We are living in uncertain times. The geopolitical landscape is chaotic, and the ripple effects throughout the economy are challenging at best. Tech entrepreneurs have the added uncertainty driven by the implications of AI, and the resulting skittish buying community. In my last post, I quoted Yogi Berra “The future ain’t what it used to be,” and I heard a lot of feedback from CEO friends.
What was most concerning was that several CEOs were quite fatalistic about the prospects for the future. They pointed to imbalances in the economics of innovative technology companies that will lower their potential valuation and significantly raise the bar to reach sustainable financial success. Here are several of the issues they raised:
Special Sauce - Every successful company needs a unique value proposition and defensible competitive differentiation — their special sauce. As one CEO put it, we can think about a software product in terms of three layers: data, application logic, and interface. An AI-based natural language interface layer has become nearly ubiquitous across applications. In just the last year or two, the ‘gee wiz’ of being able to ask questions or give directions to a platform in natural language has vanished, and it has become table-stakes for nearly every product. In other words, it has become hard to defend a competitive differentiation with a cool interface.
At the same time, it has become much easier to create application logic using AI tools. The pace of delivering capabilities has accelerated, so the longevity of competitive differentiation has diminished. What was easy for you to build will be easy for a competitor to cover. There is still value in subject matter expertise, but the scale of the engineering effort is no longer the competitive barrier.
So that leaves the data layer as the last potential competitive barrier. This one is interesting because companies with deep, proprietary troves of data have a unique asset that has value and may be defensible. Additionally, with AI tools, companies that have a deep data trove can mine it to effectively turn it into a competitive advantage that start-ups with a bright idea but no historic data cannot match. The potential counter to this argument is the ready availability of publicly-sourced data and agent-based tools to access and integrate it. My bet is on the companies with rich proprietary data, but time will tell.
The bottom line is that it is becoming increasingly difficult to defend a unique value proposition and competitive differentiation. As a result, it is challenging to outpace competitors and consistently grow at a rate that will attract a high valuation.
Stickiness - In a recurring revenue relationship, customer retention is critical. It is not uncommon for SaaS companies to spend as much or more than the first year’s revenue to acquire a customer. They are betting on client retention in subsequent years to be the source of profits. We calculate the lifetime value to customer acquisition ratio (LTV/CAC) as a key SaaS metric. However, in the evolving AI-driven market, customers are less sticky. The replacement cycle has accelerated. With natural language interfaces, the barrier to learning a new system is lower, and with agentic activity management, the level of user-learning and involvement to accomplish tasks is lower. The result is buyers can hop from one application to another fairly easily.
Profit Margins - So far, AI-driven companies are just not that profitable. Typical SaaS companies have better than 80% gross margin, and once they are established they can generate decent net income. AI-based companies have a different cost structure, most have gross margins well below 60% and underperform significantly on net income. AI requires massive computing costs and pricing models have not settled into predictable patterns. From the vendors’ perspective, consumption-based pricing seems to be the only way to manage costs and preserve margins, but many corporate buyers struggle to budget for unpredictable consumption costs, particularly when no one can explain how actions equate to consumption. The whole token-based AI pricing model is so complicated and unpredictable that application software companies are being trapped between customers that want a simple predictable licensing fee, and AI costs that are eroding their gross margins. Lower margins and profitability equate to lower enterprise value for raising capital, and lower value on exit, which is not an attractive model for investors or entrepreneurs.
Pivot - What most entrepreneurs are facing is the need to pivot their business strategy and model to incorporate the new dynamics introduced by AI. They need to figure out how to skate to where the puck will be, which may be a very different trajectory from how they have structured their business to date. Investors who made investments in the last five years are seeing expectations evaporate. Going forward, investors need to change their criteria and investment models to reflect the evolving new realities, but at this time, things are still pretty unsettled, so it is hard to predict what an investment made today will look like five years from now.
The best defense for entrepreneurs and investors is to return to fundamentals and manage a solid business every day. We see this in the dramatically increased focus on cash flow and positive EBITDA. Gone are the days when investors tolerated losses in search of growth at any cost. Unit economics and capital efficiency have become the principal metrics, and growth goals have become more modest. At the same time, we are seeing declining valuations and reduced expectations for multiples at exit. The net is that the landscape has become less attractive for entrepreneurs and investors. The message from the CEOs I spoke with recently was that launching and growing a tech business was already hard work, but the task has become even more daunting, while at the same time potentially less rewarding. The market will sort it all out in the next few years, but for now we are going through a transition that is exciting on one hand and frightening on the other. Buckle Up!