PARTNER CONTENT
Not all that’s called AI is true IP – knowing the difference helps investors avoid costly missteps, says Lukasz Lazewski, CEO of LLinformatics.
Digital transformation is never linear. Indeed, sticking to the plan is often the biggest challenge – the journey from proof-of-concept (POC) to return on investment (ROI) being fraught with barriers and budgetary constraints.
For PE sponsors, a botched transformation can spell a difference of many multiples of return. As such, a forensic approach is critical across the investment lifecycle: from spotting risks and red flags during technology risk diagnostics; through tracking performance and value levers in the transformation phase; to implementing easy valuation wins before exit.
According to Lukasz Lazewski of LLinformatics, it’s easy for firms to misread the signals at various phases of investment.
Pre-deal
A common misconception during due diligence relates to the viability of AI-enabled transformations. Lazewski says: “AI is often seen as a silver bullet solution that can immediately generate efficiencies for most businesses – a perception that frequently drives investment. However, it is crucial to first determine whether those efficiencies truly stem from AI, or whether they are the result of other factors, such as process automation or cost-saving measures like outsourcing.”
Secondly, AI comes with a host of secondary regulatory considerations. “A healthtech company, for instance, could develop a fantastic AI product but fail its regulatory clearance due to data compliance and governance issues. In some cases data governance laws may dictate that companies have to work with local or regional data providers wherever they operate – which entails a host of additional costs that may not be immediately apparent.”
Another consideration with the rapid evolution of AI is the gap between the real and perceived value of a technology stack. A company may present its digital infrastructure as AI-powered and highly differentiated – an attractive prospect for investors. Yet, if in practice the “AI” is merely an integration with an external provider, rather than proprietary IP, the actual value of that stack may be far lower than it appears.
“On the flipside, AI’s ability to dramatically improve digital infrastructure can and should be seen as a value driver where relevant.”
Outside of AI, common red flags during technology risk diagnosis include unrealistic propositions and slower-than-expected tech development. It is important to distinguish between symptoms and root causes in these cases, since the underlying drivers are often more complex than the apparent underperformance of product or development teams.
The holding period
Lazewski notes that firms frequently discover material underperformance in certain functions – not just a few individuals, but entire teams where output is 15–20% below what their headcount would suggest. “Their natural response? To hire more developers or double down on their investments – which treats the symptoms rather than the cause.“
There are often very clear people-related reasons for slow progress. “It could be as simple as staff slacking off, or more political factors such as developers/leadership slowing down tasks to prolong their relevance. Lacking the tech expertise to understand the true scope of work, companies often fail to identify these blockers.
Beyond that, the structure of a development team also plays a critical role. Many firms rely on a large pool of coders with limited product insight, guided by just a few ‘product owners’. In such cases, the bottleneck isn’t developer capacity but product leadership, and adding more product expertise, rather than more developers, drives better outcomes.
Indeed, having the right people in place – from the CTO down to the coder – is critical to driving long-term value in a portfolio company. When teams are well-structured, companies can avoid inefficiencies and achieve results far beyond what raw headcount alone would suggest, according to Lazewski.
Pre-exit
In the run-up to an exit, companies should shift their focus from building long-term fundamentals to unlocking the value levers that can rapidly boost valuation. One example, Lazewski notes, is shifting from service-based or manual-delivery models towards productised solutions – such as Software-as-a-Service subscriptions.
“Companies can introduce automations so pricing models can work on their own without protracted processes. If clients, particularly established ones, can self-serve and pay as they go then the entire proposition becomes several times more valuable at exit”.
At this stage, Lazewski suggests that integrating AI in meaningful ways can also help signal an asset’s future potential, with tangible applications often amplifying market perception and valuation. Lazewski suggests AI can also be leveraged for its hype value. Implementing some form of AI-enabled efficiencies can rubber stamp an asset and demonstrate the scope for future value creation.
With genuine tech expertise becoming increasingly critical in PE, LLinformatics supports firms with identifying the above challenges and opportunities across their portfolios – diagnosing common technology risks and pitfalls that may hinder long-term progress.
Lukas Lazewski, CEO of LLinformatics – Lukas is a software engineer and entrepreneur with nearly 2 decades of experience working with and advising technology companies across Europe and the United States. He began his career as a software engineer at AOL and Advertising.com before moving into CTO and other senior leadership roles at companies including Viewlabs, Brandnew IO, and Zenloop. In these roles, Lukas developed a reputation for using software and technology to solve complex technical challenges, helping to scale startups into international players as well as driving sector-defining product innovation for regulated industries. In 2012, Lukas co-founded LLInformatics, where he now serves as CEO. Headquartered in New York, Berlin, and Warsaw, the firm has grown to 130+ senior-level specialists serving a variety of companies and organizations. Under his leadership, LLInformatics has become known for rescuing failing projects, modernizing legacy systems, and delivering compliance-ready software products on time and on budget. The company boasts a 120% YoY rebooking rate and 3+ year average client relationships. Lukas holds a Master’s degree in Computer Software Engineering from the Polish-Japanese Academy of Information Technology and a postgraduate degree in Business Intelligence from SGH Warsaw School of Economics. He is also a member of the Entrepreneurs’ Organization (EO).
