Jamie WestonInvestors Are Done Betting on Software and That Should Worry You
March 19, 2026
I got in at 7:30 this morning. My dad was already here. I still don't understand how he does that.
I'd been up late reading about the software funding situation - not for any particular reason, nobody asked - and by the time I sat down with my coffee I had maybe four browser tabs I actually needed and about thirty I'd opened because one thing kept leading to another. The headline I started with was about investors pulling back from software bets. By the time I was done, I was looking at a structural shift that I think most business owners are going to feel before they understand what hit them.
Here's what's actually happening, and here's why I think the calm, measured takes on this are wrong.
The Numbers Aren't Being Read Right
On the surface, 2024 and 2025 look fine for venture capital. Investors poured $425 billion into more than 24,000 private companies in 2025, up 30% year over year from $328 billion in 2024. People see that number and relax. They shouldn't.
Because when you crack the headline number open, what you find is that the money isn't going to software - it's going to AI, and those are not the same thing anymore. OpenAI, Scale AI, Anthropic, Project Prometheus, and xAI each raised more than $5 billion in 2025. Those five companies alone took $84 billion - or 20% of all venture capital funding that year. That's not a rising tide lifting all boats. That's a tsunami crashing into one specific beach while the rest of the coastline dries up.
The percentage of overall funding going into deals of $100 million or more hit a record high in 2025, with 60% of global and 70% of U.S. venture capital going to those mega-rounds. If you are not an AI foundation model or directly adjacent to one, you are being squeezed out of a market that used to have room for you.
Meanwhile, the software category that investors spent a decade treating like a sure thing? It's getting a very different reception.
The SaaS Hangover Is Real and It's Not Going Away
The software-as-a-service sector, which had experienced a vibrant 21% annual revenue growth just a few years ago, saw that metric sink to just 12% in 2024. And that was before things got worse. The first quarter of 2025 was even tougher, with sector-wide SaaS revenue growth at negative 2%, indicating that the slowdown is not temporary but structural.
Negative. That's not a correction. That's a category reconsidering itself.
The valuation story is just as stark. According to the Bessemer Venture Partners Cloud Index, the median revenue multiple of a publicly traded SaaS company sat at 7.5x in early 2025, down from a peak of 18.43x in September 2021. And it's continued sliding. At the beginning of 2025, the median revenue multiple for software firms still stood above 7. By early 2026, it had dropped below 5.
I told Derek about this the other day. He was not concerned. Derek is currently very focused on a fan theory about the Mandalorian timeline that I don't fully understand. But the valuation collapse is not a fan theory. It's a fact with receipts.
By early February 2026, the IGV ETF, which tracks North American software, was down almost 20% year to date and close to 30% off its September peak. The sector's forward P/E had collapsed from roughly 35x at the end of 2025 to around 20x, back to levels the market hadn't really seen since 2014. That's twelve years of multiple expansion gone in months.
And the stock market is often six months ahead of reality on the ground. Which means the businesses that buy software - that's you, reading this - are about to feel it too.
The 150 Companies Nobody Is Talking About
This is the part that I think is genuinely underreported, and it has direct consequences for anyone whose business runs on software tools built by venture-backed companies.
Per Crunchbase data, more than 150 boom-era U.S. software and software-related companies with $100 million or more in equity funding have not raised capital in over four years, remain private, and have not been acquired. Collectively, they pulled in over $51 billion in aggregate funding.
Think about what that means. These aren't tiny startups running on a shoestring. They raised enormous sums at peak valuations between 2020 and early 2022. During that boom, software companies in particular routinely raised megarounds at rich valuations. Then the market turned. By mid-2024, Crunchbase recorded only 21 SaaS funding rounds exceeding $100 million in the prior 12 months - a significant drop from 147 mega-deals in 2021.
If a venture-backed company has gone more than four years between funding rounds, the forecast generally looks dim. It becomes increasingly unlikely that it will secure another good-sized financing or a sizable exit.
Some of those 150+ companies are names you've heard of. Carta, the equity and fund management software platform, raised close to $1.2 billion in total funding but hadn't reported a new round since 2021. Calendly secured $350 million in 2021 and hasn't raised a round since. These are tools companies actually use. Tools people depend on. And their future is quietly uncertain.
If you're running a business that relies on vendor-backed software - and you almost certainly are - this is the part where you should pay attention. We've written before about what happens when the platform you've built on changes ownership or loses its footing. It can feel very abstract until it's suddenly very personal.
The Concentration Problem Is the Real Story
The reason I keep coming back to this isn't just the individual companies going quiet. It's what the whole pattern signals about where investment is flowing and - more importantly - where it isn't.
Most VC fundraisers slogged through 2024. They raised $76.1 billion, making it the lowest fundraising year since 2019. Established firms continued to attract the majority of capital, as LPs increasingly chose a "flight to quality." The top 30 funds secured 75% of the year's total.
What that "flight to quality" actually means in practice: money flows to the biggest names chasing the biggest AI bets, and everything else - including the thousands of mid-tier SaaS companies that quietly power B2B operations - gets starved of oxygen.
The biggest shift in SaaS investment strategy came from growth-stage and crossover investors. Firms such as Tiger Global and Coatue, which had aggressively invested in SaaS during the peak years of 2021-22, significantly slowed their SaaS funding activity. These weren't small players. They were the engine of the SaaS boom. And they've moved on.
Linda mentioned to me this week that her husband Gerald just switched accounting software because their old vendor "just kind of stopped improving." She said it matter-of-factly, the way she says everything. But I've been thinking about it. That's what starved investment looks like at ground level. The product just... stops getting better. The support gets slower. The roadmap goes quiet. And one day you're looking for alternatives.
AI Isn't Just Stealing the Investment - It's Eating the Revenue
Here's the thing that makes this more than just a funding story. Investors aren't just choosing AI over traditional software because AI is shiny. They're doing it because AI is beginning to replace what traditional software did.
In 2024 and the first half of 2025, the dominant story was that AI would augment existing enterprises. Every software company talked about embedding AI and boosting user productivity. By late 2025 and early 2026, that story quietly flipped. The reality on the ground was that the pace at which incumbents were adding AI lagged the pace at which native AI tools were beginning to replace them.
That's not hype. While Wall Street closed 2025 as one of the best years in its history, led by the Magnificent Seven, the software industry lagged far behind. The SaaS index, representing companies that sell subscription-based software to enterprises, fell 6.5%, compared with a 17.6% rise in the S&P 500.
We've been going back and forth on this one internally. I wrote a section about it four times. The version you're reading is the one where I stopped trying to be balanced and just said what I think: the market is pricing in a scenario where a meaningful percentage of seat-based software subscriptions become unnecessary within the next three to five years. Companies with usage-based or data and infrastructure-linked models are better positioned than those reliant on per-seat subscriptions, which face direct pressure as AI agents reduce human headcount. If you're paying per seat for a tool that a single AI agent could theoretically replace, investors are already pricing in the risk that you'll eventually notice.
We asked vendors directly about this not long ago. The answers were genuinely wild. Nobody wanted to say yes, but some of them basically said yes.
What This Means If You Buy Software for a Living
I want to be direct about who this matters to, because I think a lot of people will read this as a VC industry story and tune out. It isn't. This is a vendor stability story.
When a software company's investor base dries up and its revenue growth drops to negative territory, a few things happen in sequence. First, hiring freezes. Then product development slows. Then support quality erodes. Then the company either quietly winds down, gets acqui-hired for parts, or gets sold to a private equity firm that immediately cuts costs and raises prices. None of those outcomes are great for the customer.
Deal slippage is now common as clients delay approvals or scale back project scopes. Sales teams are reporting deals stuck in late-stage review, purchases postponed until budget clarity improves, reduced contract values, and lengthened evaluation cycles. This is the buyer-side response to an uncertain market. But it's also a feedback loop - when enterprise customers slow their software spending, the vendors serving them slow their investment in product, which makes the tools worse, which makes customers even more reluctant to expand their spend.
Tory - our optimism-is-everything guy - told me yesterday that all downturns create opportunity. He's not wrong. He's also living in his sister's guest room right now, but his optimism is structurally sound. The opportunity here is real: this is an excellent time to audit which of your software vendors are actually in a solid financial position versus which ones are quietly on the clock. Vendor risk is a real thing. We used to call it concentration risk when it was about suppliers. The software version of this is the same concept with a subscription model attached.
This is also where I'll just say plainly: tools that generate real, defensible revenue data and have strong network effects are probably fine. CRM platforms with huge install bases and years of customer data aren't going anywhere tomorrow. The more vulnerable vendors are the mid-tier niche tools that raised big in 2021 and have been living off that cash ever since. Some of them are already on borrowed time.
We wrote about the software stock crash and the way nobody wanted to say the word "contagion" out loud. That piece still holds up. The private market version of the same story is just quieter and slower, which makes it easier to ignore until it's too late.
My Take, Plainly
The headline says investors are done betting on software. That's a little dramatic - they're not done, they're just done betting on the kind of software that defined the last decade. Horizontal SaaS with per-seat pricing and a TAM slide that assumes you'll eventually sell to every company on earth? That era is closing. Deal counts declined about 16% year-over-year in 2025, with just under 10,500 reported rounds, as more capital concentrated in larger rounds. Fewer bets, bigger bets, almost entirely on AI.
The companies getting funded now are either building AI infrastructure or are so deeply embedded in enterprise workflows that switching costs make them effectively permanent. Everything in the middle is competing for scraps while trying to pretend it isn't.
If I'm running a business, what I'm doing right now is making a list. Every software vendor I depend on. Checking when they last raised money. Checking their revenue trajectory where I can find it. Thinking about what happens to my operations if two or three of those vendors quietly stop investing in their products or get sold to someone whose priorities are different from mine.
That's not paranoia. That's vendor management in 2025. The market is telling you something. The smart move is to listen before the product roadmap email stops arriving.
My dad read a draft of something else I wrote last month and said "you should keep going." I'm not sure that means what I want it to mean. But I keep going anyway.