Listen to the audio version of this article (generated by AI).
Two months ago, software stocks were in freefall. Today, they’re within striking distance of all-time highs.
Same companies. Same AI threat. Completely different prices.
Something changed. The question worth asking — before you chase this rally — is what, exactly, that something was.
The SaaS Comeback Nobody Saw Coming — and What It Actually Means
The iShares Expanded Tech-Software ETF (IGV)— the benchmark index for software stocks — just made an unexpected comeback.
Back in April, IGV was sitting nearly 40% below its all-time highs, in bear market territory. People began questioning if the entire sector had a future. And some of those questions were legitimate — more on that in a moment.
But then the rally began. IGV ripped 45% off its lows in a matter of weeks. It blew through its 50-, 100-, and 200-day moving averages like they weren’t even there. Today, it sits less than 10% off its all-time highs.
This type of reversal off the 200-week moving average has only happened a handful of times in the past 15 years. Each time — in late 2011, early 2016, and early 2023 — turned out to be generational buying opportunities.
Technically speaking, this rally looks like the real deal. Institutional money came back hard and fast. Positioning is no longer washed out. The macro backdrop — no recession, tariff de-escalation, the AI capital expenditure cycle running full steam — is supportive.
For traders, fighting this tape in the near term is likely a losing game.
What the Rally Did Not Fix: The Three AI Waves Still Threatening SaaS
Now, here’s the uncomfortable truth: the stock prices recovered. The fundamental risks did not.
For years, Wall Street loved the Software-as-a-Service business model. Businesses would pay per employee per month to access a software platform that managed some part of its operations — sales pipeline, expense reports, project timelines, creative assets.
Recurring, predictable, high-margin revenue.
AI is now dismantling that very business model that made these companies worth hundreds of billions in the first place.
The disruption is playing out in three distinct waves, each more threatening than the last:
Wave 1: The Point Solution Wipeout
AI agents can now perform tasks without a SaaS subscription attached. Why pay $15 per employee per month when AI can now manage tasks for far less? The lowest-value software offerings are being hollowed out first, and the pace is accelerating.
Wave 2: The Pricing Compression Squeeze
For mid-market horizontal platforms — i.e. project management, customer relationship management, collaboration tools — the threat is subtler but equally dangerous. AI is reducing the switching cost of leaving these platforms.
If an AI agent can replicate much of what a software platform does at a fraction of the price, customers don’t necessarily churn immediately. But they start to negotiate. Renewal rates slip. Pricing power evaporates.
These are low-multiple businesses masquerading in high-multiple clothing.
Wave 3: The Business Model Disruption
Even the biggest, most entrenched platform companies — the ones with large enterprise relationships and genuine data moats — will survive. Though, to do so, they will have to transform from seat-based subscription businesses into consumption-based AI platforms.
Usage-based pricing sounds modern and exciting. It is also inherently lower-margin and lower-multiple than the model Wall Street has been paying 30x revenue for. This is not a crisis. But it is a permanent structural reset.
Beta vs. Conviction: How to Tell Which Software Stocks Deserve the Rally
IGV went up as a block. It will not come down that way.
When macro sentiment flips — fear turns to greed, institutional money re-risks — it buys everything in a sector first and asks questions later. That is what happened with IGV.
However, inside that ETF, there are companies with genuinely AI-native business models that will compound through this transition, and there are companies bouncing on pure beta that will re-test their lows the next time AI demonstrates its abilities.
The Compounders: SaaS Stocks That Benefit as AI Proliferates
The names worth holding are those that make up the nervous system of the AI economy — the infrastructure, security, observability, and physical-world data.
Palantir (PLTR), CrowdStrike (CRWD), Palo Alto Networks (PANW), Datadog (DDOG), Axon (AXON), Samsara (IOT): these businesses benefit directly from a world where more AI agents are running, more data is being processed, and more attack vectors need strong defense.
The Faders: SaaS Stocks Bouncing on Macro, Not Fundamentals
Then there are the names worth fading on this bounce. And this is where the SaaSmageddon thesis bites hardest.
Broader SaaS incumbents — legacy CRM platforms, creative tool suites, HR and payroll software, project management tools — face a more complicated road. Some are investing aggressively in AI and may survive the transition. But many are bouncing on macro tailwinds rather than fundamental improvement, and their pricing power story is getting harder to tell with each AI capability improvement.
Workday (WDAY), HubSpot (HUBS), and Adobe (ADBE) each face acute pressure from the second and third waves — bouncing hard on macro tailwinds while their pricing power stories quietly erode.
One of the most ironic shorts in the market right now is UiPath (PATH) — a company whose entire business is automating workflows, now being disrupted by better automation. UiPath built its model on robotic process automation: software bots that mimic human clicks, keystrokes, and navigation across legacy enterprise systems. It charged enterprise customers handsomely to deploy and manage those bots. Now, AI agents can do the same work — and increasingly more — without the rigid rule-based scripting UiPath requires, at a fraction of the cost, and without a dedicated implementation team. The product that was the future of automation is being made obsolete by the next version of it. The robots are eating the robot-makers.
The Bottom Line: Own the Nervous System, Fade the Workflow
In the near term, there is no reason to be aggressively bearish on software stocks. The technical setup is as good as it has been in years, institutional positioning supports continuation, and the macro backdrop is not fighting the tape.
But looking out 12 to 24 months? The fundamental reckoning that the market postponed is still coming. Enterprise AI adoption data — renewal rates, churn patterns, pricing concessions — will start to surface in earnings calls over the next several quarters. And as that happens, the distinction between AI-native compounders and beta-driven bounces will become impossible to ignore.
Own the nervous system. Fade the workflow.
Software that becomes more valuable as AI proliferates — security, observability, data infrastructure, physical-world intelligence — deserves a permanent place in your portfolio. The rest deserves skepticism, regardless of how good the chart looks today.
One thing this rally made clear? The market doesn’t wait for permission.
When institutional money decided software was worth owning again, it came back all at once, in a matter of weeks, before most retail investors had time to react.
The same thing will happen when OpenAI and Anthropic file their S-1s — except the repricing won’t be contained to one sector. It’ll ripple across the entire AI ecosystem simultaneously.
I’ve already mapped where I think that money lands first. Not the IPOs themselves — the companies underneath them that Wall Street will be forced to reprice the moment the filings go public.
.
