swyx gave simple numbers for the results of his Kill My SaaS competition. He put the price of a piece of software his team was about to buy at $40,000 a year. His original brief for the bounty said the team was proposing to pay more than $40,000 a year for enterprise software it had never used. Replacing it cost about $1,000. He said the replacement is going into production for his next conference.
swyx hosts the Latent Space podcast, and his company runs the AI Engineer conferences. He raised the example at AI Engineer New York while talking with Alex Lieberman and CJ Hess of Tenex, a firm that helps large companies adopt AI. The conversation appeared in a Latent Space episode published October 21, 2026. swyx asked Lieberman to tell him where the plan was going wrong. Lieberman didn't say it was wrong. Instead he offered a test for when building your own replacement makes sense, and said most large enterprises are unlikely to try it.
The bounty and its result
SaaS stands for software as a service: a program that a vendor runs online and rents out by subscription. Kill My SaaS was swyx's bounty for a replacement of event software his team had been proposing to pay for, with a $10,000 prize. In an earlier interview on The Cognitive Revolution, he described the bounty, the work of judging the entries and how his skeptical events team came around.
On the show, swyx called the tool "mid-tier": "probably not that trivial, but it's not Salesforce. It's somewhere in the middle." He said there is a lot of software like that, and it is what people mean when they talk about "the death of software." According to swyx, his team had become the kind of customer that wants a tool made just for it, and it was building something specific to its own business. "So far, it's going okay," he said.
Then he joked about how many story points the project was worth. Tenex pays its engineers for story points, which are agreed estimates of how much work a task involves, rather than by the hour. Lieberman passed the question along: "CJ can size up the story points on it."
Why Lieberman thinks Salesforce and SAP are safe
Lieberman's view is that the further up the market you go, the harder it gets to rip out software and build a replacement from scratch. He mentioned "a very small group" who say Salesforce and SAP are "really at risk" and said he doesn't "agree with that at all." Salesforce sells customer-relationship software, and SAP sells systems that run companies' finance and operations. In his view, both businesses "accrue value in this moment in time."
His reason is the cost of failure. If a large company gets it wrong when it tears out one of these systems, he said, "you're talking about potentially hundreds of millions or billions of dollars." The conversation noted that buyers facing those stakes will be risk-averse, and Lieberman agreed.
The size of the prize
For mid-tier tools like the one swyx replaced, Lieberman said the decision comes down to the size of the opportunity. He imagined an enterprise paying a few million dollars a year for one of them. The first question, he said, is whether that spending, as a share of revenue or costs, would "ever move the needle enough" to justify putting engineers on it. He expects the answer to depend on the company, with some enterprises deciding yes and others no.
He drew on Tenex's own experience. The company built its own CRM (customer-relationship management software) from scratch, and its own internal operating system too. Lieberman said Tenex learned that part of software's value is "not just the utility that you get from it." It is also "all of the headaches you don't have to think about in maintaining it and improving it when it's not in your core."
He put the test this way: "is the size of the prize big enough to justify the cognitive load" of having people in your company look after something that is not core to the business? Lieberman said that for swyx's company, the difference of "basically $39,000 a year" means swyx is implicitly answering yes. That figure is the gap between the roughly $40,000 annual price and the roughly $1,000 spent on the replacement.
Defense and offense
swyx said the decision was "not just about costs." He called the savings defense. The offense is that "our experience could be a lot better for literally every party concerned," because his team is its own customer and can iterate "with, like, minutes and not months."
Lieberman agreed and split the case in two. One part is being overcharged. The other is that "the product has been mediocre for a long period of time." Put those together, he said, and the question becomes: "What would the 10X better version of this product look like?"
Who can move this fast
Lieberman thinks only some companies can act on that question. In his view, startups and "high-growth, mid-sized companies" can be nimble enough. He said large enterprises are "so entrenched in a certain capability" that they will rarely decide to build a 10x better version of a SaaS tool they use, let alone spin it out as a product of its own. "I just don't think enterprises function that way," he said.