The Leading Public Software Companies Are Now Down -50% in the Last 6 Months

The SaaStr.ai Index of the Top 25 Public Software Companies just crossed -50.5% over the past six months. October 2025 to April 2026. Half the market cap of the leading public B2B software companies — gone.

Not -20%, which would feel like a correction. Not -30%, which we’ve seen before. Down Fifty percent.

This is not a blip. This is a structural re-rating of an entire asset class.

What the SaaStr.ai Index Is Telling Us

The SaaStr.ai Index tracks 25 leading B2B and software companies — the names founders and investors have benchmarked against for a decade. Companies like Datadog, Zscaler, GitLab, Rubrik, Klaviyo, UiPath, Workday.

In March, the index had declined approximately 23.6% year-to-date, with DigitalOcean as the lone standout at +49.6% YTD, and Atlassian as the biggest laggard at -53%. By April 10, the 6-month cumulative loss hit -50.5%.

That number represents something that has never happened in our industry before.

For the first time ever, public software companies now trade at a P/E discount to the S&P 500. Not at parity. Below. We’ve never seen this. Not in 2022 when rates spiked. Not in 2008. Not even in the dot-com unwind, which was a speculative bubble collapsing — not a structural attack on the business model itself.

The forward P/E multiple for application software has collapsed from 84x in 2021 to 22.7x today. The market’s implied long-term growth rate for a representative public SaaS company has dropped from 4.7% just three months ago to 1.1% today.

The market is effectively saying: we no longer believe software is a premium business.

Two Forces Are Hitting at the Same Time

It’s important to separate these, because conflating them leads to bad strategy.

Force 1: Budget displacement.

CIOs have finite spend. When Anthropic is now at $19B+ in annualized run rate — up from $9B at year-end 2025, growing $6B in a single month of February — that spend has to come from somewhere. Approximately 75% of new hyperscaler infrastructure spending in 2026, over $450 billion, is targeting AI infrastructure. That money used to flow to Salesforce seats, ServiceNow modules, HubSpot licenses, Datadog tiers. It is being redirected right now, this quarter, measurably.

Force 2: Substitution fear.

AI agents may make traditional software structurally less valuable over time. Seat-based models depend on headcount growth. If agents replace seats rather than complement them, the revenue model doesn’t just slow — it reverses. The market is pricing this in at the terminal value level, which explains why current earnings don’t fully explain the magnitude of the decline. Roughly 85-95% of enterprise value in a SaaS DCF comes from terminal value. Investors aren’t debating this quarter. They’re debating whether this business has a sustainable moat at all.

Both forces are real. The second one may be more fear than reality at this exact moment. But when $2 trillion in market cap disappears, the market has made a judgment.

Who Is Actually Getting Crushed

Look at what’s happened to the category leaders over the past 6-12 months:

Atlassian (TEAM): Down 50%+ YTD. A company that basically built developer collaboration. Jira is in virtually every engineering org on the planet. HubSpot (HUBS): Down 50%+ over the past year. Wrote the book on inbound. Grew to $2B+ ARR on the back of one of the best go-to-market motions in B2B history. Workday (WDAY): Down 30%+ in Q1 2026 alone. Owns HCM at the Fortune 500. Deeply embedded workflows. Salesforce (CRM): Down 30%+ in Q1. The defining CRM platform of the last 20 years. ServiceNow (NOW): Down 30%+ in Q1. Despite actually accelerating — RPO growth outpacing revenue, Agentforce-style AI positioning arguably the most credible in enterprise software. Adobe (ADBE): Down from $638 to under $350. The creative cloud monopoly. Every designer, every marketer. UiPath (PATH): Down dramatically from its highs. The RPA leader that was supposed to be the automation winner.

These are not speculative bets. These are cash-generating, deeply embedded, category-defining businesses. And the market is treating them like they face existential risk.

The Bifurcation Within the Chaos

Not everything is down equally. Within the wreckage, a clear pattern has emerged.

The companies holding their own — or in some cases still growing — share a few characteristics:

Palantir (PLTR): +135% in 2025, now cooling with the broader market but still dramatically outperforming peers on a multi-year basis. Their Rule of 40 score hit 127 in Q4 2025. Revenue growth at 70% YoY, U.S. commercial up 137%. They guided 2026 at 61% growth as a floor. The market is doing them no favors right now, but the underlying business has definitively proven it is not just another seat-based SaaS vendor.

Cloudflare (NET): Guided 2026 at $2.79B revenue, 28-29% growth. Their infrastructure sits behind 20%+ of the web. AI agents are generating an order of magnitude more outbound requests than traditional user-driven apps, and all of that flows through Cloudflare. This is what “becoming infrastructure for AI” looks like.

DigitalOcean (DOCN): Up nearly 50% YTD in an index down 50%. Smaller companies, simpler stack, and apparently a product that developers building AI apps actually want.

CrowdStrike (CRWD) and Zscaler (ZS): Cybersecurity is non-discretionary. You can be in budget freeze and still have to fund your security stack. These companies are down from 2025 peaks but have held far better than application software.

The winners are infrastructure plays — companies where AI workloads create demand rather than substitute for their product. The losers are horizontal workflow software, productivity tools, anything where the core value proposition is “help humans do their job more efficiently” and the obvious question is: what if agents do that job instead?

What This Means If You’re Building

The public markets are a leading indicator, not just a scoreboard.

The repricing happening at the public level is coming to the private market. It already is. The Redpoint data shows median NTM revenue multiples for public SaaS now at 4.1x, down from 22x at the 2021 peak. Private markets at Series B/C are still pricing at 60x ARR for the highest-growth AI-native companies — a 528% premium over public software. That gap will narrow. Some of it will close because public software recovers. More of it will close because private valuations come in as the AI-native winners separate from the pretenders.

Three things to take from this if you’re a founder right now:

1. Seat-based growth is over as a primary growth thesis. If your business plan says “we’ll grow as customer headcount grows,” that assumption is broken. The CFOs know it. Investors know it. You need to be able to articulate a growth vector that doesn’t depend on companies hiring more humans.

2. Infrastructure beats application. The companies holding up are the ones where AI runs on top of their product, not the ones where AI replaces their product. If you sit in the workflow layer and you don’t have proprietary data or years of embedded integration, your category is in play.

3. The budget is moving, not disappearing. $40-50B in new annual enterprise spend has materialized in two years flowing to Anthropic and OpenAI. That money came from somewhere — traditional B2B software. But it didn’t evaporate. It went to AI infrastructure and foundation models. Which means if you’re building AI-native, the budget exists. CIOs have already started authorizing it. The question is whether your product captures a piece of that shifted spend.

A Buying Opportunity Still? Or a Lost Generation?

Here’s what I keep asking: is this a buying opportunity or a fundamental re-rating?

In February 2016, LinkedIn dropped 44% and Tableau dropped 50% in a single day. Salesforce fell 13%. It felt like the end of the world. Four months later, Microsoft bought