Dynatrace: The AI Recovery Is Real, But the Multiple Has Run Ahead of the Billings

Dynatrace's stock has made the round trip in eight months. It was crushed alongside the software group in February, when the market panicked that AI could make whole categories of software redundant; the shares bottomed near $31.64. Now, at about $55, the stock sits just under its 52-week high of $55.76, up roughly 51% over the past four months. The engine of both moves is the same: AI. The difference is that this time Wall Street is betting Dynatrace's observability platform—the tooling companies use to monitor, secure, and fix the IT systems behind their applications—is a place where AI creates paid demand instead of destroying it.

The bull case is measurable, not a slogan. In the quarter ended June 30, Dynatrace said more than 1,000 customers now use it to monitor AI and large language model workloads, up from about 850 the prior quarter. Agentic automation—software that acts on its own—went from roughly 500 customers a quarter ago to more than 800. Management also pointed to record new-logo growth, with average deals above $285,000. This is the kind of evidence that separates an AI narrative from AI demand: usage, customer counts, and deal sizes, not a slide deck.

The financials back parts of the story. Revenue grew 16.2% to $554.5 million, a modest beat, and adjusted profit of $0.48 per share beat by about 8%.adjusted profit of $0.48 per share beat by about 8%. Annual recurring revenue rose 17.2% to $2.14 billion. The quality of the model is genuinely good: gross margin is above 80%, and free cash flow came in near 27% of revenue—a real cash-generating business, not a burn machine.

Here is where the recovery gets harder to swallow. Even as it beat the quarter, management trimmed full-year revenue guidance to about $2.31 billion at the midpoint, down roughly 0.6%. Cutting the target while beating is a tell that the back half is softer than the story implies. And the leading indicator says the deceleration is real: billings in the quarter rose just 7.8% year over year—less than half the pace of ARR growth. Billings are the bookings that arrive before the revenue does; when they run far below ARR growth, the top line tends to catch down to them, not the other way around.

So what does the stock's rebound actually cost now? After a 51% run in four months, Dynatrace trades at roughly 7.6 times trailing sales and about 83 times forward earnings. That is a healthy multiple for a company growing 16%, and it is the crux of the current decision. The February crash created the classic "cheap enough" setup the moment the multiple reset faster than the operating reality. That window has closed. At $55, the price is once again paying for the AI land-grab in advance—while the reported billings, the most direct evidence of that land-grab, are growing at half the rate of the headline.

For context, the comparison cuts both ways. Datadog, the closest comparable, trades near 21 times sales but grows revenue at more than 30%. Measured per unit of growth, Dynatrace is actually the cheaper of the two—roughly half the sales multiple for half the growth. But "cheaper than an expensive peer" is a weak reason to buy a stock at a 52-week high. The relevant question is whether Dynatrace's 7x-plus multiple is already discounting the acceleration it has not yet shown.

That makes the next two to four quarters the proof window, and the metric to watch is billings growth closing the gap with ARR growth. The AI adoption numbers are encouraging—customers who are already paying for LLM observability, agentic automation, and a log-management business management says nearly doubled in the last two quarters. If that usage converts into faster billings, the multiple can be supported. If billings stay in the high single digits while ARR keeps sliding toward them, the stock has re-rated ahead of the evidence.

Dynatrace is a good company, and the market is not wrong that AI has given its platform a durable reason to exist. But a good company and a good stock are different things. After the recovery, the margin of safety the February reckoning provided is gone. The honest read at $55 is not "sell"—the balance sheet is clean, cash flow is strong, and the product direction is sound. It is "too early to chase." The buy-the-dip window was opened by the crash and has largely been closed by the rally; the next entry is earned the old-fashioned way, when the bookings catch up to the narrative.