Software companies were, until recently, all the rage on Wall Street and Silicon Valley. More than a decade ago, web browser Netscape’s creator Marc Andreessen, now a prominent venture capitalist, penned a piece titled Why Software is Eating the World arguing a new golden age was dawning for the sector.
“We are in the middle of a dramatic and broad technological and economic shift in which software companies are poised to take over large swathes of the economy,” he wrote. “More and more major businesses and industries are being run on software and delivered as online services — from movies to agriculture to national defence.” Over the next 10 years, he argued, “many more industries (will) be disrupted by software, with new world-beating Silicon Valley companies doing the disruption in more cases than not.”
If you had invested in the benchmark iShares Expanded Tech-Software Sector ETF when Andreessen wrote his seminal piece in August 2011, you would have over 655% return. The stock of software giant Microsoft, the world’s most valuable company, is up 2,230% since then. The key to Microsoft’s resurrection after its lost decade in the aftermath of the tech bubble burst in March 2000 was its pivot to the cloud. Companies like Microsoft, Salesforce, Adobe, ServiceNow, Palo Alto and their ilk began providing their software as a service (SaaS) or a solution that enterprises such as JPMorgan Chase, Walmart, Exxon Mobil and others could purchase on a pay-as-you-go basis from a cloud service provider like Amazon Web Service, Microsoft’s own Azure or Google Cloud.
Software is ubiquitous these days. Before the advent of smartphones, cellular phones were just a chunk of hardware with a keyboard for dialling. Software transformed a clunky cellular phone into a powerful Internet-enabled computer in your pocket. These days, software runs cars that were once just chunks of metal. “Software runs engines, controls safety features, entertains passengers, guides drivers to destinations and connects each car to mobile, satellite and GPS networks,” Andreessen adds. “The days when a car aficionado could repair his or her car are long past, due primarily to the high software content.”
Wal-Mart, the world’s largest retailer, uses software to power its logistics and distribution capabilities, which it has used to crush its competition. Indeed, FedEx is a software network that just happens to have trucks, planes and distribution hubs attached. Andreessen notes that even agriculture is increasingly powered by software, including satellite analysis of soils linked to per-acre seed selection software algorithms.
So why is software suddenly no longer in vogue? Software sector stocks were on a tear during the Covid-19 lockdown in 2020 and 2021, when they surged 72%. In 2022, when the US Federal Reserve raced to raise interest rates to fight runaway inflation, software stocks, one of the most overvalued sectors within tech, took a hammering, with the benchmark software index down 35.6%. Some of the software names were down nearly 70% in 2022. Last year, with the advent of ChatGPT and generative artificial intelligence or Gen AI, the software sector rebounded with a vengeance, with firms that had some exposure to AI seeing their stocks soar again. The software index jumped 54% in 2023.
See also: Australia’s social media ban for under 16s to become law
Just as a new AI-powered bull market gets underway on Wall Street, the software sector is seen as a laggard. Even software startups are reeling as venture funds have dried up. AI is sucking all the oxygen in the room, leaving little for once high-flying sectors like software. Investors are taking money out of software names and piling on AI names like Nvidia, AMD, Super Micro Computer, search behemoth Alphabet, social media supremo Meta Platform, e-commerce pioneer Amazon and iPhone maker Apple. About 60% of software stocks are down this year. The software index is up just 1.1% this year compared to a 13% gain for the broader S&P 500 index and a 15.4% upside for the tech-heavy Nasdaq100, which includes the biggest tech stocks.
Blaming AI for software’s winter?
So, what’s going on with software stocks that are being touted as AI plays? A sell-off in enterprise software stocks in recent months has dragged down names like Salesforce Inc to SAP, Adobe, UIPath, ServiceNow, Intuit and Cadence. Salesforce is down 8% this year and 25% from its March peak because it missed on earnings two weeks ago. Adobe is down 22% this year, UI Path has been down 50% since January, ServiceNow has been down 14% since its February peak, and Intuit has been down 15% since its February peak. Only the top software player Microsoft, which is in the process of reinventing itself as an AI powerhouse, has fared better. Its stock is up 14% this year.
Ironically, companies like ServiceNow and Adobe were until recently being touted as AI plays. Investors who missed the boat on Nvidia hoped they could always hitch a ride on software firms leveraging AI. Unfortunately, even after its massive run over the past two weeks, Nvidia stock trades at 40 times forward earnings while ServiceNow trades at 51 times next 12 months earnings. Even Microsoft is only slightly cheaper than Nvidia at 34 times forward earnings.
See also: UBS-Credit Suisse integration opens up new tech for bigger plans
During their earnings calls over the past two weeks, CFOs and CEOs of Salesforce, Adobe, UIPath, and Intuit stepped out to explain that their customers had suddenly become a little more reluctant about spending more money on enterprise software. Corporate budgets around the world are being rejiggered to give priority to AI-related spending. Every company wants its own AI strategy. But instead of expanding their overall budgets, they are just redirecting more of their software spending towards AI. Even as their AI spending grows, investors are raising questions about the effectiveness of AI-powered tools like Copilots or other Generative AI offerings that enterprises are racing to adopt to keep ahead of their competitors. It is unclear whether software firms will suffer from a switch to rival AI-powered services or will be better off because they adopt AI in many of their existing products.
As the software stocks began reeling a few weeks ago, Chris Paik, a General Partner at New York-based venture capital firm Pace Capital, published an essay titled The End of Software arguing that in the age of AI, software’s run was over. He likened software to old pre-Internet media. “Before the internet, media behaved very differently — it was expensive to create,” Paik adds. “You had to pay people to make content, edit it, and distribute it. Because content was expensive to create, it had to make money. And consumers paid (for) newspapers, magazines, books, cable, and pay-per-view (TV). (Billionaire investor) Warren Buffett famously loved newspapers. Who wouldn’t love a predictable subscription business with local monopolistic dynamics?”
Then, in the late 1990s, the internet arrived. Media firms saw it as a way to reach broader audiences and drastically slash their distribution costs. “Internet not only reduced distribution costs to zero, but it also drove the cost of creating content to zero,” Paik notes. User-generated content on Instagram, YouTube, TikTok, Snap, Pinterest and Twitter (now X) flourished. “When content doesn’t cost anything to create, it no longer has to make money,” he argues. But what happens when content no longer has to make money? “The relaxation of this economic constraint (has) led to a Cambrian explosion,” he says. “You can take a picture of a cup of coffee, post it to a million views or none at all, and the market clearing price is still met. That has produced a deluge of content that none of us can reasonably consume,” he adds.
Little wonder, then, that user-generated content platforms like TikTok, YouTube and Instagram have pushed products to users’ direct attention and routed them effectively for maximum impact. Media firms were competing for the same attention of users, but with a strictly higher cost of goods sold, Paik argues.“The more people you had on your payroll creating content, the more exposed you were to being flanked by user-generated content platforms. What the internet did was it pushed the marginal cost of media distribution to zero. Structurally, investing in media has been a losing value proposition ever since and value creation has shifted entirely to the platforms that control distribution.”
Video killed the radio star. Then, the internet killed videos and movie theatres and decimated newspapers and magazines. Now, it’s a Netflix world, and all the other media companies live in it. There is also YouTube, TikTok and Instagram, which mostly rely on user-generated content. Will AI kill software? Or dramatically change the face of the software industry?
Software is not like a newspaper. Believe me, as a journalist, I know what I am talking about. For one thing, software is much more expensive to create. “You have to pay people to create it, maintain it, and distribute it,” Paik argues. “Because it is expensive to create, software has to make money.” Companies pay for software licences, Software as a Service or SaaS subscriptions and per-seat pricing. The more people using a particular software in your company the more money you pay. Not surprisingly, software margins have historically been more than 90% with zero marginal cost of distribution.
Paik also makes the point that software is expensive because good software developers are expensive. “They are skilled translators — they translate human language into computer code and vice-versa. Large language models or LLMs have proven themselves to be remarkably efficient at this and will drive the cost of creating software to zero.” With AI Copilots writing all the computer code or barking orders to Siri on your iPhone to write your next software, the value of coders and software engineers will ultimately decline.
AI-driven disruption
Sink your teeth into in-depth insights from our contributors, and dive into financial and economic trends
What happens when software no longer has to make money? Paik says the world will experience a Cambrian explosion of software, the same way it did with content. “Vogue wasn’t replaced by another fashion (magazine); it was replaced by 10,000 influencers. Salesforce will not be replaced by another monolithic customer relationship management or CRM software firm. It will be replaced by a constellation of things that dynamically serve the same intent and pain points.”
As Paik sees it, software firms will be replaced the same way media companies were, giving rise to a new set of platforms controlling distribution. “SaaS and ARR (annual recurring revenue) are shorthand to understand the old model of business building in software, one where the expense associated with creating software was a moat,” the New York VC notes. “The invisible hand has stayed in software for a long time but LLMs will usher in its swift, familiar corrective force. Majoring in computer science today will be like majoring in journalism in the late 1990s.”
Ben Thompson, who writes the Stratechery blog, concedes that software firms face real challenges in the new AI era. He cites the per-seat software license fees, which he says “make a lot less sense when AIs are delivering value by eliminating seats, not adding them.” However, he argues that one of the advantages that companies like Databricks, an analytics and AI firm, have is that their business models are based on usage. “Software needs to be stable and predictable and have infrastructure to run on,” Thompson notes.
That is a lot easier to buy from a company than to manage yourself. “Businesses don’t want to be IT departments; they want to achieve business results, and time spent trying to get stuff to work is a waste of resources.” That doesn’t mean there won’t be upheaval as software firms figure out how to leverage AI on both the supplier and consumer sides.
Indeed, he argues, there may very well be a different set of winners, but as long as there are jobs to be done, there will be opportunities to sell software — whatever it might look like. Just because AI can code doesn’t mean the end of software. It just means you and I will write our own software, which will be as user-friendly as we want it to be.
Assif Shameen is a technology and business writer based in North America
