You've seen the headlines. $300 billion wiped from software stocks in a single session. Salesforce down 27% year-to-date. Atlassian off 35% in a week. The SaaSpocalypse has dominated every feed for the past six weeks.

And like most narratives that move this fast, it's partly right, partly overblown, and mostly missing the point.

The signal underneath the noise

The market is finally pricing in something that's been true for a while: not all software is created equal. The companies getting hammered hardest are the ones built on seat-based models with low switching costs, the horizontal tools that sit on top of workflows rather than inside them. The "nice to have" layer.

Meanwhile, CIOs are doing exactly what you'd expect. They're cutting the tools that don't demonstrate clear ROI and doubling down on the ones already embedded in critical operations. AI spend is up over 100% year-on-year while total IT budgets are growing at barely 3-4%.

The distinction that matters now is the one between software that people experiment with and software that people depend on. Between tools that generate curiosity and tools that generate revenue.

That's not a new idea. But the SaaSpocalypse is making it impossible to ignore.

Why this is a seed-stage story

Most of the coverage has focused on public markets. But the real implications are playing out at the earliest stages of company building.

AI has compressed the time it takes to build and ship software. Smaller teams can now get to product-market fit faster, reach revenue milestones sooner, and embed themselves in enterprise workflows before incumbents even notice. The gap between startup and scale is narrowing.

At the same time, enterprises are heading toward a consolidation cliff. CIOs don't want more tools. They want fewer, better ones. They want orchestration layers that reduce complexity, not add to it. And when they consolidate, they acquire the point solutions already embedded in critical workflows.

Companies with genuine commercial traction, real recurring revenue, customers who rely on them daily, become disproportionately valuable as acquisition targets. Not because of the technology alone, but because of the proven demand.

Where we focus, and why it matters now

If you've followed Haatch for a while, you'll know our thesis has always been anchored in one place: the application layer. We back B2B SaaS companies solving real problems for large organisations, at a point where commercial traction already exists.
We don't back AI narratives. We back AI that's already earning revenue.
The infrastructure layers, the semiconductors, the cloud platforms, the foundation models, those are becoming commoditised. The value is being built at the application layer, where software meets specific, persistent enterprise problems. That's always been our bet, and the SaaSpocalypse is reinforcing it, not undermining it.

Regulated industries are where the moat gets deeper

There's another layer to this that doesn't get enough attention. The SaaSpocalypse is largely a story about horizontal software, the tools that compete on convenience rather than necessity. But in heavily regulated sectors like financial services, banking, and compliance, the dynamics are fundamentally different.

Regulation doesn't just create complexity, it creates dependency. When a company builds software that helps enterprises stay compliant with FCA requirements, or embeds working capital infrastructure directly into a bank's workflows, or automates finance operations for brands navigating complex cash flow cycles, that software doesn't get cut in a rationalisation exercise. It becomes load-bearing. The switching cost isn't just about workflow disruption, it's about regulatory risk.
This is a space we're increasingly drawn to across the portfolio.

Adclear automates marketing compliance for financial services companies operating under FCA regulations, covering everything from investments and crypto to CFDs and spread betting. They launched in 2024 and are already trusted by Lloyds Banking Group, PensionBee, Plum, and Trade Nation. At Plum, they reduced approval times from ten days to same-day and enabled an 18x increase in marketing output. Their ARR has grown 10x since we led their pre-seed, and they've since closed an oversubscribed £2.1m seed round with backing from the ClearScore founder and the MD of Coinbase. As AI-generated marketing content explodes in volume and regulators increase scrutiny, the need for what Adclear does only accelerates. We use it ourselves for our own financial promotions.

Bourn is tackling a different but equally persistent problem in banking. Their Flexible Trade Account reinvents the business overdraft for SMEs, combining the flexibility of an overdraft with the risk profile of invoice finance, all automated and embedded directly into bank workflows. It's infrastructure built with banks, not around them. They're already piloting with Investec and a host of other tier-one clients, and NatWest took a strategic minority stake in their most recent £3.5m round. When you're embedded in the working capital infrastructure of major UK banks, you don't get swapped out because a CIO watched a demo of an AI agent.

Triffin is building the finance operating layer for consumer brands, using AI agents to automate pay runs, collections, purchasing, and reporting. They hit a £2m revenue run-rate within three months of launch, with 91% agent adoption across active accounts and a CAC payback under three months. They've also partnered with Froda to embed a £100m line of credit directly into their workflows, meaning brands can access growth capital without leaving the platform. Finance automation for consumer brands sits at the intersection of compliance, cash flow management, and operational complexity, exactly the kind of deep, persistent problem that doesn't disappear when budgets tighten.
These are the kinds of companies that thrive in a correction. They solve problems that are regulated, recurring, and hard to replace. The pain doesn't go away when the market gets nervous. It gets worse.

The takeaway

The SaaSpocalypse isn't the death of software. It's a correction, and a useful one.

It's forcing the market to ask better questions. Not "is this company using AI?" but "is this company solving a problem that someone is already paying to have solved?" Not "what's the TAM?" but "what's the actual customer dependency?"

For us, the thesis hasn't changed. If anything, the current environment has made it sharper. Back companies with real revenue, embedded in real workflows, solving real problems. Let the froth clear. The fundamentals will do the rest.

The companies doing real work inside enterprises today are the ones that will be acquired tomorrow.

The rest is noise.