We just got the 17th edition of KeyBanc’s private software benchmark. For those who don’t follow the venture debt or equity research world closely, this report is usually the sober morning-after pill for SaaS hype. This year, they leaned heavy into the AI intersection, and the numbers are a reality check for anyone still thinking we are in the 2021 bull market.
The Rule of 40 is Dead (For Now)
The first thing that jumps out is the sheer lack of efficiency in the current market. Only 6% of the companies surveyed are hitting the Rule of 40. For the uninitiated, that’s the golden metric where your growth rate plus your profit margin equals 40 or more. If only 6% are hitting it, the industry isn’t just slowing down; it’s struggling to find a sustainable identity.
We are seeing an average growth rate of about 24%. In the old world, that would be respectable for a mid-market firm, but in the context of the massive capital being poured into AI features, it feels sluggish. It tells me that builders are spending a lot of money to stay in the same place. We are running on a treadmill that keeps getting faster, but our actual forward progress is minimal.
The Rise of Vibe Coding
There is a specific stat in the KeyBanc report that should keep every B2B founder awake at night: 43% of vendors say they are watching their own customers "vibe code" replacements for their products. This is the first time we’ve seen this data quantified at scale.
Vibe coding is what happens when a non-technical manager uses an LLM to spin up a custom internal tool that does 80% of what a specialized SaaS product does. It might not have the polished UI, and it definitely doesn't have the enterprise security features, but it's free, it's custom, and it’s "good enough." When nearly half of software vendors admit their customers are experimenting with building their own alternatives, the moat is officially under siege.
The Cost of Being an AI Company
Builders often talk about AI as a margin-expander. The theory is that you automate support, speed up engineering, and magically become more profitable. The survey suggests the opposite is happening in the short term. AI costs are eating into margins faster than the efficiency gains can save them.
We are seeing a shift in where the money goes. It’s no longer just about headcount; it’s about compute, API credits, and the specialized talent needed to make these models actually work in a production environment. Most companies are still in the "investment phase," which is a polite way of saying they are burning cash to see if an AI feature can actually drive a price increase.
Pricing Power or Pricing Pressure?
The report touches on the shift in pricing models. For a decade, per-seat pricing was the law of the land. AI breaks that. If an agent does the work of five people, you can't charge for five seats. You have to charge for the value or the outcome.
However, the survey shows that vendors are struggling to make this transition. Customers are used to predictable monthly bills. When you move to usage-based or outcome-based pricing, the sales cycle gets longer and the CFOs get more nervous. We are in a messy middle ground where the old way of charging is dying, but the new way hasn't been perfected yet.
Internal Adoption vs. External Value
One of the more interesting takeaways is how these companies are using AI internally. Almost everyone is using Copilot for dev work and some form of LLM for marketing copy. But there is a massive gap between using AI to write a blog post and using AI to fundamentally change the unit economics of the business.
For builders, the lesson here is clear: stop focusing on the "wrapper" features. If your AI strategy is just a chatbot in the corner of your dashboard, you are part of the 94% missing the Rule of 40. The companies that will survive this transition are those using AI to do things that were previously impossible, not just things that were previously slow.
What This Means for Builders
If you are building in the crypto or AI space right now, you have to look at this data as a roadmap for what to avoid. The "SaaS-ification" of everything is hitting a wall. Here is how I see the path forward:
- Focus on the Moat: If a customer can vibe-code a replacement for your tool in a weekend, you don't have a product; you have a feature. You need to lean into deep integrations, proprietary data, or workflows that are too complex for a prompt to handle.
- Watch the Margins: Don't assume scale will fix your compute costs. If you aren't profitable at a small scale with your AI features, you likely won't be at a large scale either.
- Solve for the CFO: In this economy, software is being cut ruthlessly. Your tool needs to be a "must-have" that either directly generates revenue or replaces a much more expensive legacy cost.
The era of growth-at-all-costs is over. The era of 'good enough' custom-built software is here. Builders who don't respect that shift will be the ones wondering where their churn is coming from in 2026.
The Takeaway
The KeyBanc report is a warning shot. The software industry is facing a dual threat: a macro-economic slowdown and a technological shift that empowers the customer to become the competitor. 24% growth sounds fine until you realize the cost of achieving it is higher than ever.
We are moving toward a world of "thin" software. The winners won't be the ones with the most features; they will be the ones who provide the most utility with the least amount of friction. If you can't beat the vibe coders, you might need to build the tools that help them vibe code better. Either way, the old SaaS playbook is ready for the shredder.
Read the original at SaaStr →