Product Management
March 17, 2025
9 min read
How can tech startups survive today? How to find a good idea that will rock the market? Who can help you to guide your team if you have a limited budget?
To discuss these topics, Innovantage podcast host Max Golikov, who is also the CBDO at Sigli, invited Laimonas Sutkus to join him in his studio. Laimonas is a person with robust expertise in helping businesses launch their projects and manage tech teams in such highly competitive fields as AI, fintech, health tech and others.
In his career, he has gained experience as a software developer, tech advisor, CTO, and fractional CTO, working with businesses at different stages of their development. In this episode of the Innovantage podcast, Laimonas spoke not only about his professional path and the peculiarities of the tech industry landscape today but also shared valuable insights and practical recommendations for startup founders.
Laimonas began his fractional career in early 2024. As he admitted, before that he even hadn’t known that such roles exist nowadays. According to him, he discovered the concept by chance through a LinkedIn post from another fractional CTO. This inspired him to explore the field.
A fractional CTO operates as a hands-on consultant and provides technical leadership to companies that don’t require a full-time CTO. This role is particularly beneficial for non-technical businesses like marketing agencies and small pharma companies, as well as early-stage tech startups. Such teams may not need a full-time executive but they still require expert guidance to avoid common pitfalls.
Unlike a traditional CTO, a fractional CTO is available on a part-time basis. It can be a few hours per day or even just a few hours per week.
What is important to highlight here is that this person is not a third-party consultant. This specialist is a full-scale team member, despite the limited hours that he or she devotes to your business per week.
This expert helps businesses navigate technical challenges, streamline processes, and make informed decisions.
The fractional model extends beyond CTOs to other executive roles, such as fractional CMOs and CFOs. And all these roles follow the same principle. These professionals provide their strategic expertise without being full-time employees.
For a little bit less than a year, Laimonas worked as a fractional CTO. Nevertheless, now he has a full-time job. And here are the key pros and cons of a fractional role that he defined.
Among the benefits, Laimonas highlighted the flexibility and security that come with a fractional career. Fractional employees can choose their projects and work with multiple clients.
Moreover, this approach helps to reduce financial risk. If you lose one or two clients, it doesn’t mean that you will lose all your income at once.
In other words, a fractional executive operates as a one-person business and can maintain great autonomy.
However, this independence also comes with challenges. Fractional professionals must handle not just their core expertise but also a wide range of other tasks, including sales, marketing, and client acquisition. All these activities are traditionally managed by entire departments in a business.
As a result, Laimonas shared that a significant portion of his time was spent on prospecting, lead generation, and outreach rather than on his actual technical work.
For specialists like fractional CMOs, CFOs, or CTOs, the ideal scenario is to focus solely on their expertise. In Laimonas’ case, his passion lies in technology, not in sales or marketing. Constant business development efforts could be very draining and that’s the key disadvantage of this career path.
As artificial intelligence remains one of the most widely discussed topics today, Max and Laimonas also couldn’t omit it in their conversation.
Laimonas joined the AI space long before this technology became mainstream. He has been building AI-based products since 2014.
Over the years of his work, he witnessed how AI development has changed with the emergence of large language models like ChatGPT. Previously, AI required hands-on data science, machine learning experimentation, and model deployment. Today, AI is more accessible. Developers can integrate it into products with simple API calls, avoiding the need for complex model training. This shift has allowed businesses to incorporate AI quickly and transform non-AI products into AI-powered solutions sometimes in a matter of hours.
According to Laimonas, earlier many startups approached AI as a standalone product rather than a tool. Laimonas mentioned Rabbit R1 and AI Pin as examples. These are gadgets designed to function as AI-powered assistants. Nevertheless, they failed. It happened because they lacked a strong foundational business model.
Today, it has become obvious: AI is not a product in itself but a feature that can enhance existing solutions.
Laimonas believed that in the future AI will continue to be a powerful tool for gaining a competitive advantage. However, success will depend on integrating AI into solid business ideas. It will work much better than just relying on AI as the core offering.
According to the article published by Sequoia, one of the biggest VC firms, the vast amount of capital poured into AI-based solutions now requires an additional $500–600 billion in revenue across these companies for investments to break even. At the moment, it’s difficult to say whether this target is achievable or not. However, it brightly highlights the significant financial pressure on the AI sector.
Laimonas mentioned that the gap between business profitability and AI investments exists not only for startups but also for major players like Google, Meta, and Microsoft. These tech giants lead AI development today because only they can afford the immense costs of training large-scale models. Such efforts often require tens or even hundreds of millions of dollars.
Despite such a market situation, investors remain optimistic. This can be seen in the steady growth of the S&P 500 index, which tracks the stock performance of 500 of the largest companies listed on the US stock exchanges. However, here we can observe a notable concentration on the so-called “Magnificent Seven”. Seven major tech firms (Microsoft, Meta, Tesla, Amazon, Apple, NVIDIA, and Alphabet) make up nearly 30%-35% of the index.
The last time when such a concentration was observed was in the dot-com bubble era.
Laimonas sees obvious similarities between the current AI hype and the early 2000s internet boom. The internet was also a revolutionary technology. It went through a speculative bubble that eventually crashed before stabilizing into long-term growth.
Could this happen to AI as well? The expert believes AI is following a similar trajectory. There was an initial boom. Now we can expect a likely correction that will ultimately result in a lasting impact.
According to Laimonas, AI is definitely a very good technology. Nevertheless, it is already being weaponized. Deepfake videos of world leaders, AI-generated propaganda, and automated disinformation campaigns are becoming widespread. Large language models, when integrated into social media platforms, further amplify misinformation. That’s why it’s also worth taking into account this “darker” side of AI while analyzing its role in our society.
Laimonas emphasized that one of the most important lessons for new founders is accepting that their initial ideas can be flawed. In the beginning, a startup’s vision is rarely perfect, and founders must be willing to refine it. Instead of treating an idea as something sacred, they should focus on building a minimum viable product (MVP), testing it, and gathering feedback.
The reality is that most early concepts will fail. However, failure is part of the process. Founders must continuously iterate. This should include seeking feedback, adjusting the product, and repeating the cycle. All this should be done again and again until product-market fit is achieved. The key is to remain adaptable and recognize when something gains traction.
However, not all feedback is equally valuable. Some users may explicitly state why they don’t like a product. For example, they may explain that they stopped using a product because the price is too high or because it doesn’t address some of their needs. That’s a very helpful type of feedback.
Nevertheless, more often, the feedback is implicit: users simply don’t engage. In such cases, founders must investigate why it has happened. This requires reaching out to former or inactive users, analyzing usage patterns, and identifying the reasons behind low adoption.
Deep, specific feedback is crucial to making the necessary improvements that lead to success.
In early-stage startups, achieving product-market fit requires rapid iteration cycles. The faster a startup can implement and test changes, the higher its chances of success will be. The chosen technology plays a crucial role in this process. It can either accelerate development or become a bottleneck. It is the responsibility of a technical co-founder, fractional CTO, or experienced consultant to ensure the right technological choices are made to support fast iteration.
Traditionally, technical teams are structured with dedicated backend developers, frontend developers, QA specialists, and sometimes mobile engineers. While this model worked well in the past, it is often too slow for modern startups that need a competitive edge.
As a response to such market needs, full-stack frameworks and technologies have started gaining popularity. They integrate multiple aspects of development into a single streamlined system.
Frameworks like Next.js and Vercel provide infrastructure, frontend, and backend capabilities in one codebase. As a result, they enable faster deployment and iteration. However, these technologies come with some pitfalls, such as vendor lock-in. To fully unlock Next.js’s benefits, software developers often need to use Vercel, which can be costly and restrictive.
Other frameworks, such as Remix, offer an alternative approach. For instance, Remix allows developers to write frontend and backend logic within the same file. This might seem disorganized at first. However, following strong design principles can result in a well-structured and efficient system.
A single full-stack developer in such a case can often outperform a traditional five-person team consisting of separate frontend, backend, and QA engineers. The key advantage lies in eliminating communication overhead and reducing knowledge gaps. In other words, one developer can deliver all features without dependencies on other specialists.
This shift toward full-stack development, combined with AI-assisted coding tools, significantly shortens iteration cycles. Features that previously took a full day to implement can now be developed in a fraction of the time.
For startups aiming to stay agile and efficient, prioritizing generalist developers, who can build entire features independently, is more effective than hiring narrow specialists. Specialization should come later when the team grows to a size where dedicated roles in infrastructure, frontend, backend, and QA become necessary. Initially, focusing on generalists ensures maximum speed, flexibility, and resource efficiency.
Startups must strike a balance between focusing on immediate survival and planning for the future. While long-term vision is important, over-prioritizing future scalability at the expense of present execution can be fatal. If resources are not managed well and iteration cycles are too slow, a startup risks running out of cash before it ever reaches the future it envisions.
The priority should always be profitability and survival.
Scalability issues, expansion challenges, and the need for team specialization are all positive problems. They signal that the business is working, clients are coming in, and revenue is growing. Growth problems indicate success, whereas failure to manage short-term sustainability can lead to an early shutdown.
Some kind of uncertainty is an inherent part of the tech industry. Tech teams constantly need to solve scalability problems. While the nature of these problems evolves, the challenge itself never disappears. Mature IT leaders and software developers must recognize this uncertainty and design solutions, architectures, and infrastructures that accommodate future changes.
A well-structured codebase should reflect the uncertainties of the business. It must be flexible enough to adapt to different directions as the company expands. Designing with adaptability in mind ensures that as business needs shift, the technology can keep up without requiring a complete overhaul.
For early startups, it is also vital to have people who will professionally guide them at least at the initial stages of their development.
While many mentorship services are available online, they often lack a very important element. This key element is trust. It is difficult to assess a mentor’s true experience, expertise, and quality of services without firsthand knowledge.
Instead of relying solely on external help from the internet, startup founders should first turn to their personal networks, including friends, former colleagues, business partners, and industry acquaintances. These trusted connections can either offer direct guidance or introduce founders to experienced professionals within their networks.
Human connections are invaluable. In the startup world, relationships often open doors to mentorship, partnerships, and new opportunities that wouldn’t be accessible otherwise. Entrepreneurs should prioritize building and maintaining strong professional relationships, as these connections often prove more beneficial than any formal mentorship services.
The journey of building a tech startup is filled with challenges: from finding the right idea to managing scalability. According to Laimonas Sutkus, flexibility and readiness for iterations are among the key components that can drive a tech startup to success.
Want to learn more about technologies and their role in the business world? Don’t miss the next episodes of the Innovantage podcast where its host Max will welcome new experts in his studio.
.svg)
.svg)
.svg)
.svg)
.svg)
.svg)
.svg)
.svg)
.svg)
.svg)
.png)
.png)
