A product launch and a large contract are both better starting points than a vague promise to “lean into AI.” They are still starting points.
For software, I want to know whether a feature wins incremental revenue, protects an existing subscription or simply becomes another cost of staying competitive. Those are three legitimate business outcomes, but they deserve different valuations. A feature that customers expect for free cannot carry the same profit forecast as a product they choose to pay extra for.
For infrastructure, the arithmetic reverses. The contract may be visible before the owner has spent all the money needed to serve it. A large headline value says little about timing, cancellation rights, utilization guarantees or the residual value of the equipment at the end. I would rather see a smaller contract with clearer economics than use a bigger number as a substitute for them.
Friday offered a more useful AI discussion because there were actual developments to inspect. My job is to follow each one far enough down the income statement and cash-flow statement that the word “demand” means something measurable. Otherwise I have replaced a technology slogan with a contract slogan and called it analysis.