Software stocks are plunging—should you patiently hold or buy the dip?
```
The software sector is undergoing a brutal sell-off—not just a fluctuation in market sentiment, but a deep reassessment of the industry’s ultimate fate.
According to Chase Wind Trading Desk, UBS pointed out in its heavyweight research report released on February 4 that investors should not rush to "catch falling knives," but rather remain patient. The core logic is: the rapid acceleration of AI technology (such as Gemini 3, Claude 4.5, and upcoming OpenAI releases) is fundamentally disrupting the traditional SaaS business model. Although the market has long anticipated AI-driven transformation, the reality now is that the pace is faster than expected, while the income growth curve for software companies has yet to rise.
For investors, this means the "terminal value" risk for SaaS and application software stocks is surging. UBS clearly states, in the short term, avoid application software companies based on "seat" pricing, as they are at the eye of the AI disruption storm. If investors still want to seek opportunities in tech, UBS recommends switching focus to infrastructure, data, and cybersecurity (e.g., Microsoft, Snowflake, Datadog, Okta). While these areas have also fallen with the broader market, their customer spending trends are healthier and are less exposed to direct AI substitution. Simply put, the current strategy is: wait and see for application software, selectively absorb infrastructure.
The Illusion of Growth Shattered: Not Only AI Fears, But Also Cyclical Downturn
The market tends to blame the plunge in software stocks entirely on AI disruption fears, but this overshadows harsher fundamental truths. UBS emphasizes, even before investors debated who the AI winners are, the growth engine of the application software industry had already stalled. This is no longer an era of “growth at any cost,” but a mature, saturated market.
Data doesn’t lie. UBS statistics show, large SaaS and application software companies have ended their pre-pandemic era of 15-20% high growth; current average organic revenue growth has dropped to about 12-13%. Salesforce, once a star at stable 20% growth, has now fallen to 9%, with no bottom seen; Adobe fell from 21% to 10%. When an industry's growth rate plunges from high to mediocre, a valuation multiple compression is the inevitable physical reaction. Now, Fortune 500 CIOs are no longer keen to expand SaaS coverage but are focused on cutting software spending to free up budget for expensive AI infrastructure and data modernization. Therefore, unless we see a renewed upward growth curve, buying just because “the stock price looks cheap relative to history” doesn’t hold water.
Valuation Trap: Profitability Still Weak From GAAP Perspective
If you think software stocks are already cheap, you may be looking at the wrong metric. UBS sharply points out, most software stocks are still valued using "non-GAAP" PE ratios (median around 23x), masking the reality that huge stock-based compensation (SBC) has beautified the income statements.
Switching to the stricter GAAP perspective, many companies’ profitability shrinks instantly. Aside from Salesforce successfully transforming from high growth to high profit (expected GAAP EPS to reach $7.25 for FY2026), most companies in the industry still have high non-cash stock incentives (over 15% of revenue). In this situation, investors cannot even reasonably use non-GAAP multiples for evaluation. What's more worrying, as stock prices halve, companies might be forced to reprice equity or issue more cash to retain talent, further eroding shareholder value. The reason PE giants remain inactive despite large declines is that many SaaS companies have such high stock-based compensation costs that they are “not financially acquirable.”
The Truth About AI Monetization: Lots of Talk, Little Action; Income Share is Tiny
AI hype has been noisy for three years, but there's little to show for it in software companies' earnings. UBS statistics: The total disclosed AI revenue of public application software companies is only $5.6 billion. Excluding Microsoft Copilot and GitHub Copilot (about $3.8 billion), the entire industry's AI revenue drops to roughly $2.0 billion.
Given the top 10 SaaS companies’ total revenue base of $290 billion, $5.6 billion in AI revenue is only around 2%—a drop in the bucket. This explains why AI buzz is deafening but software companies’ growth curves remain untouched. In contrast, hardware and semiconductor firms derive tangible revenue from the AI boom. More grimly, incremental AI spending doesn't fully go to established SaaS giants, but gets split between model providers like OpenAI, Anthropic, and AI-native startups like Sierra and Glean.
UBS data shows, these AI-native companies' total revenue already matches the AI revenues of public software companies. This is a zero-sum game—traditional software companies' moats are being eroded by new players and the trend of enterprises "DIY" AI applications.
Seat Compression Risk & The Battle of Enterprise Spending
The most direct threat AI poses to the SaaS model is “seat compression.” UBS surveys of Fortune 500 firms (including hotels, manufacturing, tech, retail, etc.) reveal that companies are indeed using AI to reduce the number of human customer service and low-skilled employees, meaning SaaS seats billed per user will drop sharply.
For example, a large hotel group using Salesforce said its human seats would drop 10% in 2025 and another 30% in 2026. Yet the flip side is, companies must pay extra fees to deploy AI agents (such as Agentforce). Several interviewed companies noted seat numbers fell, but AI functionality spending rose, resulting in a net overall payment to software vendors going up (net increase possibly high single to low double digits). This may be the only positive signal for now: AI may not kill existing software giants, but will force them through painful transformation—from selling "head count" to selling "AI output."
UBS Advice: Avoid the Eye of the Storm, Seek Safe Harbors
In the face of such high uncertainty, UBS’s ultimate advice is: do not try to find winners in the "too difficult" basket of SaaS application software, at least not now. Only when revenue growth accelerates, earnings forecasts rise, or SaaS firms prove they can coexist with AI giants, will it be time to enter.
If you must allocate to software stocks at current levels, UBS is clear about three preferred directions:
- Infrastructure & Data (top pick): Microsoft, Snowflake, Datadog. These are the “pick-and-shovel” players in the AI rush, with healthy customer spending and relatively low disruption risk.
- Cybersecurity: Okta, Zscaler. Compared with the highly valued CrowdStrike or Palo Alto Networks, these two have more attractive valuations, and demand remains solid amid AI-driven security threats.
- Non-seat pricing & niche markets: Focus on companies utilizing usage-based pricing (Twilio, Braze), or verticals migrating to cloud like Autodesk; these are at the edges of the AI disruption debate, relatively safe.
~~~~~~~~~~~~~~~~~~~~~~~~
The above excellent content comes from Chase Wind Trading Desk.
For more in-depth interpretation, including real-time analysis and frontline research, please join [Chase Wind Trading Desk · Annual Membership]
Risk Warning and DisclaimerThe market has risks; investment needs caution. This article does not constitute personal investment advice and does not consider the unique investment goals, financial situation, or needs of individual users. Users should consider whether any opinion, viewpoint, or conclusion in this article suits their specific circumstances. Invest accordingly, at your own risk. ```