Startups & Entrepreneurship
August 26, 2025
10 min read
AI and related technologies are rapidly developing. These advancements inevitably attract the attention of businesses interested in innovative solutions to improve the efficiency of their processes. But how is it possible to make this implementation smooth and trouble-free, especially when a company has no opportunity to attract investor funding? This became the topic of a new episode of the Innovantage podcast, hosted by Sigli’s CBDO, Max Golikov. He invited Denis Leysen to his studio to discuss it.
Denis is an AI adoption strategist and a former enterprise consultant with deep roots in family entrepreneurship.
But apart from that, he is also a co-founder of noah., a startup that helps businesses identify which AI solutions are actually relevant and effective for them.
The project was founded in late 2024, and by January – February 2025, the first product prototypes were underway.
Traditional consultancy often involves time-consuming interviews and lengthy reports. Meanwhile, noah. offers a fully automated, personalized, and contextualized approach to AI opportunity discovery. The goal is to save companies time and effort in exploring the AI landscape while delivering actionable, tailored insights.
Denis believes bootstrapping isn’t a rigid philosophy but a practical approach, especially in today’s startup environment.
While there is always the option to raise funding, Denis emphasized that most investors, VCs, and incubators advise founders to delay fundraising as long as possible. Bootstrapping can give founders more flexibility and control.
However, financial obligations and personal constraints can make external funding necessary.
The good news is that starting a company has never been more accessible than today. Software development costs have dropped significantly. Meanwhile, software is increasingly becoming a commodity.
With AI-powered tools, founders can generate most of the code on their own, hire freelancers to fine-tune it, ingest client data, and launch with minimal upfront investment. As a result, bootstrapping a tech startup is increasingly feasible, even for solo founders.
As Denis highlighted, it is a great time to start something, test it, and see where it goes, with very little capital required.
As a result, the barrier to entry is lower, but the risk of being copied is higher.
This environment makes startup defensibility more important than ever before. One of the biggest challenges for early-stage ventures is ensuring their solution can’t be easily replicated, especially by large players like OpenAI or Perplexity that can quickly release new features.
noah. is still in the early stages of establishing a strong product-market fit. After running some pilot projects, the team is now working with a few paying clients through structured use tracks.
The product relies on AI to conduct internal interviews within a company and collect insights directly from employees via calls or screen prompts. This process helps map out the organization’s current workflows, tools, and challenges. In parallel, noah. continuously scans the AI landscape for proven use cases and establishes partnerships with consultancies and solution providers to understand what technologies are working in the market. By combining such insights, noah. offers clients a personalized and up-to-date AI opportunity report.
But what comes next after identifying the opportunity? The team is now exploring ways to connect clients directly with the right implementation partners, who can build solutions for them.
Moreover, noah. sees potential beyond AI. The same underlying technology could be applied to other domains, such as supply chain optimization or workflow improvements. This will make the platform adaptable across industries.
noah. primarily targets larger organizations. In such organizations, the complexity and scale of operations make AI opportunity discovery significantly more challenging. In multiple departments, it becomes difficult to track what employees are struggling with or where AI could make a difference. Given this, the value of noah.’s offer can’t be underestimated.
As Denis noted, AI adoption in large organizations is progressing, but it is happening unevenly. Most companies have moved past the initial skepticism, where tools like ChatGPT were blocked due to perceived risks. This early caution was the first stage of enterprise engagement with AI.
The second phase involved experimentation. Enterprises were testing major tools like Microsoft Copilot and piloting small use cases.
However, many companies are now pausing or even pulling back. For instance, some of them abandon co-pilots due to high costs and underwhelming results. This has led to a third phase that can be described as disillusionment.
Currently, AI adoption in enterprises tends to be fragmented. While most have some experience, organization-wide AI solutions are still rare. Success is typically found in very niche use cases where the business problem is clear and the solution is highly targeted.
Denis explained that the current hype around AI is justified. Unlike past technological waves, AI has quickly become embedded in everyday life. The widespread exposure across platforms like TikTok or in quick DIY app-building tools has made AI feel valuable and already widely adopted.
While the enthusiasm is understandable, Denis also cautioned against overexcitement. The situation echoes elements of past tech bubbles, such as the dot-com era. Amid this hype, it’s essential to stay focused on meaningful, well-defined use cases.
In their discussion, Max and Denis also mentioned a major shift in how organizations approach AI, data, and content governance. Just a few years ago, enterprise AI efforts were tucked away in isolated teams with little visibility. Initiatives like data warehouses and governance frameworks were often seen as costly and disconnected from business value.
Today, that perception has reversed. AI has become a board-level topic. Management is now actively asking IT teams about their AI strategies. Meanwhile, employees can independently experiment with new AI tools.
Now, IT departments are often under pressure to support innovation while maintaining data security and governance.
Some IT teams made a decision to freeze AI tool usage. At the same time, others are more open and try to find balanced approaches and ensure both innovation and control.
Denis sees the increasing importance of governance frameworks as AI adoption grows within enterprises. High-quality, well-annotated data is essential, and organizations are beginning to understand why data governance matters.
Businesses typically work with huge volumes of unstructured data, including slide decks, PDFs, and other files. Quite often, these documents are outdated, duplicated, and poorly maintained. When AI models are trained on this disorganized content, the result is often low-quality outputs.
Moreover, another concern is about how people treat data. Employees frequently input sensitive data into third-party AI tools (and quite often, they don’t even realize it). This growing concern around data misuse highlights the need for stronger policies, awareness, and change management efforts.
For large enterprises, the complexity multiplies. They have different generations of workers with different backgrounds, various data types, and inconsistent governance practices. It’s a space where classic consultants still have an important role.
Denis mentioned an opinion voiced by Simon van Teutem, a young former McKinsey consultant and author. He argues that much of the industry’s brainpower is wasted. He suggests that if the talent currently tied up in consulting were redirected toward social initiatives, startups, or directly impactful roles, the societal benefits would be immense. Simon also criticizes the Dutch government for its heavy reliance on consultants.
Nevertheless, Denis disagreed with the idea that consultancy’s value is overstated. Experienced consultants bring fresh perspectives across industries and drive meaningful change. However, certain parts of the consultancy business model are likely to be disrupted by AI. This is especially true when it comes to repetitive tasks like interviewing, note-taking, meeting coordination, transcription, market research, and reusing past deliverables.
The consultancy sector is undergoing significant change. Companies now demand AI-enhanced services that deliver faster results and deeper insights. To stay competitive, firms must adapt. They need to specialize in high-value niches that justify premium rates or leverage offshore and nearshore talent to maintain quality at lower costs.
The world of entrepreneurship was familiar to Denis since childhood. Over 40 years ago, his father founded one of Belgium’s first IT consultancies. The company was focused on the sales of early computers and Microsoft training at a time when most businesses still relied on typewriters. It was an era of pure change management. And one of the tasks was to convince clients why digital tools could replace traditional methods. Today, Denis remains on the board, while the company is led by his brother.
Growing up in such an environment proved invaluable. Business discussions around the dinner table, early exposure to risk-taking, and seeing close family members start companies made entrepreneurship feel natural. There was no pressure to join the family business. However, it was an open opportunity.
Thanks to this, he was not afraid of starting new ventures. He always felt ready to take calculated risks. Moreover, he has direct access to a trusted network for quick feedback. And he considers this one of the greatest advantages in their career.
Denis doesn’t think that all organizational data must be 100% accurate. Instead of obsessing over perfection, it’s more important to focus on clear objectives. If these targets can be met with existing data, then over-engineering governance processes add little value.
Strong foundations and basic policies are important. But it’s crucial to avoid governance for the sake of governance.
At the same time, it’s worth highlighting that strict privacy rules, policies, and regulations can slow innovation in large organizations.
Their solution is to introduce controlled flexibility. They need to create safe environments where teams can experiment with new products and solutions without risking client relationships or compliance breaches. This could include dedicated innovation teams, collaborations with interns and students, or co-creation projects with existing clients.
It’s also crucial to stay close to the real problems. Quite often, it means just speaking directly with clients to understand what truly matters.
Denis mentioned Nokia, Blockbuster, and Intel as examples of companies that missed critical opportunities to pivot into new business models. Without the ability to disrupt themselves and explore new markets, organizations risk to face a slow and painful decline.
He contrasted two approaches: forced pivots (like Nokia’s late search for a new niche) and proactive innovation.
He also recollected Disney’s move into streaming. While Netflix reshaped the industry, Disney responded years later by launching its own platform. Disney saw this as innovation. Nevertheless, for Denis, it looks like replication. Now, Disney is trying to compete directly against pioneers.
True innovation isn’t about copying a competitor’s proven model. It’s about creativity, differentiation, and the courage to do something new before the market demands it.
For Denis, innovation is a broad term that covers speed, adaptability, and originality. He said that Disney could have pursued many paths beyond simply copying Netflix. With its iconic brand, Disney had the potential to leverage unique assets like the Mickey Mouse franchise to create something distinct.
Denis explained that in the current conditions, businesses will need to reinvent themselves far more frequently than in the past. As software and technology become commodities, both startups and large enterprises must think years ahead. Today’s advantage could be easily replicated tomorrow.
In his opinion, today, the VC mindset is quite popular. It is based on the idea of simply copying an existing business model to further apply it to a new market, scale quickly, and sell. This approach can be profitable, but it often prioritizes only short-term financial gain over building something truly original and sustainable.
But true breakthrough founders are those who discover unique solutions that defy conventional VC rules and established business models.
Denis mentioned Odoo as an example of this rare approach. The company ignored standard playbooks. They built open-source software and avoided aggressive early fundraising. They didn’t hire external managers to boost their growth and even expanded into India instead of Silicon Valley. And that was the company’s unique path to success.
While speaking about the role of AI in today’s jobs, Denis compared the automation of manufacturing to the impending transformation of white-collar work. In the past, car assembly lines relied on many workers, each of whom added a component. Today, machines handle most of the process, while humans act primarily as operators.
Denis believes a similar shift is coming for desk-based roles. Tasks like creating marketing plans, handling customer support, and generating financial reports may soon be performed primarily by AI agents. The human role will evolve into monitoring, verifying, and optimizing these systems rather than doing the work directly.
This change raises profound questions:
The complexity is heightened by the opaque nature of large language models. Unlike traditional machine learning, their reasoning processes are often unexplainable.
Acquiring AI knowledge is one of the first steps that should be taken before introducing AI into business processes. But the real challenge lies in embedding this technology effectively within organizations. It’s essential to ensure AI tools are safe, practical, and operational.
Denis advises people to start by identifying repetitive tasks in their own roles that could be automated. Even non-technical employees can experiment with AI tools like ChatGPT by asking simple, step-by-step guidance on automating those tasks.
Scaling this approach to teams and departments can lead to meaningful organizational impact.
The key recommendation is to begin with small, manageable experiments within a single team or department and build from there. It’s quite risky and challenging to implement complex, ambitious AI models that aim to disrupt everything at once. Incremental adoption will bring much more benefits.
Want to explore more about the role of technologies in businesses and the value of innovation? The next episodes of the Innovantage podcast are coming soon! Don’t miss them.
.svg)
.svg)
.svg)
.svg)
.svg)
.svg)
.svg)
.svg)
.svg)
.svg)
.webp)
.webp)
