The Observability Layer Sitting Behind Every AI Workload

Every AI model in production needs someone watching it. One cloud software name runs that layer across the enterprise.

August 29, 2026

The AI trade everyone talks about is chips. The one nobody argues about at dinner? Who watches the models after they go live. One cloud infrastructure name owns that layer, and its next earnings print lands in early November.

AI wealth (Sponsored)

Some of the world's most influential tech leaders are investing heavily in what many consider the next phase of artificial intelligence.

According to one analyst, this shift could create new winners and losers across the stock market.

He's released a free Hotlist and Hitlist highlighting the stocks he believes may be best positioned—and those that may face the greatest challenges.

Get the Free Hotlist + Hitlist

Want to flag a company most people file under boring cloud software and completely miss the AI angle on?

It sits behind the scenes of nearly every major enterprise cloud deployment you can name, and with a Q3 print landing in early November, there's a real window to get positioned before the AI observability story resurfaces in analyst notes for Datadog (NASDAQ: DDOG).

Before you scroll, hear me out. This isn't the crowded, obvious AI trade. It's the picks-and-shovels layer underneath it.

Every enterprise pushing generative AI into production needs a way to watch what those models actually do once they're live, and Datadog's observability platform sits directly on that workflow.

Action: build the position in stages ahead of the Q3 print in early November.

The most recent quarter set the bar with revenue of $1.12 billion, up 36% year over year, non-GAAP EPS of $0.65 against guidance of $0.57 to $0.59, and full-year growth guided around 30%. Size the entry before the report, not after it.

What Just Happened

Datadog most recently posted another beat-and-raise print, and the details matter more than the headline. Revenue growth came in at 36% year over year. Management lifted full-year guidance again.

Trailing net revenue retention sits in the low 120s, meaning existing customers are spending materially more, not less. That's the number you care about most in a usage-based model.

The other piece: AI-native customers now make up a meaningful and growing chunk of consumption. Free cash flow margin sat in the low-30s, and gross margins stayed above 80%.

Nothing about the setup looks like peak growth to you. It looks like the AI cohort hasn't fully shown up yet.

New opportunity (Sponsored)

A major policy change is reshaping the U.S. energy landscape.

While incentives for several energy technologies are changing, one energy source continues to receive long-term support.

A new presentation explains why one analyst believes this development could create opportunities—and the company he's watching most closely.

Watch the Free Briefing.

What The Business Actually Does

Datadog runs the monitoring plumbing for modern cloud applications. When your banking app freezes at 3 PM, when a streaming service stutters, when an AI assistant starts spitting nonsense, engineering teams pull up Datadog dashboards to figure out why.

The platform ingests logs, metrics, traces, and now the behavior of AI models themselves.

Here's why you care. Every new cloud workload adds data to monitor. Every AI model deployed to production is the heaviest workload most engineering teams have ever managed.

GPUs run hot, inference costs stack up fast, and one broken pipeline can burn six figures overnight. The observability layer that watches all of it isn't optional. That's the flywheel, and the AI wave is pouring rocket fuel into it.

High-Rated AI (Sponsored)

Louis Navellier’s proprietary Stock Grader system helped flag major winners years before they became household names.

Now, that same $9 million system is flashing its highest rating on one AI stock with 28% year-over-year sales growth and more than 30,000 patents.

He is giving away the name, ticker, and full analysis for free.

See the free AI stock pick here.

Why The Market Cares Again

AI observability is expanding fast. If you're pushing LLMs into production, you're figuring out you need to watch model latency, cost per token, hallucination rates, and infrastructure load together.

Datadog rolled out LLM Observability specifically for this, and adoption has been quicker than most Street models assumed.

Enterprise cloud budgets are unfreezing too. For two years, CFOs slashed cloud spending and consolidated vendors. That cycle has clearly turned. Recent commentary from the hyperscalers points to reaccelerating consumption, and you can see Datadog's revenue tracking that curve closely.

Tool consolidation favors the platform. Enterprises are cutting the number of monitoring tools they run. If you're the platform absorbing workloads from Splunk, New Relic, and a half-dozen point tools, that's a tailwind you don't have to earn. You just have to not screw it up.

Action: Watch Q3 results for AI cohort commentary and any incremental disclosure on LLM Observability adoption.

What The Financials Are Signaling

Net revenue retention in the low 120s. This is the single most important number in the story. It tells you that existing customers are expanding usage, which, in a consumption model, means the AI workloads landing this year are still ramping. Look for management to reaffirm or nudge this higher on the November call.

Free cash flow margin in the low-30s. Datadog is one of the rare SaaS names that grew through this cycle without lighting cash on fire. That discipline means they can invest in AI product without needing to dilute you or take on debt.

Customer count over $100k in ARR is still growing. Big customers keep landing. The mix continues to shift toward enterprise, which is where the AI budgets actually live. If this number keeps climbing at the current pace, it validates your whole thesis.

The Valuation Problem You Can't Ignore

Not cheap on any traditional measure. You're paying a premium multiple for Datadog, plain and simple. The stock trades at a level that assumes execution stays clean and AI workloads keep flowing.

Any hiccup, any miss on retention, any hint that AI-native customers are pulling back, and the multiple compresses fast.

Usage-based cuts both ways. When customers scale, revenue scales. When customers optimize, revenue drops in the same quarter. No long-tail subscription cushion here. That's the trade-off you accept for the AI upside.

What Needs To Happen Next

Q3 earnings in early November. This is your main catalyst. Watch for another retention beat, AI cohort commentary getting more specific, and a guidance raise into year-end. If management sounds more confident on AI monetization, the stock re-rates.

DASH conference product updates. Datadog's own user conference typically produces the year's biggest AI product announcements. New agentic monitoring features and cost governance tools for LLMs would meaningfully expand the addressable market.

Enterprise deal size expansion. Any commentary about seven- and eight-figure enterprise contracts closing tells you AI budgets are landing where the bulls hope they land.

The Risks You Should Take Seriously

Cloud spend rationalization returns. If the macro softens and CFOs start optimizing cloud bills again, Datadog's usage-based revenue is the first thing to feel it. You saw the 2023 version of this movie. You don't want the sequel.

Hyperscaler competition. AWS, Azure, and Google Cloud all have first-party monitoring stacks and every incentive to bundle them into enterprise deals. Datadog has won this fight before, but it's a permanent overhang.

Concentration risk in AI-native cohort. The AI customer base is exciting but concentrated. If a couple of large AI labs cut infrastructure budgets or move workloads in-house, you'll see it in Datadog's numbers before anyone else's.

Action: Hedge with a broader cloud infrastructure basket or large-cap tech ETFs to shield against hyperscaler competition and macro cloud spend risk.

How I'd Frame A Position

Build in tranches, not all at once. Split your intended position into three pieces. Take the first at current levels, keep dry powder for a pullback toward the lower end of the entry zone below, and save the last third for post-earnings volatility in November either way.

If you already own it, sit tight through the print. The setup into Q3 looks favorable and the AI narrative is only getting stronger. Consider trimming only if the stock runs another 20%+ into the report without a fundamental update.

Where I'd get out. If net revenue retention drops below 110% on any print, or if management sounds cautious on AI cohort spend, the thesis is broken. Reassess below.

Where This Leaves You

Datadog is the observability layer for the AI buildout, and the market keeps pricing it like a standard SaaS business. Q3 earnings in early November are the next catalyst, retention is holding, and the AI-native customer cohort is still ramping.

Yes, the multiple is rich. But if you want AI infrastructure exposure without paying the semiconductor markup, this is where you want to be looking right now.

Setup Scorecard

Entry Zone: $231–$248

Target: $279

Stop Loss: $215 — a weekly close below that breaks the trend and the thesis needs a rethink

Catalyst Timeline: Q3 earnings early November, DASH product conference in Q4, hyperscaler capex commentary through fall earnings season

Confidence Level: Medium-High. Fundamentals are solid and the AI tailwind is real, but the valuation leaves less margin for error than I'd like.

That's our coverage for today; thanks for reading! Reply to this email with feedback or any tech stocks you want me to check out.

Best Regards,
—Noah Zelvis
Tech Stock Insider