Apologies if this topic has already been done to death. I was prompted to dig a bit deeper after the very strong quarterly report from Service Now (ticker NOW).

I set my research assistant (Claude) the following task:

There is increasing sentiment in financial markets that SaaS companies future revenue and profitability will be threatened by the increasing sophistication and ease of do it yourself coding using tools like Claude code and Claude cowork.

Intuitively, the greatest threat should be to SaaS companies that provide a low code platform for businesses to build their own workflows, for example Service Now, UIPath and Appian.

Yet Service Now have just reported very healthy revenue growth in the current quarter, and have increased their guidance for the full financial year.

Provide an in depth analysis of reasons why the prevailing narrative that SaaS companies are threatened with extinction does not seem to be accurate for companies such as Service Now.

This was the response:

This is an excellent observation that highlights a significant gap between market narrative and business reality. Let me provide an in-depth analysis of why ServiceNow and similar enterprise platforms are proving far more resilient than the "AI will replace SaaS" thesis suggests.

## The Enterprise Integration Moat

ServiceNow isn't just a low-code platform—it's deeply embedded infrastructure. They've spent 15+ years building integrations with thousands of enterprise systems (SAP, Oracle, Salesforce, legacy mainframes). When a company has ServiceNow managing their IT service desk, HR workflows, and asset management, with data flowing between these systems and their ERP, replacing it isn't a matter of coding ability—it's an extraordinarily complex migration project. The switching costs are measured in millions of dollars and years of effort, not lines of code.

AI coding tools make it easier to build *new* applications, but they don't solve the problem of ripping out and replacing systems that have become organizational nervous systems.

## Data Network Effects and Institutional Knowledge

ServiceNow's real value proposition has evolved beyond the platform itself. Their systems contain years of organizational knowledge: incident patterns, resolution workflows, approval hierarchies, compliance audit trails, and performance metrics. This accumulated data trains their own AI models and informs best practices. 

A company building something from scratch with Claude Code would start with a blank slate—no historical data, no pre-configured workflows reflecting industry best practices, no accumulated institutional knowledge about what works in enterprises similar to theirs.

## The "Build vs Buy" Calculation Hasn't Changed as Much as Assumed

Yes, AI makes building easier, but several traditional enterprise factors remain unchanged:

**Ongoing maintenance burden**: Building your own system means you own the maintenance, security patches, feature updates, and scaling challenges forever. ServiceNow employs thousands of engineers continuously improving the platform. For most enterprises, redirecting engineering talent to maintain internal workflow tools rather than building competitive differentiation remains unattractive.

**Compliance and security**: ServiceNow maintains SOC 2, ISO 27001, GDPR compliance, and dozens of other certifications that enterprises require. Building compliant systems from scratch—even with AI assistance—requires significant specialized expertise.

**Skills and staffing**: While AI lowers the technical barrier, someone still needs to design workflows, make architectural decisions, handle edge cases, and maintain the system. The shortage of technical talent hasn't disappeared.

## ServiceNow's AI Integration Strategy

Critically, ServiceNow hasn't been standing still. They've aggressively integrated AI capabilities into their platform, including their own generative AI features. This means customers can get AI-powered workflow automation *within* ServiceNow rather than replacing it. They're positioning themselves as the enterprise AI layer rather than being disrupted by it.

Their "Now Assist" AI features provide natural language interfaces, automated case resolution, and intelligent workflow suggestions—essentially offering customers the benefits of AI without requiring them to rebuild everything.

## The Low-Code Platform Paradox

Here's an underappreciated insight: low-code platforms like ServiceNow actually *benefit* from AI coding tools in certain ways. These platforms are designed for citizen developers and business users. As AI democratizes coding, it also lowers the barrier to *using* platforms like ServiceNow more effectively—writing custom scripts, building integrations, and extending the platform becomes easier for the existing user base.

## Market Maturity and Greenfield vs. Replacement

The "AI threatens SaaS" narrative is most credible for:

- New startups choosing their initial tech stack

- Simple, single-purpose tools with shallow integrations

- Companies with strong technical teams and unique requirements

But ServiceNow primarily serves large, established enterprises with complex, interdependent systems. The decision matrix for these organizations weighs reliability, vendor support, and de-risking far more heavily than technical feasibility alone.

## The Analyst and Investor Perspective Gap

I suspect part of what you're observing is that market sentiment is driven by forward-looking concerns (what *might* happen in 2-3 years), while ServiceNow's current results reflect actual enterprise decision-making happening today. There's a lag between when a technology becomes possible and when enterprises actually change their purchasing behavior.

Additionally, investors often overweight disruption narratives because the potential magnitude of change is so large, even if the probability and timing are uncertain. The "AI will disrupt everything" story is compelling and gets attention, even if the reality is more nuanced.

## Not All SaaS Is Equally Vulnerable

The companies most vulnerable to AI-assisted DIY solutions are likely:

- Point solutions with shallow feature sets

- Tools primarily valued for their UI rather than underlying complexity

- Products with weak network effects or data moats

- Software serving technically sophisticated customers

ServiceNow, Salesforce, SAP, and similar platforms are essentially enterprise operating systems with decades of accumulated functionality, integrations, and organizational dependency. They're fundamentally different from a project management tool or form builder that could more easily be replicated.

## Conclusion

The prevailing narrative isn't entirely wrong—AI coding tools *will* change the build-versus-buy calculation at the margins, particularly for new companies and simpler use cases. But for enterprise platforms like ServiceNow, the moats remain formidable: switching costs, integration complexity, compliance requirements, ongoing maintenance burden, accumulated data and institutional knowledge, and continuous platform evolution all combine to make replacement far less attractive than the pure "can we build this?" question suggests.

ServiceNow's strong results indicate that enterprises continue to see more value in established, integrated platforms than in redirecting resources to build and maintain custom alternatives, even as those alternatives become technically easier to create. The threat is real but gradual, and established players have time to adapt—which many, including ServiceNow, are actively doing by integrating AI into their platforms rather than being disrupted by it.

A few thoughts on AI and software businesses

There’s a growing narrative that AI will allow small businesses or even individuals to simply “vibe code” their own software and replace incumbents like Xero, Sharesight, or other SaaS platforms. I think that view misses some important realities.

Most small business owners are busy running their businesses. They don’t have the time, desire, or risk appetite to build and maintain custom software just to save a few dollars on a subscription. Software like Xero exists to remove friction, not add another problem to solve.

When you buy SaaS, you’re not just paying for code. You’re paying for:

• Embedded domain knowledge
• Regulatory and tax compliance
• Ongoing updates as rules change
• Reliability and accuracy
• The confidence that “this is done properly”

That “how-to” and accumulated IP is the real product.

Take Sharesight as a simple example that many Strawman readers will relate to. Are you realistically going to vibe-code your own Sharesight replacement? Sure, maybe you could build something that tracks transactions — but do you really want the hassle of:

• Maintaining it year after year
• Ensuring capital gains are calculated correctly
• Handling edge cases and tax changes
• Still needing to manually check everything anyway

I personally pay for Sharesight so my tax returns are easy and low-stress. The whole point is not having to think about it.

Of course, some software companies will be disrupted, and others will need to disrupt themselves. AI agents, automation, and workflow changes are coming, no doubt. Margins may compress in some areas. But many of these businesses are mission-critical, deeply embedded, and supported by recurring revenue models.

This reminds me of the narrative when GLP-1 drugs were supposedly going to “destroy” ResMed. The market often extrapolates a new technology to its most extreme conclusion, ignoring nuance, adoption friction, and real-world behaviour.

AI will absolutely change the landscape. There will be new winners and new losers. Some companies will be disrupted, others will adapt and thrive. But the idea that most businesses will casually replace trusted, compliant, mission-critical software with DIY AI solutions feels overstated.

On a related note, I recently watched the DeepMind documentary The Thinking Game — well worth a watch for anyone interested in where AI is genuinely heading:

https://www.youtube.com/watch?v=J5XFYMsczy8

Happy to hear the contrarian view.

One angle on "death of SaaS" we have not yet discussed on this forum (probably because it was obvious to everyone one,except me), and that is contributing to the market sell-off without doubt (in my mind), are the reports that enterprises are spending some of their IT budgets directly on AI solutions, at the expense of purchasing from legacy software providers.

In a world of limited capital, unless IT budgets are being expanded, then every $ spent on AI products directly, contributes to a SaaS-seat-licence not purchased, a module not upgraded to, etc. etc. or at least to delays.

So in a zero sum equation we have:

Assume $IT(yesterday) = $IT(today) = $IT

$IT = $SaaS (yesterday) = $Saas (today) + $AI(today); where $AI(yesterday)=$0

if $AI(today) > 0

then $SaaS(today) < $SaaS(yesterday).

Of course, there is nothing to say that we are in a zero sum world. So, if increasing the investment in IT increases overall productivity, there there doesn't have to be a loser, i.e., $AI can be >0, and $SaaS(today) can be > $SaaS(yesterday).

Mutiple reports have started to look for this phenomenon. So far, in aggregate, relatively low precentages of enterprise IT budgets, reportedly less than 1 percent.

Of course, weighing against this, is the report from MIT late last year from a survey of 300 enterprises, where 95% so far saw no measurable returns from their AI initiatives.

This is further supported by a round table of some WSJ Journalists from Davos (sorry behind the paywall) that I watched earlier this week, where a journalist reported attending a meeting CEOs of major banks in Davos. Basically, the journalist said that the banks had been counting on major investments in AI to help them dramatically reduce headcounts. BUT that they weren't seeing progress. This was put down to the problems of having multiple enterprise systems, and the challenges of integration. (Something we have discussed here in some detail!)

Of course, this is going to evolve over years. At least a few and maybe many. But clearly the market has taken an initial view of the future. Is it prescient, premature or an over-reaction?

So What?

A key actionable insight from this is to look for changes in SaaS Revenue growth and ARPU in the current and future results rounds. And perhaps ahead of the ASX season, to have a look at the ongoing Q'ly reports in the US happending now.

The market has clearly moved to anticipate something. But what is the evidence that it is actually starting to occur?

In the reading I have done, clearly seat-based pricing appears to be under threat. The "consensus" if I can call it that, is that AI - whether purchased directly, or embedded in existing software tools - is going to make workers much more productive, requiring fewer human "seats" to do the same work.

So one question for SaaS CEO's is "how are you anticipating that changes in your customer productivity through using AI will require fewer seats for your licences?" "How is your commercial model taking account of that?"

Clearly, $WTC saw it coming, and they have already pivoted their commercial model overnight. They clearly saw the risk as material enough, and very highly geared to seat-based pricing that they decide the risk was high enough to warant risking a significant customer backlash - which anecdotally, seems to be happening at some level. It will be very interesting to hear $WTC reporting on the feedback from their customers. I hope there are candid and honest reponses.

A nice piece today on this topic in the WSJ.

My Takeway from the Article: I'm not expecting a change in market sentiment any time soon for software, unless AI materially comes off the boil. Each SaaS company is going to have to earn its multiple through its own delivery (Margins, EPS and FCF growth), and unlikely to be carried by sector multiple read across. Strong balance sheets will be important.

Wall Street Has Fallen Out of Love With Software Stocks

Declines show how quickly sentiment can shift as money pours into the next hot investment

By Sam Goldfarb Jan. 25, 2026 5:30 am ET

Quick Summary

• Software stocks, including Salesforce, Adobe, and ServiceNow, have declined by at least 30% since the beginning of last year due to AI competition.
• An S&P index of small and midsize software stocks is down over 20% in the same period, with accelerated declines this month.
• Thirteen software companies have defaulted on broadly syndicated loans in the past two years, a new trend for the sector.

Software companies’ pitch to investors could use an upgrade.

Once a favorite of Wall Street, software stocks have been sliding lately, with investors increasingly concerned about how the sector could be upended by their newest crush: artificial-intelligence companies.

Rocked by the emergence of “vibe coding”—the practice of using AI tools to quickly produce apps and websites—software heavyweights Salesforce CRM -0.02rease; red down pointing triangle

, Adobe ADBE 0.45%increase; green up pointing triangle

 and ServiceNow NOW 3.54%increase; green up pointing triangle

 are all down at least 30% since the start of last year. 

An S&P index of small and midsize software stocks is also down more than 20% over that period, with declines accelerating this month after the introduction of Anthropic’s Claude Code, an AI tool that industry insiders have said can dramatically shrink the time it takes to build even complex software.

At a time when many investors are wondering if the AI investment boom itself might be a bubble, software’s slump is a reminder of how quickly fortunes can change on Wall Street.

“The narrative has really shifted,” said Rishi Jaluria, a software analyst at RBC Capital Markets. 

Investors, he said, have gone from initially thinking that software companies could benefit from AI to asking, “Is AI just the death of software?”

Investors will gain insight into the overall state of the tech sector in the coming week, with Apple, Meta Platforms and Microsoft set to report earnings. The Federal Reserve will also meet, though it isn’t expected to change interest rates.

Just a few years ago, it was software, not AI, that was at the center of an investing frenzy. 

Over the previous decade, the sector had seemingly made good on Marc Andreessen’s promise that software would come to “eat the world.” Fueling its rise was the growth of high-speed internet and cloud computing, in which software companies could rent storage space from the likes of Amazon.com rather than own data centers themselves.

Software startups blossomed to serve every niche imaginable, from helping yoga studios handle scheduling and billing issues to securing businesses from cyberattacks. 

Once seen as volatile, software came to be viewed by Wall Street as the paragon of stability, because businesses seemed loath to switch products once they became embedded in their workstreams. Investors also liked the emergence of multiyear subscription contracts that offered steady revenue.

Not only did software stocks soar, the sector became the center of a credit boom, with debt investors eagerly funding a surge of private-equity buyouts. The pandemic only supercharged the dynamic, thanks to the move to remote work and a drop in interest rates, which made it even easier to borrow money.

But enthusiasm began to wane with the rate increases in 2022 and the shift back to the office. For debt investors, software companies also began to look less invincible, with competitive pressures and highly leveraged balance sheets driving more businesses into distress.

Before this decade, software defaults were practically unheard of, partly because lending to software companies was so new. But in the past two years, 13 software companies have defaulted on loans that were broadly syndicated to investors, including both traditional defaults, such as bankruptcies, and out-of-court restructurings known as liability management exercises, according to PitchBook LCD.

One of those companies was Quest, the maker of OneLogin software that authenticates employees logging into their work platforms. Bought by Clearlake Capital in early 2022 with the help of $3.6 billion in loans from investors, the company benefited from the shift to remote work but was pressured by the weight of its debt and competition from the likes of Okta, a larger, publicly traded company. It struck a debt-restructuring deal with lenders last June.

Default rates for software loans are still lower than they are for buyout loans generally, and investors have hardly abandoned the sector. Still, the extra yield, or spread, that investors demand to hold software loans over a benchmark short-term interest rate has ticked higher over the past 15 months, even as overall loan spreads edged lower, according to PitchBook LCD.

“The investor base is definitely scrutinizing these software names much more closely,” said Vince Flanagan, a portfolio manager and senior leveraged-finance research analyst at Seix Investment Advisors.

The emerging threat from AI has only added to that caution. The big risks are that existing software companies could face increased competition from new entrants and that businesses could develop more software themselves rather than paying specialists for it.

In reality, few investors and analysts think that software companies will become obsolete in the foreseeable future. The more pressing risk is that it could become more difficult to increase revenue, as customers experiment with other options rather than paying more for the usual updates and add-ons, RBC’s Jaluria said.

Ultimately, Jaluria said, AI could damage “fat, lazy incumbents” while helping others that are innovative and can use AI to improve their products. 

Uncertainty about the future of software is compounded by the fact that investors lack answers to basic questions about what the emerging AI industry will look like. 

Excitement over AI has played a major role in propelling stocks to new records in recent years, with investors especially eager to buy the stocks of so-called AI hyperscalers like Alphabet and Microsoft. Investors, though, have grown more discerning in recent months in choosing what stocks to buy even among those companies.

As companies start borrowing heavily to fund the build-out of AI infrastructure, debt investors have also taken a more cautious approach than they once did with software companies—forcing big-spending companies like Meta and Oracle to pay high interest rates on new bonds relative to their credit ratings.

“Investor sensitivity is high,” said Rich Gross, a senior fixed-income analyst at Columbia Threadneedle Investments. 

Investors, he added, are asking “are these investments sustainable? Are they going to be profitable? Are there going to be cash flows, or will there not be?”