NOW down 25% after Q3 — my thesis with full numbers. Please try to break it.
retained text
Disclaimer : I redacted my post partially with AI as English isn't my main language. I own every argument stated in this post and used my own way of calculating the future share prices. Please feel free to criticize my method. I bought ServiceNow around $100 a few months back and held through earnings were NOW had become my strongest position with 60% upside and 10% of total portfolio (hesitated to sell at 160 but didn't as my model kept showing more upside as you will see. That's the very model I want to question now). For the record, I did the same calculation with Atlassian TEAM during the same period : Bought at 70, sold at 200. Currently, I believe the stock is to expensive. I used the exact same method that I will display now to make this assumption. **About NOW :** Reddit seems to agree that cRPO is collapsing, management is bad, Armis was a mistake and GAAP margins are falling apart (saw all these arguments in this very sub, on GOOG and NOW posts). I want to test my thesis against these arguments, maybe try and confort some bagholders like myself, or be grateful to be taught a lesson. So here are my assumptions. Tell me where they’re wrong. **1. The moat, and what threatens it** *Why I think it holds* * ServiceNow is the system of record for IT, HR and customer workflows in large companies. It runs on one data model, and once it is wired into a company’s processes, ripping it out is a multi-year project. I've been working 10 years in IT and worked a lot with Service Now Products. I know people don't like it, but they don't like any of the software products in my experience. * Renewal rate is still 98%, when of the best of the industry * The AI pitch (the “control tower” for enterprise agents) builds on that position. If companies deploy agents, they need governance, permissions and audit trails. NOW already owns those workflows. *Real risks, which I price in* * **Seat cannibalisation:** AI agents could shrink the per-seat model. Management says customers keep seats and add consumption, with no demand for outcome-based pricing, but that is their word, not proof. * **Capital allocation:** Armis cost about $7.6B in cash. Interest expense went from $6M to $81M a quarter, buybacks stopped in Q3, and Armis costs \~0.75 pt of operating margin in Q4. * **Quality of earnings:** GAAP operating margin fell to 9.4% from 16.8%. SBC was $633M in Q3, 15.4% of revenue, up 28% year on year against revenue up 20.5%. This is the bear point I take most seriously. **2. Revenue growth assumptions** * **Q3:** subscription revenue +20.5% constant currency (cc), above guidance; cRPO +21% cc; total RPO $29.7B, +22%. * **Q4 guide:** cRPO +18.5%. That is a slowdown. Management attributes \~0.75 pt to technical effects: * early renewals pulled into Q3, about half of the Q3 beat; * on-prem public-sector deals, −0.25 pt; * termination-for-convenience clauses that exclude \~0.5 pt of backlog from cRPO but not from revenue. * **Subscriptions:** Q4 is guided at +21.5% cc, of which \~1.75 pt is Armis. That leaves about 19.75% organic. * **Organic cRPO:** Armis’s contribution is not disclosed. Assuming it is similar (\~1.75 pt), I estimate underlying organic cRPO around 17%. * **AI:** the net new ACV target was raised to >$1.7B, up from $1B at the start of the year. AI net new ACV is +65%, agentic consumption ×13 and AI deals over $1M ×4.5. Management says this shows up in revenue in 2027, not before. My Base case is global revenue +17.5% / +16.5% / +15.5%. That is below the current organic run rate and **slightly below 2027 consensus**. Bear assumes AI eats the seat model and growth falls to 12% a year. Bull assumes AI consumption starts to convert in 2027 (+19.5% / +18.5% / +17.5%). **3. SBC and buybacks** I value the company on **FCF minus SBC**, never on non-GAAP EPS, because SBC is a real cost. Guided FY26 FCF margin is 36%, which leaves about 21% of revenue after SBC. * **SBC:** I assume 15% / 13% / 11% of revenue in FY29 (Bear / Base / Bull) according to what management said. Base needs SBC to grow \~10% a year, slower than revenue. If SBC keeps outgrowing revenue in 2027, my Base is wrong. * **Share count:** with no Q3 buybacks and Q4 diluted shares guided at 1.05B, I assume no net shrink in Base (1.05B), net dilution in Bear (1.08B) and small net buybacks in Bull (1.03B). **4. Exit multiples (on FY29 FCF − SBC)** * **Bear 24×** (\~13× reported FCF): priced like mature software, no premium. * **Base 35×** (\~23× reported FCF): a business still growing mid-teens after 2029, with 98% retention and \~37% FCF margins. In line with historical valuation * **Bull 40×** (\~29× reported FCF): Because that's a bull case :D **5. The numbers** Base year FY26 expected : revenue $16.19B ; \~1.04B diluted shares. **Scenario** |**Revenue growth Y+1 / Y+2 / Y+3** |**FCF margin (Y+3)** |**SBC CAGR** |**Net share count CAGR** |**Diluted shares (Y+3)** |**FCF − SBC / share** |**Exit multiple** |**Price (FY29)** Bear |+12% / +12% / +12% |32% |+11.4% |+1.3% |1.08B |$3.58 |24× |$86 Base |+17.5% / +16.5% / +15.5% |37% |+10.4% |+0.3% |1.05B |$5.85 |35× |$205 Bull |+19.5% / +18.5% / +17.5% |39% |+6.3% |−0.3% |1.03B |$7.32 |40× |$293 Weighted 25/50/25, that gives \~**$197 by late 2029, roughly +74% from \~$113, or \~20% a year.** This, IMO, is in line with value investing and mid to long term hold, or even adding more following recent dips **So, bears: which assumption is too generous?** Growth, SBC, share count or multiple? Numbers welcome. EDIT : thanks for the comments TLDR is that my multiple assumptions are to high, especially on high interest rates and with the business getting more and more mature (slower growth at 5, 10 years…) Basically better assumptions should be Bear : 15x multiple : 53,$ Base : 22x multiple :128$ Bull : 30x multiple : 219$ Average is : 145$
document fetched 2026-10-30 00:47 · canonical https://reddit.com/r/ValueInvesting/comments/1xit19a/now_down_25_after_q3_my_thesis_with_full_numbers