We follow 81 publicly traded SaaS companies, of which 35 have had their Q4 earnings calls. We went through those calls and pulled all the quotes and comments regarding AI to get a real feel from software companies themselves on AI’s impact (note we will update this when all 81 co’s have had their calls). Below are the key themes, and yes we absolutely dumped those quotes into Claude for this analysis.
Software incumbents are big beneficiaries AI. Across sectors ranging from project management to healthcare revenue cycles to e-commerce, executives are no longer talking about what AI might do. They’re reporting what it already did, measured in dollars, users, and automation rates that would have seemed implausible just a few years ago. The scale of adoption is striking. Atlassian reported that its AI assistant, Robo, surpassed 5 million monthly active users, and found that customers using AI code generation tools were creating more tasks in Jira, growing their user bases faster, and expanding seats at a noticeably higher rate than those who weren’t using AI tools at all. ServiceNow surpassed $600 million in annual contract value for its Now Assist AI products, with deals including five or more Now Assist products increasing by more than tenfold year over year. Paylocity saw average monthly usage of its AI assistant more than double in a single quarter. Enterprise buyers are not just purchasing AI features built by software companies, rather they’re leaning into them heavily.
Data is a differentiator. Data has emerged as the decisive competitive asset. Company after company made the same fundamental argument: their proprietary data is what makes their AI valuable, and that advantage is nearly impossible for a new entrant to replicate. ZoomInfo was explicit about this, noting that while AI development tools have lowered the cost of building software, they don’t erode advantage at the data layer. SPS Commerce echoed the sentiment, with customers telling them that data is absolutely key to their AI initiatives. Blackline described its unified trusted data ecosystem as the essential fuel for its agentic AI capabilities. Overall, the moat isn’t the model, it’s the data the model is grounded in. As Microsoft put it simply: for agents to be effective, they need to be grounded in enterprise data and knowledge.
AI is becoming more agentic. Systems that don’t just respond to prompts but take autonomous action dominated the conversation across nearly every company. This marks a meaningful maturation beyond the chatbot wave of 2023. Qualys introduced an AI agent called Val that doesn’t just assess vulnerabilities but autonomously runs safe exploits to confirm whether a vulnerability is actually exploitable in a customer’s environment. Bill.com deployed an AI agent that autonomously executes card payments based on supplier preferences and can fully code complex invoices, reducing the steps required by 90%. Blackline launched Vera as a supervisor over an entire agentic workforce executing high-value finance tasks. ServiceNow went so far as to build kill switches to shut down any agent that goes rogue.
The impact is quantifiable. Waystar, which operates in healthcare revenue cycle management, reported that its AI prevented more than $15 billion in denials for clients in 2025, reduced appeal time by 90%, and drove double-digit increases in denial overturn rates. Approximately 50% of Waystar’s solutions now leverage AI, and nearly 40% of its revenue is driven by AI embedded in mission-critical workflows. The company’s framing was notably direct: “Our AI is not experimental. It’s embedded, monetized, and delivering measurable outcomes inside mission-critical workflows our clients rely on every day.” Similarly, Bill.com’s AI-powered fraud detection stopped 5.3 million fraudulent attempts in a single half-year period while reducing manual fraud reviews by 40%.
Some are not keeping up though. Not every company is at the same stage of the journey, and the candid executives acknowledged it. Doximity noted it had not yet commercialized its AI tools outside of hospitals and had not included any AI revenue upside in its current guidance, even as it celebrated becoming one of the most widely used AI tools among physicians within a single quarter of acquiring Pathway.ai. Appfolio cited survey data showing that more than half of AI users report they cannot rely on AI features embedded in their core property management systems, framing that trust gap as an opportunity.
Other responses to AI are varied. Some companies, like Wix, are leaning hard into democratization enabling anyone to build software through natural language in what Wix called Vibe Coding, with the ambition to do for software what it once did for websites. Others, like JFrog, are focused on the infrastructure layer, building systems of record that govern AI models and artifacts across the software supply chain, including a partnership with NVIDIA and Hugging Face to serve as a secure model registry. RingCentral is reinventing itself entirely as an agentic voice AI platform, rebranding its product vision around AI solutions that operate before, during, and after every conversation. Strategic responses to AI vary.
Overall, AI no longer a feature being bolted onto existing products, as Workiva was careful to distinguish about its own approach. It is becoming the structural core of how enterprise software creates value. The companies moving fastest are those that had deep proprietary data, built trust with customers in high-stakes workflows, and are now channeling that foundation into autonomous agents that act. We continue to view software companies as one of the biggest beneficiaries of AI, as they have the dev, customers, CS teams, sales teams, leadership, and actually know the problem set to solve.
We will update this post as the rest of the SaaS companies we monitor do their earnings calls.
Thank you for your readership. See more blogs and SaaS data at blossomstreetventures.com. Email the author at sammy@blossomstreetventures.com.