Barriers

Lack of Access to Specialized Training and Resources

Many SMEs face significant challenges in acquiring the knowledge necessary for effective AI integration. With limited access to specialized training and resources, many companies have turned to self-learning as a way to bridge the gap. One respondent highlighted this approach, stating, “I handle everything with my knowledge of economics and business management (...) I learned from YouTube” (ManuEnt1). While platforms like YouTube offer valuable resources, they often lack the depth and personalized guidance required for strategic AI adoption. Another interviewee expressed the difficulties of relying solely on self-directed learning, adding, “The research we spend days doing could likely be done in a few hours by someone who knows AI. We are struggling on our own. Training would be very helpful” (ManuEnt3). This highlights the limitations of informal learning methods and the growing need for more tailored, expert-led education.

​

Even though there is a clear need for education, as one interviewee stated, "Yes, we need training (…) But how do we use it [AI]? It’s still unclear to us. Of course, we want to use it and receive training, but how do we adapt it to our company? How do we get immediate results, especially fast?"(ManuEnt3) many SMEs struggle to find affordable and effective solutions: "When we ask other companies for help, they give us ridiculously high prices" (WelEnt1).

​

Moreover, some point out that the available training often falls short of their needs: "Iresearched some educational institutions claiming to provide AI training. I personally met with the people who would provide the training, but unfortunately, the information is often no more than what you could find with a simple Google search. They lack the detailed data and32 understanding needed to guide people effectively." (TechEnt1). This reflects the broader challenge of finding both accessible and high-quality training that can truly support businesses in integrating AI technologies.

 

Lack of Financial Resources

A significant barrier to AI adoption in SMEs is the lack of financial resources, which restricts their ability to invest in the necessary tools and infrastructure. Interviewees highlighted that limited budget forces SMEs to prioritize immediate operational needs over digital transformation initiatives. One respondent noted, “We can not put our focus on AI only, we would have the financial resources as long as we can keep our traditional way of making money (...) we are tight on budget when it comes to such digitalization” (LegEnt1). This illustrates the challenge SMEs face in balancing ongoing business operations with the financial demands of AI implementation.

 

Additionally, SMEs often lack the capital to hire dedicated specialists, resulting in existing employees taking on multiple roles. As one interviewee explained, “We have to do everything with our own capital and savings (...) One person often has to do the work of 3 or 5” (ManuEnt3). This financial constraint forces SME owners and employees to balance multiple responsibilities, limiting their ability to focus on strategic areas like AI. Moreover, costs associated with data preparation, hiring skilled consultants, and managing technical tasks can be prohibitively high, as highlighted by another respondent: “For example, in Germany, you might have one person working on this, …, there are people who label the data; they cost money, too. Plus, you’d need a senior Python consultant to review the extraction code.” (InsEnt1). Collectively, these insights reveal that financial limitations are a critical barrier for SMEs, hindering their capacity to invest in the human and technical resources required forsuccessful AI adoption.

​

High Costs of External Expertise

The high cost of external expertise further compounds this issue; as one participant stated, “When we seek outside help, we encounter high costs (...) For small businesses like ours, budgets are tight, capital is limited, and resources are scarce” (WelEnt1). On top of not having sufficient financial resources, the high cost of external expertise makes it even more difficult for SMEs to adopt AI.

 

Overwhelming AI-Implementation

Another barrier encountered during the AI adoption process was the attempt to implement too many AI use cases simultaneously. This approach, while initially appealing as a strategy to enhance efficiency across various business functions, quickly became overwhelming. As one interviewee explained, "One of the biggest barriers we faced was trying to implement too many AI use cases at once. (…) It seemed like the perfect way to enhance our overall efficiency and stay ahead of the curve. But what ended up happening was that we were completely overwhelmed" (CnsltEnt1).

 

Knowledge Gap About AI’s Capabilities

A significant barrier to AI adoption in SMEs is the lack of knowledge about what AI can do and how it can be implemented effectively. Many SMEs struggle to understand the potential applications of AI within their operations, leading to hesitancy and missed opportunities. As one interviewee expressed, “I think we might be able to use AI in our company, but first, we need to fully understand what AI can do (...) I didn't know how AI could be used in production, nor did I know how to implement it” (ConsEnt1). This highlights a widespread gap in knowledge about AI’s capabilities, making it challenging for SMEs to envision how AI could enhance their existing processes. Additionally, SMEs often find themselves unsure of how to structure their data or prepare their operations for AI integration. One participant noted, “We don't know how to introduce this to AI or how to categorize it (...) we don’t know how to use it with AI, nor what we need to do as preliminary work” (LegEnt1).

 

The knowledge gap surrounding AI’s capabilities is further highlighted by contrasting perspectives among SME owners about AI’s effectiveness in specific business functions, such as marketing. One interviewee expressed skepticism about AI’s value in marketing, stating, “I used ChatGPT and Google’s AI; for example, I asked, ‘How can we increase our market share?’ but it gave me the same answers I already knew (...) I don't think AI could provide us with anything significantly different in sales and marketing” (ConsEnt1). In contrast, another interviewee found AI to be highly beneficial in enhancing marketing efforts, particularly in areas like search engine optimization (SEO). They noted, “I would say (we can use it more in) marketing (...) AI has taught me a lot about SEO, optimization, and better marketing” (LogEnt1). This divergence in viewpoints often stems from varying levels of familiarity and experience with AI tools. Those who have taken the time to explore and experiment with AI applications tend to see its value, while others may dismiss AI prematurely due to limited exposure or past experiences that failed to demonstrate meaningful impact.

 

Complexity of AI Tools and Interfaces

The complexity of AI tools and interfaces further compounds this issue. One respondent shared that the interfaces are overwhelming and difficult to navigate, stating, “The user interfaces of these applications are very complicated for me. There are many icons, each doing different things, and there’s no clear description of what each does. Since we’ve come into these technologies later (…) we don’t fully understand what to ask from it.” (WelEnt1). This reflectsa broader challenge often associated with age, where individuals who have not grown up with advanced technologies may find it harder to adapt to and fully utilize AI tools, leading to a steeper learning curve and increased resistance.

 

Uncertainty about AI Integration Steps

Moreover, even when SMEs recognize potential areas for AI application, they are often uncertain about the specific steps required for implementation. For instance, one interviewee discussed the need to improve production efficiency but admitted, “We certainly have areas, but I don’t know how suitable our infrastructure is for it. For example, to increase production efficiency and speed, the speed of orders, both in buying and selling, is very important…But honestly, I don’t know exactly how we would use it. Does it need to be integrated into our machines, or do we need to input data from there to analyze?” (ManuEnt3). This uncertainty often stems from a lack of technical expertise, inadequate guidance, and insufficient knowledge about how AI can be customized to fit existing workflows. As a result, SMEs struggle to bridge the gap between identifying AI opportunities and executing them effectively.

 

Data Quality

Data quality emerged as a significant barrier in the AI adoption process, particularly in sectors like finance, where the volume and variety of data are substantial. One interviewee captured the challenge succinctly: "One of the biggest barriers is data quality (…) AI is only as good as the data you feed it. Financial markets are tricky because you’re dealing with vast amounts of data from different sources—some reliable, some not so much. Cleaning and processing that data before feeding it into the AI model is a huge task" (FinEnt1). This challenge underscores the critical importance of data management in AI implementation.

 

Lack of Employee Knowledge

Another significant barrier to AI adoption in SMEs is the lack of knowledge and expertise among employees, which limits the ability to fully leverage AI technologies. This issue extends beyond financial constraints; as one interviewee noted, “It’s not just about the money; the human resource is critical. If you can’t reach key people, no matter how much money you have, how will you develop these things?” (InsEnt1). This concern often stems from the challenge of attracting and retaining skilled personnel who understand AI’s complexities, especially in smaller businesses where resources for hiring specialized talent are limited. Even when technical staff possess some AI-related skills, there are often gaps in knowledge among operational employees, which hampers effective implementation. As stated, “Our technical staff are providing AI-related services, but there are deficiencies among the operational staff” (InsEnt1). This knowledge gap often arises from insufficient training and a lack of continuous learning opportunities, particularly for non-technical staff who may not be directly involved in AI initiatives.

 

Lack of Practical AI Implementation Experience

A lack of practical experience in AI implementation among employees can be a significant barrier to successful AI adoption. One company faced this challenge when one of their developers, despite having a strong theoretical understanding of AI, struggled with the hands-on aspects: "One of our developers was really good with AI in theory, but when it came to hands-on implementation, there were gaps in their knowledge" (CnsltEnt1). This highlights a common issue in AI integration, where theoretical knowledge alone isn’t enough to tackle the complexities of real-world application.

 

Lack of Employee Openness

The lack of understanding and openness to technology among employees can create furtherchallenges. One respondent shared that they often proceed independently rather than involving others, explaining, “People’s approach to technology, especially in this sector and considering their education level, would make it [transformation process] difficult, so I always do it alone” (LogEnt1). This reluctance to involve employees reflects a broader issue where insufficient knowledge and resistance to new technologies hinder collaborative efforts in AI integration.

 

Technological Awareness

Additionally, there is a noticeable gap in technological awareness even among younger employees. One interviewee noted a surprising trend: “I realized that [new graduates] knew even less about new technologies than I did (...) For example, someone 24-25 years old, who should be in the middle of this right now, knows less about AI than I do” (ConsEnt2). This gap in knowledge among younger, presumably more tech-savvy employees, indicates a broader problem within the workforce, where even those expected to be familiar with AI fall short of the necessary understanding. This lack of engagement often stems from a fear that AI could threaten job security, leading employees to avoid fully engaging with these technologies. Many prefer to stay within their comfort zones, resisting new tools that could potentially change their roles or make them redundant. As a result, even younger employees, who are often assumed to be more adaptable to technological change, may distance themselves from AI.

 

Employee Resistance to Change

Employee resistance to change is a significant barrier to AI adoption in SMEs, driven by a reluctance to leave familiar practices and embrace new technologies. As one interviewee noted, “In digitalization, people don’t want to leave their comfort zones” (TechEnt2). This resistance is often evident when new technological initiatives are introduced, with employees viewing AI as an added burden rather than a valuable enhancement. One participant observed, “Our employees are quite resistant to technological innovations. Top management, aside from myself, even though they are older than the employees, constantly follow technological developments and try to keep up, but the first reaction we get from employees is always resistance. Because of this, we don’t make announcements about technological steps unless we are sure, to avoid seeing resistance and disrupting our daily work” (LegEnt1).

 

Some interviewees pointed out that this resistance is more common among younger employees. One respondent shared:“The biggest challenge we face when adopting AI is the attitude of employees (...) we see this especially among younger employees” (LegEnt2). Another interviewee explained that this resistance might stem from the unique experiences of the younger generation, particularly those who missed out on traditional university and social experiences during the COVID-19 pandemic. They commented, “It’s normal for the young team to have this fear because I see that they missed out on university and social life during the pandemic. They couldn’t experience the real university life, and now they feel their work is not valuable. At the same time, they believe they can get rich quickly without working too hard (…) developing technology scares them.” (TechEnt1).

 

The challenge of overcoming ingrained work habits was further highlighted by another interviewee who shared, “I’m more curious about how it will be received by the people using it after we’ve created the product (...) Sometimes our work habits keep us from changing, even

when it’s something important.” (InsEnt1). This illustrates how established routines and comfort with existing processes can obstruct the adoption of new technologies, even when these innovations are designed to improve efficiency or solve significant problems. An extreme example of this resistance was described by a respondent who faced considerablepushback from employees despite introducing a tool that was intended to make their jobs easier. “I told the salespeople, ‘Use this; look, it’s great (...) But they didn’t use it for a long time. I didn’t understand why, so I offered rewards, made rules, threatened punishments (...) Finally, I snapped and made phone calls, saying, “If you don’t use it, I’ll fire you from the company”(...) They’re good employees with strong portfolios; I’m not talking about lazy workers. I then asked them why, and they said that transitioning from one system to another is difficult. This is something we often overlook in project transitions; once a workflow is established, how does one change that workflow? After doing everything, I resort to force; I create crises and that’s how I make it happen. I haven’t discovered any other method.” (InsEnt1). This underscores the deep-rooted resistance employees can have towards change, even when the new system is

demonstrably beneficial, highlighting the often-overlooked human challenges in digital transformation.

 

Fear of Job Loss

Resistance to AI adoption in SMEs is often fueled by employees' fear of job loss, as many employeees perceive AI as a direct threat to their employment. Although this concern is rarely expressed openly, it significantly influences attitudes toward AI. As one interviewee noted, “In the industry, people hesitate to express this directly as ‘AI will take my job,’ so they frame it as ‘AI can’t do my job anyway.’ But it certainly can. For example, everything that an intern does can already be done by AI” (LegEnt2). Another interviewee acknowledged the validity of these concerns, stating, “The reason is purely that they are concerned about AI taking their place. I don't look at it entirely wrong because AI is coming in full force and may take jobs, and that would mean AI could also manage the company.” (TechEnt1). Some managers believe that AI will not entirely replace jobs but rather transform the nature of work. As one respondent explained, “I don’t see AI replacing jobs in our company in the near future. AI will likely change how we work by automating repetitive and time-consuming tasks, but there will always be a need for human oversight, especially in decision-making and customer relations” (ManuEnt4).

 

Lack of Trust in AI Accuracy and Reliability

Trust issues regarding AI’s outputs also represent a significant barrier to its adoption in SMEs, as employees and managers often question the accuracy and reliability of the information provided by AI systems. One interviewee expressed skepticism about the credibility of AI-generated information, stating, “The main concern is whether the information is accurate and the source. At the end of the day, it's a computer providing the information (...) [AI] researching, finding, and reading them often takes longer than if you just did it yourself” (ManuEnt1).

​

This lack of trust is further compounded by the need for constant oversight of AI’s work. As another participant noted, “The main problem we face while trying to use AI models is that they do not actually reduce our workload (...) we still have to double-check the AI’s work” (LegEnt2). Similarly, another respondent voiced concerns about AI’s ability to improve work processes, stating, “I don't think it can do our job better than us or help us work faster. I still have to check what it does, which wastes my time” (ConsEnt1). These comments reflect a broader hesitation to rely on AI outputs without manual review, reinforcing the perception that AI lacks the reliability needed to be trusted fully.

​

Concerns also extend to the potential consequences of errors or oversights by AI systems. One respondent highlighted the risks associated with AI-enabled automation, especially in contexts where precision is crucial: “The slightest miscalculation here or something AI overlooks could lead to work accidents (...) Trying to speed it up could cause us a lot of trouble” (ConsEnt2).

​

Moreover, even when AI provides outputs that could be beneficial, trust issues persist, particularly when the AI’s suggestions conflict with the user’s own experience and knowledge. As one interviewee explained, “If AI says something contrary to what I know, based on years of experience, I trust the correctness of what I know more and continue to do the work as I know it” (ManuEnt3). While AI can serve as a valuable research tool, users are reluctant to fully trust its recommendations, particularly when those recommendations involve technical matters or areas where the user lacks expertise.

 

Client Skepticism

Client skepticism, especially among older clients, poses a significant barrier to AI adoption. As one interviewee explained, “But, on the flip side, clients also need to trust the AI, and some of our older clients are still a bit skeptical” (FinEnt1). This resistance stems from a lack of familiarity with AI technologies and concerns about their reliability. Clients who are used to traditional methods may be hesitant to trust AI-enabled systems, fearing that automated processes could make mistakes or overlook important details. This skepticism can slow down the adoption of AI tools in client-facing areas, as businesses must first build confidence in the technology’s accuracy and benefits.

 

Fear of Failure

Fear of failure is a significant barrier to AI adoption in SMEs, stemming from the rapid pace of technological change and the high stakes associated with investing in unproven technologies. One interviewee highlighted this concern, stating, “This AI thing is changing so quickly that you might be working on something in R&D, and suddenly, something new comes up, right when you’re ready to use something. A more suitable tool you didn’t know about might just pop up. It’s so fast; its speed of change is incredible. You might realize one day that you made the wrong decision with the technology you were using and that you invested a lot of money in something wrong. That risk exists… So, I’m moving slower and taking fewer risks. If one day my technical people come to me and say, “Sorry, man, we tried, but it didn’t work, and your money went to waste,” that would make me really unhappy, to put it mildly” (InsEnt1). The unpredictability of AI advancements can make it difficult for companies to keep up, and the fear of investing in the wrong tools or technologies creates a significant reluctance to fully commit to AI initiatives.

 

Negative Experiences with Failed AI-projects

Past experiences with failed AI projects further exacerbate this fear. One participant shared their disappointment after working with an AI team that failed to deliver on promised results: “We were previously working with a two-person AI team, but their work was very inefficient (...) They promised us for about six months that they would create the AI applications I just mentioned, but it never materialized. They came up with many excuses, like high costs, data requirements, etc. That’s why I’m hesitant to take this risk again.” (ManuEnt2). This experience not only led to financial losses but also eroded trust in AI investments, leaving the company hesitant to take similar risks in the future. The inability to understand AI processes and outcomes made the company vulnerable to being misled, reinforcing fears that AI adoption could result in wasted resources and failed projects.

 

Organizational Overload

Another barrier to AI adoption in SMEs is the organizational overload and complexity involved in integrating new technologies into existing work processes. One interviewee highlighted the challenge, stating, “My team is too swamped with projects right now (...) It’s not enough just to buy the technology and set it up (...) The integration into sales processes, explanations, planning on top of it, and feeding it back into other processes are all technicalities with significant organizational costs beyond just obtaining the technology” (InsEnt1). This issue is primarily driven by a lack of time and personnel, as SMEs often struggle to balance AI integration with their ongoing operations. Since their primary focus must remain on traditional activities that generate revenue, they lack the resources to dedicate to the complex process of AI adoption. This results in AI projects being sidelined or delayed, as immediate business needs take precedence over long-term technological investments.

 

Inter-departmental Conflicts

Inter-departmental conflicts are a barrier to AI adoption in SMEs, often exacerbated by differing levels of acceptance and resistance to change within the organization. These conflicts arise when certain departments embrace new technologies while others resist, leading to a lack of alignment across the organization. As one interviewee noted, “One department is happy, while another is not. Inter-departmental conflicts were a major issue, compounded by resistance to change, which isn’t necessarily age-related” (TechEnt2). This friction is often driven by varying priorities, work habits, and comfort levels with technology among different departments. Departments that see immediate benefits from AI are more likely to adopt it, while others, fearing disruption or lacking confidence in the technology, resist changes that alter established workflows.

 

Lack of Tailored AI Solutions for SMEs

Another barrier to AI adoption in SMEs is the perception that existing AI solutions are not tailored to the specific needs and limitations of small businesses. One interviewee highlighted this concern, stating, “No matter how well our business goes, we will never have a budget as big as Microsoft, Apple, or IBM (...) the most important factor for me would be a solution that suits small businesses' budgets and is easy to use” (ConsEnt1). This reflects a common sentiment among SMEs: while AI has potential, the solutions available often feel designed for larger enterprises with extensive resources, making them difficult for smaller businesses to adopt without significant financial and operational strain. As another interviewee noted, “While there are more AI tools entering the market, many of them don’t address the specific challenges SMEs face” (ManuEnt4). The lack of customized, accessible, and affordable AI solutions creates a sense that AI is better suited for larger companies, discouraging SMEs from investing in these technologies.

 

Challenges in Customizing AI Systems

The difficulty of customizing AI systems to meet specific organizational or industry needs is another challenge. Even when companies have digital data, tailoring AI to understand and process this information accurately can be a daunting task. One interviewee highlighted this challenge, stating, “We think there will be challenges in making this digital data recognizable (...) the AI needs to be tailored to the Turkish legal system. For instance, it needs to understand what a decision is, what an appeal is, who the parties are, what annulment means, and what re-litigation after annulment entails” (LegEnt1).

 

Top Management Resistance

While top management recognizes the importance of their role in guiding digital transformation, resistance at the leadership level poses a significant barrier to AI adoption in SMEs. One interviewee noted, “We can’t expect much from employees without clear guidance from management” (LegEnt1) highlighting the critical role of leadership in driving technological change. However, some top managers themselves resist this transition. As stated,“It’s not just employees at the lower levels; top management also resists this issue (...) top management lacks knowledge in this field, which complicates implementing strategies and regular planning” (LegEnt2). This resistance and lack of understanding can create a top-down challenge that hinders digitalization efforts, even when attempts are made to introduce new technologies through seminars and market penetration strategies. 

 

An example of this hesitation is seen in one manager’s account: “To be honest, I didn't see the need for something like this from the start, maybe because of my old-fashioned thinking or my age, but AI didn't seem to teach me anything new. However, due to the insistence of two young employees we recently hired, we are now slowly focusing on this issue.” (ConsEnt1). This reflects how reluctance at the top can slow down the adoption process, even when younger employees push for innovation.

 

Another reason for top management's resistance lies in the potential impact on BMs, especially in service-oriented firms where revenue is linked to billable hours. One respondent explained, “Reducing a task that would normally take us three hours to 15 seconds isn’t always something management wants (...) if we finish in half an hour and bill for half an hour, the customer would love it, but it’s a big risk for us” (LegEnt2). This reluctance to shorten billable time illustrates how financial incentives can clash with the operational efficiencies AI offers, creating a conflict between maintaining traditional revenue streams and embracing more efficient technologies.

 

Also, the resistance within management is not always uniform; some managers are open to change, while others resist, leading to a divided stance on AI adoption. As one interviewee described, “[Is there a top management support in the transformation process?]. It’s divided. The management level is broad, so some parts support it, while others don’t, and this conflict is reflected at every level” (LegEnt2). Even when there is no outright resistance, a lack of knowledge can prevent top management from fully embracing AI. “I don’t think there will be much resistance from management. However, there is a weakness in management: they can’t fully grasp AI's capabilities. We need to present them with simple applications in their fields to spark their interest and increase their enthusiasm because solving problems using traditional methods seems easier for them.” (ManuEnt2). This knowledge gap often results in hesitance to adopt AI unless its benefits are clearly demonstrated in simple, applicable ways.

 

Lack of Technological Readiness and Infrastructure

Many SMEs are still in the early stages of digital transformation and are not yet equipped to integrate AI effectively into their operations. So, the lack of technological readiness and adequate infrastructure to support AI implementation is another barrier. Even companies that provide technical and engineering services struggle with inadequate infrastructure for AI. A respondent noted, “Even though we offer technical and engineering services, our infrastructure is not equipped for AI” (TechEnt2). This illustrates that without the right technological foundation, even businesses with a strong technical background face significant hurdles in adopting AI.

 

Privacy and Confidentiality

Privacy and confidentiality are significant barriers to AI adoption in SMEs, especially for those handling sensitive data. SMEs often face challenges in ensuring that data is adequately protected, anonymized, and secure, which can complicate AI integration. As one interviewee highlighted, “One of the biggest challenges in AI integration is the confidentiality and security of healthcare data (...) protecting and anonymizing personal data correctly can be challenging” (HealthEnt1). This concern is particularly acute for industries like healthcare, law services where the stakes for data breaches are high, and stringent privacy standards must be maintained.While larger organizations with international collaborations have developed more robust strategies to manage these concerns, smaller SMEs struggle significantly. For instance, LegEnt2, uses its own developed large language model (LLM) primarily to handle confidential data internally, while leveraging specialized external AI models for non-confidential tasks. In contrast, smaller SMEs often face bigger challenges in managing data privacy. As one respondent noted, “There are personal data concerns; we can't transfer everything as it is—it needs to be filtered. Data anonymization is crucial for us; it’s a non-negotiable requirement.” (LegEnt1). These privacy constraints make it difficult for smaller companies to fully utilize AI, as they must constantly navigate the complexities of data filtering and anonymization.

 

Risk Aversion

Minimizing risk is a critical concern for SMEs when adopting AI, as uncertainties about the technology often lead to cautious approaches. Many SME leaders are wary of venturing into AI due to the perceived risks involved, especially when lacking familiarity with the technology. One interviewee expressed this hesitation, stating, “For me, the risk is significant. It’s more about ensuring the work is tidy rather than how much money I’ll earn. And what does risk mean for me? For example, placing a 3-ton air conditioning unit on a roof isn’t a risk because I know that process from previous work. But as I said, the risk in AI is all a risk for me. Since it’s an area I don’t know, it wouldn’t be easy for us to spend a lot of money on it” (ConsEnt2). This reluctance is driven by the desire to maintain control and avoid making costly mistakes in an unfamiliar domain.

​

​

Enablers

​

Top Management Support

Leadership emerged as a vital factor enabling SMEs to transition toward AI-driven business models. Managers’ proactive involvement significantly influences organizational change. One respondent emphasized, *“The main element of digital transformation is that a business manager sees, understands, and measures what they are doing; the more you understand, the better you can manage it” (InsEnt1).* Another affirmed the impact of visionary leadership, stating, *“Honestly, without wanting to boast, one of the most important assets of our company in making this transition is myself and my vision. I believe that a company can progress as far as its leader's vision. I invest heavily in myself and my colleagues” (TechComp1).* Effective leadership further motivates employees, as another interviewee noted, *“It’s also about how people are motivated and encouraged. If you’re showing effective leadership, you can persuade your colleagues” (TechEnt2).* Encouraging a culture where failure is acceptable fosters innovation: *“We’ve made it clear that it’s okay to fail as long as we’re learning from those failures. This has given employees the freedom to explore AI tools without feeling like they’re risking their job if something goes wrong” (CnsltEnt1).*

 

Employee Selection

Selecting the right employees with appropriate technical and industry-specific knowledge was highlighted repeatedly. Interviewees stressed combining domain expertise with technical skill: *“It has to be someone who combines law and technology so that we can communicate properly (...) It has been hard to communicate our needs correctly to the service providers” (LegEnt1).* Another explained the need for domain-specific guidance, *“The problem is that the friend knows programming but doesn’t know what to program (...) Programming alone isn’t enough; someone who knows the job needs to guide them” (ManuEnt3).* The presence of skilled individuals within teams significantly facilitates transformation: *“My CTO—directly my CTO. The CTO working in my company, he previously worked with a team on such a vision in another company, a German company” (InsEnt1).*

 

Resource Allocation

Strategic financial investment greatly impacts AI integration. Companies see clear benefits in AI-related resource management: *“Yes, I pay a certain cost, but if I had to hire more people to ensure this coordination, the coordination would be much more expensive” (TechEnt1).* Allocating clear budgets also sends essential signals within the organization: *“Management also allocated specific budgets for AI adoption and training, which showed the rest of the team that this wasn’t just a passing fad. They were serious about it, and that seriousness trickled down to everyone else” (CnsltEnt1).*

 

Training and Guidance

Continuous training and educational support are pivotal in AI adoption. Companies recognize the general lack of readiness: *“Honestly, who is ready for AI? Who’s ready for digital transformation? There’s no such thing; people were completely caught off guard in every way, and everyone needs it \[training]” (InsEnt1).* Practical guidance tailored to specific industry needs is essential: *“Absolutely. While we know our medical field, we don't know how it can be adapted to technology. We need training, and then we need someone who can support us full-time after the training” (HealthEnt1).* Visible outcomes from training further motivate continuous investment: *“Absolutely, and we’re already investing in that \[training]… The better equipped our team is, the smoother the transition to new technologies. And even with such a small investment, I can see that it does actually very positively affect our digitalization” (ConsEnt2).* Some companies prefer internal training due to privacy and customization concerns: *“Yes, we desperately need it. However, the company’s current policy is to provide training internally (...) training is ongoing” (LegEnt2).*

 

Collaboration and Peer-to-Peer Learning

Collaborative learning among SMEs and internal teams significantly supports AI initiatives: *“Partnering with other small businesses in similar industries. We pool our resources to bring in external experts for shared workshops. It’s been a win-win because we get high-quality training at a fraction of the cost” (CnsltEnt1).* Internally driven peer-to-peer learning also proves effective: *“We’ve also started encouraging peer-to-peer learning, where one employee who’s mastered a tool teaches the others. It’s been quite effective” (FinEnt1).*

 

AI-Powered Tools

Using AI to manage employee concerns and demonstrate direct benefits facilitates smoother adoption: *“We implemented an AI-powered wellness app that helps employees manage stress through mindfulness exercises and mood tracking. It’s a bit ironic—using AI to reduce stress about AI—but it’s been well-received” (CnsltEnt1).*

 

Gamification and Structured Engagement

Gamification significantly reduces resistance to AI by engaging employees positively: *“Gamification really helped, to overcome fears that some of our employees were facing. We introduced internal competitions where employees could showcase how they used AI to improve their workflows. We gave out small rewards, like gift cards, to the best examples” (CnsltEnt1).* Structured sessions such as regular meetings further support integration: *“We also set aside time every week for ‘AI Fridays,’ where we’d discuss the latest AI tools and how we could use them” (CnsltEnt1).*

 

Establishing AI-Leadership

Dedicated leadership roles enhance clarity and coordination: *“We appointed a ‘chief AI officer’ of sorts—someone who’s responsible for overseeing all our AI tools and making sure they’re working in harmony” (CnsltEnt1).*

 

Client Openness

Client receptivity to AI enables smoother adoption and experimentation: *“Some of our clients are quite tech-savvy and are actually excited about the integration of AI. They’ve been willing to be more experimental, which gives us room to try new things without too much pressure” (FinEnt1).*

 

Together, these enablers, reflected in the real-world insights of SME stakeholders, create a comprehensive foundation to support SMEs' transition into AI-enabled business models, turning potential barriers into actionable steps for innovation.

​

​

Framework

​

Many SMEs lack a clear strategy for transitioning to AI-enabled BMs, often feeling lost and uncertain in their digital transformation journey. Despite recognizing the importance of strategic planning, many efforts fall short due to the absence of structured guidance. One interviewee highlighted this challenge, stating, “Unfortunately, no [we don’t have a digitalization strategy], and that’s why all our attempts end up failing.” (ConsEnt1). Even companies with a strategy in place find it difficult to adhere strictly to their plans. Execution often diverges from the initial roadmap due to unexpected interventions and practical constraints. As one participant noted, “Of course, there is a strategy (...) it’s essential. But how well you follow that plan varies.” (TechEnt2). In some cases, partial adherence to the plan is seen as a success. For instance, one interviewee stated, “But when you apply it only 80%, you progress, but if you drop down to 40%, you end up out in the cold” (TechEnt2). This highlights the challenges in fully implementing strategies and the reality that reaching even 80% adherence is often considered a positive outcome.

 

Step 1: Assessing Readiness and Choosing the Right Implementation

The first step is to assess organizational, technical, and environmental readiness. As one interviewee noted, “Once we decide on our goal, we assess whether our technical infrastructure can support it. If you jump in without this check, you might find out too late that you're not equipped to handle it, which would end up costing you a lot more. So, having a clear plan and ensuring we have the technical readiness is a huge enabler for us” (ConsEnt3), SMEs must evaluate their internal capabilities, technological infrastructure, and market conditions to determine if they are prepared for AI adoption.

• Learning AI’s Role in Your Industry: It’s critical to understand AI's specific applications for your industry. SMEs can access workshops, webinars, consultations with AI experts, or leverage free online resources to learn about AI’s role in enhancing operations. This ensures that the adoption of AI aligns with actual business needs and customer expectations.

• Leadership Commitment: Ensure that top management is fully engaged, not just during the planning phase but throughout the AI journey. Leadership must foster a culture of experimentation and continuous improvement, which helps overcome resistance and align AI with business goals. Regular progress reviews involving leadership are essential to maintain momentum.

 

Step 2: Aligning AI with Business Goals

AI should be aligned with the company’s overall business strategy and existing BM to ensure that it enhances value propositions rather than conflicting with them. For SMEs, aligning AI with their business goals means thoroughly evaluating whether the AI solution they are considering will truly add value to their current operations, such as improving customer service, optimizing supply chains, or enhancing decision-making. This assessment ensures that AI investments are not just technological upgrades, but strategic tools that can drive growth and

innovation in a focused and efficient way. 

A crucial part of this alignment process is questioning whether the chosen AI solutions match customer expectations and the business's long-term vision. As one interviewee described, "We don’t start with, ‘Hey, let’s implement AI in our business model today.’ First, we define our goal—why we want to do it—and outline a general plan on how to achieve it, and then we look for the best fitting technology for our goal. We then consider whether any part of this implementation might clash with what we currently do. For instance, if we invest heavily in creating a chatbot for answering questions about construction, we need to ask ourselves if that’s really what the customer wants. Personally, I often avoid chatbots because I don’t trust them. So, while we're focused on building customer trust, some investments might not align with our overall goal" (ConsEnt3).

• Customer-Centric Approach: Evaluate AI’s role in improving customer service or delivering value without compromising trust. For example, when considering AI-enabled tools like chatbots, ensure customers are comfortable with AI interactions. Providing transparency and choice (e.g., human support options alongside AI tools) can help overcome customer skepticism.

• Alternative Revenue Models: Particularly for service-oriented SMEs where AI might reduce billable hours (e.g., through automation), explore alternative revenue models that focus on value-based pricing instead of time-based billing. AI adoption can enhance productivity without undermining revenue, allowing companies to charge based on outcomes rather than effort.

 

By carefully considering whether AI fits with the company’s current goals and operations, SMEs can avoid costly missteps and ensure that AI becomes a strategic enabler that enhances their overall BM, rather than a disconnected or ineffective solution.

 

Step 3: Simplified Data Management

Effective data management is essential for AI success. SMEs should focus on manageable data management steps that lay a strong foundation for AI integration without overwhelming their resources.

• Digitization and Data Privacy: Begin by digitizing core business data (e.g., customer records, operational data). As one interviewee shared, “We dedicated significant time and effort to manually convert all our non-digital data into digital formats (...) we recognize it as a crucial foundation for integrating digital technologies like AI into our operations” (LegEnt1), having your data digital is an important step. If your business handles sensitive data, implement privacy measures such as data anonymization to ensure security. This step is critical for industries like healthcare and finance where data sensitivity is a concern. As another participant noted, “For an institution like ours that works with sensitive data, protecting and anonymizing personal data correctly can be challenging” (HealthEnt1).

• Accessible Data Tools: For SMEs without the resources for advanced data infrastructure, focus on affordable or open-source data management tools that simplify data collection, cleaning, and structuring processes. These tools allow SMEs to prepare their data for AI without major upfront investments.

 

By focusing on these practical, foundational aspects of data management, SMEs can prepare their data for AI integration without being overwhelmed by complex governance systems. This manageable approach ensures that SMEs are ready for AI and that the transition to AI-enabled processes can be carried out efficiently and securely.

 

Step 4: Phased AI Integration with a Structured Approach

A structured, phased approach to AI integration helps SMEs implement AI gradually, reducing risks and ensuring continuous progress.

• In-house vs. Outsourcing: SMEs should decide whether to develop AI solutions in-house or outsource development. Outsourcing can provide access to specialized AI skills without the cost of hiring an internal team, as one noted;“I think it’s better to do it outside (...) I can’t manage software developers on my own” (ConsEnt2). However, in-house development allows more customization and control over AI projects, as one interviewee noted, “We did it with Python (...) the person reviewing the code gives feedback on the future of the project” (InsEnt1).

• Pilot Projects: Begin with small-scale AI projects to test the effectiveness of AI in specific areas, such as automating routine tasks or enhancing customer service. These pilots will provide insights that can refine the strategy before larger-scale AI adoption.

• Avoiding AI Overload: Implement AI systems one at a time, ensuring each solution works well before introducing new ones. This prevents the "AI chaos" that occurs when too many systems create conflicting recommendations, paralyzing decision-making.

 

Step 5: Building an AI Team, Training, and Managing Resistance

Once readiness has been assessed and the phased plan established, building a capable AI team and managing cultural resistance are essential.

• Continuous Training Programs: SMEs should implement continuous, role-specific training rather than one-off sessions. Free online training resources or partnerships with educational platforms can provide affordable, accessible AI education. Training should include clear demonstrations of AI’s benefits to mitigate fears of disruption.

• Clear and Warm Communication: While role-specific training remains essential employers should also use warmer communication strategies to alleviate fears and encourage openness. As one interviewee mentioned, “We try to communicate that AI isn’t here to take away jobs but to help us grow (...) if we can automate repetitive tasks, employees can focus on higher-value activities” (ManuEnt4). This clear and positive messaging helps employees see AI as a tool for support, not a threat.

• Gamification and Rewards: Use gamification strategies to engage employees, turning AI adoption into a positive experience. Internal competitions, rewards for innovation, and recognition programs can motivate employees to embrace AI tools. These strategies create a cultural shift, reducing fear and resistance.

• AI Tools and Emotional Support: In addition to training, consider providing emotional support through AI-powered tools like apps, helping employees manage anxiety about AI’s potential disruptions. This shows a commitment to employee well-being and fosters a smoother transition.

 

Step 6: Gradual Integration of AI into Business Processes

Once the foundational steps are in place, the next phase is the gradual integration of AI into various business processes. For SMEs, this approach allows companies to start with small, manageable AI implementations, focusing on areas that offer immediate benefits and expanding as they gain experience, As one interviewee noted, “Our strategy focuses on gradually integrating digital tools to improve operations and streamline workflows” (ManuEnt4). By starting small, SMEs can test AI’s impact, make refinements along the way, and reduce the risk of overwhelming their existing operations.

• Targeting High-Impact Areas: Start with AI applications that deliver quick, measurable results, such as automating repetitive tasks, improving supply chain management, or enhancing customer support.

• Cross-Departmental Collaboration: Ensure alignment between departments by fostering collaboration and ensuring everyone operates at a similar level of AI readiness. Avoid creating silos where some departments adopt AI rapidly while others lag behind, which can create friction and slow progress.

 

Step 7: Exploring New AI Applications

Once AI is successfully integrated into core business processes, SMEs can explore new opportunities for AI in other areas of their operations.

• Iterative Expansion: Use insights from the initial AI projects to identify additional areas where AI can deliver value. Ensure the company’s internal processes are stable and AI expertise is well-established before venturing into more complex or unfamiliar applications.

• Adaptability to AI Advancements: Monitor AI developments regularly, staying updated on new tools and trends. This step helps SMEs mitigate the risk of AI obsolescence and ensures continuous improvement, making AI a long-term driver of growth rather than a one- off investment.

 

This structured approach helps SMEs maximize the impact of AI on their growth by first solidifying their internal capabilities before venturing into new opportunities. With a strong foundation, AI becomes a key driver of both operational efficiency and future market exploration.

​