In just one month, three B2B companies from different sectors were acquired for figures close to 3,000 million dollars each. At first glance, the businesses couldn't be more different: a customer service platform, an industrial maintenance platform, and a third specialized in data orchestration for power plants. However, analyzing the details of each operation emerges a common thesis that is redefining the rules of the game in the business world: the real jewel is not the artificial intelligence model, but the proprietary data that feeds it. This phenomenon not only marks a before and after in M&A strategies, but offers key lessons for any company aspiring to build a lasting competitive advantage in the era of AI for enterprises.
The first of these operations was Autodesk's $3.6 billion purchase of MaintainX. MaintainX was born in 2018 with an idea as simple as it is powerful: to provide plant and maintenance workers with a mobile tool as intuitive as Slack. Rather than just being a maintenance management system (CMMS), the company absorbed adjacent functionalities: inspections, safety, asset intelligence, spare parts prediction, and regulatory compliance. With revenue of approximately $115 million annually in 2026 and growth of 50%, Autodesk paid out about 26 times future revenue. Why such a high price? Because MaintainX wasn't just selling software: it captured in real time how physical assets behave in real-world conditions, a flow of data that Autodesk, with its design tools, could never generate from engineers' desks. That operational telemetry is the fuel that makes artificial intelligence accurate and actionable.
The second major transaction was Salesforce's acquisition of Fin (formerly Intercom), also for $3.6 billion. Intercom had gone through a growth crisis until, after the launch of ChatGPT, it pivoted to an AI agent for customer support. Fin's AI business line reached $100 million in annual recurring revenue with 350% year-over-year growth, while the legacy messaging business was broadly flat. Salesforce paid 9 times the total combined revenue, but the operation hides a very different reality: on the AI side, the multiple shoots up to 36 times, aligning with that of MaintainX. What Salesforce bought was not the Apex model that Fin had trained, but the 30,000 customers and the history of support conversations. That corpus of data is impossible for a competitor to replicate and turns the AI agent into an asset that is reinforced with each interaction resolved.
The third case, although less mediatic, is perhaps the most illustrative. Schneider Electric acquired Cognite for $3.1 billion. Cognite, born in Norway as a spin-off of the Aker industrial group, has been solving the problem of orphaned industrial data for almost a decade: most of the information generated by refineries, power plants and factories is scattered in silos, mislabeled and therefore useless to AI. Its Data Fusion platform contextualizes decades of operational data using a knowledge graph, on which Atlas AI is based, an AI agent system that allows operators to make decisions in real time. Schneider paid 18 times Cognite's revenue, a multiple that reflects the scarcity of that layer of contextualization. Without it, any AI model lacks the context needed to be useful in an industrial environment.
The fascinating thing about these three operations is that, despite being different sectors, the valuation pattern is identical: the multiple paid is approximately half of the growth rate. MaintainX grew at 50% and was paid at 26x; Cognite grew at 36% and paid out at 18x; Fin (combined business) was growing in the mid-190s and paid at 9x, but if you isolate the AI line at 350%, the multiple is back in line with the others. This means that buyers are not paying for current revenue, but for the speed of change and, above all, for the underlying asset that drives that growth: proprietary data.
For companies that are building AI-based solutions, these acquisitions send an unmistakable signal: the real competitive moat is not the model, which is commoditized month by month, but the data that only your product can generate. If your platform accumulates information that no foundational model can obtain on its own, and if that information becomes more valuable with each use, then you have an asset that big players will be willing to pay a premium to acquire. At Q2BSTUDIO we understand these dynamics and help companies design AI solutions that turn operational data into real competitive advantages.
But it is not enough to have data; you have to know how to capture them, contextualize them and make them accessible to the models. This is where apps come into play as we build in Q2BSTUDIO. From maintenance management systems to customer service platforms, we integrate AI agents capable of learning from each interaction, generating a virtuous cycle of continuous improvement. In addition, our expertise in AWS and Azure cloud services ensures that data is available in a secure and scalable way, while our cybersecurity solutions protect your business's most valuable information.
Another relevant learning from these three transactions is the importance of clearly separating AI business lines from traditional business. In the case of Fin, the explosive growth of the AI agent was diluted within total revenues that barely grew. Shoppers are looking closely at that metric: an AI line that grows by 300% is worth much more on its own than the whole. That's why we recommend our customers structure their products in such a way that the performance of AI-based components can be isolated. This not only makes it easier to attract investment, but also positions the company for future corporate operations.
The third case, that of Cognite, reminds us that industrial data is a treasure yet to be exploited. Many companies have decades of operational information trapped in spreadsheets, legacy systems, or disjointed databases. Transforming this data into a knowledge graph that can be exploited by AI is one of the greatest challenges and opportunities of the decade. At Q2BSTUDIO, through our business intelligence services and Power BI, we help organizations visualize and model that information, laying the groundwork for AI agents to act on. And all this with a focus on process automation, which reduces costs and accelerates decision-making.
In short, these three million-dollar acquisitions are not isolated cases, but the symptom of an unstoppable trend: the value is no longer in the software itself, but in the data it generates and in the ability to turn it into actionable intelligence. Companies that manage to build products that accumulate proprietary data naturally, that grow at high rates and that know how to separate their commitment to AI from traditional business, will be the ones that dictate the next big operations. At Q2BSTUDIO we work every day so that our clients are not only spectators of this revolution, but protagonists.