ServiceNow: Can AI Really Replace It? The Invisible Moat of Switching Costs

Can AI Really Replace ServiceNow?

Not in the short term. ServiceNow's real moat is not the software itself but how deeply it is wired into how organizations actually run. To tear that out, an AI agent would have to redesign the entire organization with it.

This post unpacks that answer. The chart looks scary — down 30% year-to-date, down 50% from 2025 highs. But whether the business is broken to the same degree is a separate question.

What ServiceNow Actually Does

ServiceNow is not a consumer product. Hospitals, banks, governments, and Fortune 500 companies run their internal operations on it — IT tickets, HR workflows, legal requests, facilities management.

Why does that matter? These are not weekend-replaceable systems. They are wired into the nervous system of the organization. Ripping them out means redesigning business processes themselves.

The Real Nature of Switching Costs

Replacing QuickBooks and replacing ServiceNow are not the same problem. With QuickBooks, you move accounting data and you are done — the moat is inertia, not cost. ServiceNow tangles up internal policies, permissions, automation rules, and integration points.

That is what keeps the pricing power intact. Microsoft, Oracle, and Salesforce all compete in this space. But winning new customers and pulling existing customers away are very different fights. New deals can face price pressure; stealing entrenched accounts is much harder.

What the Numbers Show

• Market cap: ~$105B
• Enterprise value: $111B (about $6B in debt)
• Free cash flow last year: $4.5B — clears debt in ~1.5 years
• 5-year acquisition spend: $2.35B. 5-year average FCF: $2.9B/year. Growth is not bought

Growth rates:

• 10-year revenue growth: 30%
• 5-year: 24%
• 3-year: 22.5%

Very few large SaaS companies hold this pace. Revenue already past $10B and still compounding in the low-to-mid 20s.

Is a 60x PE Justified?

Let me be honest here. A 60x PE is expensive. A 23x P/FCF is materially higher than what Intuit and Salesforce trade at.

Return on capital at 6.3% is also low. The 10-year average actually started in negative territory and has climbed since. By that metric alone, the stock is not attractive.

But two things matter on the other side.

First, the combination of absolute scale and growth rate is rare. Holding 20%+ growth past $10B revenue is unusual.

Second, analysts model a doubling of revenue and earnings over four years — 19–20% revenue growth, 19–33% EPS growth. I am not that optimistic.

What the Stock Is Worth on My Inputs

My assumptions:

• Revenue growth: 9% / 13% / 17%
• FCF margin: 30% / 33% / 36%
• Exit PE: 16 / 19 / 22
• Required return: 9%

Output: low $95, fair $160, high $260. At current prices, that implies roughly a 15% expected return.

Whether that is "enough" is your call. The tradeoff I see is clear — pay a richer multiple, get a stickier moat.

FAQ

Q: What is the biggest risk to ServiceNow? A: Not AI itself — multiple compression. If growth slows into the high teens, a 60x PE is hard to defend.

Q: Is Microsoft's Power Platform a threat? A: Yes, but mostly for new customers. It is unlikely to dislodge entrenched ServiceNow installations; price pressure on net new deals is the more likely outcome.

Q: Of Intuit, Salesforce, and ServiceNow, which is safest? A: On moat strength, ServiceNow. On price, Salesforce. On balance, I would lean Intuit.

Q: Is it okay to buy at a 60x PE? A: The question is not "is it safe" — it is "does the expected return meet your hurdle." At my 9% required return, the implied ~15% return is attractive, but if you want a larger margin of safety, waiting is the rational choice.