Software stocks took a hit in 2025. That doesn’t mean investors should rush to buy.
Generative artificial intelligence is seemingly everywhere. Software companies have had to adapt—not just to stay ahead, but simply to keep pace with competitors. Many are now offering a range of AI products in hopes of boosting productivity and reshaping how enterprises operate.
On the surface, demand for enterprise software looks strong. ServiceNow posted better-than-expected third-quarter earnings and revenue in October and raised its subscription revenue guidance. Salesforce’s third-quarter revenue just missed expectations, but earnings were above consensus and guidance was strong. Creative software company Adobe also had a solid third-quarter and provided better-than-expected financial guidance for the year ahead.
All three of those companies are working hard to push out their own AI products for customers.
Despite strong financial results, ServiceNow stock has dropped 26% this year, Salesforce has fallen 24% and Adobe has lost 22%.
Software stock declines are widespread. Workday has dropped 16% this year while Atlassian has fallen 34%, Asana has declined 30%, and Monday.com has tumbled 36%. Microsoft, Oracle, and Palantir Technologies are up 13%, 13%, and 148% this year, respectively, as their business models span beyond strictly software. But all of those stocks have recently dropped from their 2025 highs.
“I think the reason that we’re seeing so much pressure on the software sector as a whole, and especially application software, is there’s this idea that AI is ‘the death of software,’” RBC analyst Rishi Jaluria told Barron’s.
Some worry that traditional software applications could be replaced by AI-driven capabilities.
Competition is another concern. Software companies must keep raising the bar to make their products worthwhile investments for clients while fending off rivals. That risk was highlighted on Dec. 14, when a KeyBanc analyst downgraded Adobe to Underweight from Sector Weight while maintaining a $310 price target.
“It doesn’t matter who the perceived winner is at any given moment, what matters for Adobe is that extremely large and well-funded companies are developing technology at a pace that is difficult for Adobe to match,” Jackson Ader wrote in a note.
Competitive concerns are just the tip of the iceberg. Software stocks are also facing fears of an AI bubble. Companies are spending billions to invest in AI, but now Wall Street wants to see some type of return on those investments. That means investors are looking for tangible evidence that AI improves productivity and that software companies are benefiting financially from businesses buying the tech that leads to those productivity gains.
Tech valuations have also soared this year. But as software stocks come down, so have their multiples. ServiceNow is trading at 37.7 times earnings expected over the next 12 months, below its five-year average of 61.3 times. Salesforce’s forward 12-month P/E of 19.6 times is below its five-year average of 34.9 times, while Adobe’s forward 12-month P/E of 14.9 times is lower than its five-year average of 28.5 times.
“I’m not going to say the market is glass-half-empty. That’s probably not fair. But it’s taking a much more balanced view about software and how to value these companies,” Steve Sosnick, chief strategist at Interactive Brokers, told Barron’s. “And so rather than giving them the benefit of the doubt all the time and then rewarding them for very consistent growth, I think investors are starting to look for warts.”
The question for investors now is what could happen next year. Analysts at RBC tell Barron’s that it’s impossible to make blanket assumptions about how software stocks as a sector will move in 2026. But what shareholders should keep an eye on is what could help the stocks rebound.
These companies need to prove they are innovating with AI, monetizing those efforts, and delivering continued acceleration in revenue growth, the RBC analysts said.
“The leading indicator that I really want to look at is, what is the AI roadmap? What are the unique use cases driving that can’t be replicated elsewhere? And not just relying on ‘data,’ but actually having innovation,” Jaluria said.
“If you start to see real usage of that, and you start to see really positive customer feedback, that to me is a leading indicator that this will eventually translate into better revenue outcomes and then better stock performance,” Jaluria added.
Comments