital Transformation and Automation: Application and Effectiveness of Chatbots in Bulgarian Entrepreneurship

235

chatbot-bggpt-bulgarias-ai-chatbot-benefits-business-

education-and-government/

Anton, C. E., Ciobanu, E., Brătucu, G., & Bucs, L. (2024).

Using Chatbots to Enhance Integrated Reporting:

Insights from Accounting and Consultancy Companies

from Romania. Electronics, 13(23), 4801. DOI:

10.3390/electronics13234801

Azmi, M., Mansour, A., & Azmi, C. (2023). A context-

aware empowering business with AI: Case of chatbots

in business intelligence systems. Procedia Computer

Science, 224, 479–484. DOI:

10.1016/j.procs.2023.09.068

Behl, A. (2024). Unleashing Success: 7 Key Metrics to

Monitor for a Chatbot Implementation. Mastech

InfoTrellis. Retrieved May 6, 2025, from

https://mastechinfotrellis.com/blogs/key-metrics-to-

look-out-for-while-implementing-chatbot

Calabrio. (n.d.). The chatbot performance metrics every

team should be measuring. Retrieved May 6, 2025,

from https://www.calabrio.com/wfo/contact-center-

ai/key-chatbot-performance-metrics/

Clim (Moga), L. M. (2025). The Role of Chatbots in the

Growth of Virtual Businesses in Romania: Implications

for Integrated Reporting. EuroEconomica, 44(1), 7–16.

Retrieved from https://dj.univ-danubius.ro/index.php/

EE/article/view/3328

Digital Watch Observatory. (2020). Bulgaria's AI

development concept (2020-2030). Retrieved May 6,

2025, from https://dig.watch/resource/bulgarias-ai-

development-concept-20202030

Dobrinić, D., Gregurec, I., & Dobrinić, D. (2021). Attitudes

of Croatian consumers about mobile messenger

chatbots. Journal of Information and Organizational

Sciences, 45(2), 579–599. DOI: 10.31341/jios.45.2.13.

EOS Group. (2021). Chatbot survey 2021: Two-thirds of

European companies already use chatbots. EOS

Solutions. https://eos-solutions.com/en/newsroom/pr

essreleases/20211117

ESG with AI. (n.d.). Home page. Retrieved May 6, 2025,

from https://esg-with-ai.com/

Horvat, T., Havaš, L., Valdec, D., & Crčić, D. (2025). The

use of AI chatbots in stem education: a study on

University north students in Croatia. Proceeding of the

17th International Conference on Education and New

Learning Technologies. pp. 3190-3197, ISBN: 978-84-

09-74218-9. DOI: 10.21125/edulearn.2025.0848

Interoperable Europe. (2024). Open-source AI for

Bulgaria: INSAIT's BgGPT can foster public and

private sector innovation. Retrieved May 6, 2025, from

https://interoperable-europe.ec.europa.eu/collection/pu

blic-sector-tech-watch/open-source-ai-bulgaria-insaits-

bggpt-can-foster-public-and-private-sector-innovation

Kostanic, A. M. (2024). Croatia’s SplxAI raises €1.8M to

enhance AI chatbot security amid growing cyber

threats. The Recursive. Retrieved May 6, 2025, from

https://therecursive.com/croatia-splxai-2m-pre-seed-

round-chatbot-security-ai/

Maderis, G. (2024). Top 22 benefits of chatbots for

businesses and customers. Zendesk. Retrieved May 6,

2025, from https://www.zendesk.com/au/blog/5-

benefits-using-ai-bots-customer-service/

McClune, S. (2024). Customer chatbots: The good, the bad,

and the ugly. Beyond Encryption. Retrieved May 6,

2025, from https://www.beyondencryption.com

/blog/customer-chatbots-good-bad-ugly

Microsoft. (2025). Empowering Bulgaria's future: AI

transformation and Microsoft's vision. CEE Multi-

Country News Center. Retrieved May 6, 2025, from

https://news.microsoft.com/en-cee/2025/02/04/empow

ering-bulgarias-future-ai-transformation-and-microso

fts-vision/

Miklosik, A., Evans, N., & Qureshi, A. M. A. (2021). The

use of chatbots in digital business transformation: A

systematic literature review. IEEE Access, 9, 106530–

106539. DOI: 10.1109/ACCESS.2021.3100885

Ministry of Transport. (2020). Kontseptsiya za razvitieto na

izkustveniya intelekt v Balgariya do 2030 g. [Concept

for the development of artificial intelligence in Bulgaria

until 2030]. Retrieved from https://www.mtitc.gov

ernment.bg/sites/default/files/koncepciyazarazvitienaii

vbulgariyado2030.pdf

NSI. (2024). Broy na predpriyatiyata [Number of

enterprises]. Natsionalen statisticheski institute.

Retrieved May 6, 2025, from https://bit.ly/3H4wYNO

Petričević, S., & Mustić, D. (2022). Communicating a

global pandemic with WhatsApp and Healthbot in

Croatia. In Proceedings of the International Conference

“Grid Technologies & Applications” (GRID 2022).

Novi Sad, Serbia: University of Novi Sad, Faculty of

Technical Sciences, Department of Graphic

Engineering and Design

SAP. (2024). What is a chatbot. Retrieved May 6, 2025,

from https://www.sap.com/bulgaria/resources/what-is-

a-chatbot

Szombathelyi, D., Jakopec, T., & Selthofer, J. (2023,

September). The impact of using ChatGPT on

employees of Croatian IT companies. In Proceedings of

the 34th Central European Conference on Information

and Intelligent Systems (CECIIS 2023) (pp. 473–480).

Dubrovnik, Croatia: Faculty of Organization and

Informatics, University of Zagreb.

Umni. (n.d.). The unlimited potential of AI chatbots in

Bulgaria. Retrieved May 6, 2025, from https://

umni.bg/en/blog/the-unlimited-potential-of-ai-chatbot

s-in-bulgaria/

Zhang, J. J. Y., Følstad, A., & Bjørkli, C. A. (2021).

Organizational factors affecting successful

implementation of chatbots for customer service.

Journal of Internet Commerce, 22(1), 122–156. DOI:

10.1080/15332861.2021.1966723

KMIS 2025 - 17th International Conference on Knowledge Management and Information Systems

236