A-Portfolio

The company in one breath

Rubrik, ticker RBRK on the New York Stock Exchange, sells software that protects company data, spots attacks, and helps a business recover after ransomware or an AI agent screws something up. Customers pay subscriptions because backups are boring right up until the building is on fire.

Rubrik now pitches itself as a security and AI operations company. The older engine protects data and restores systems. The newer bet watches AI agents, limits what they can touch, and reverses bad actions. That is a logical expansion, not an AI sticker slapped on the hood.

Why now

Rubrik reported fiscal second-quarter results Thursday night, and the business beat every target management gave it. Revenue grew 38%, subscription annual recurring revenue grew 33%, and management raised the full-year outlook. Those are category-leader numbers.

The stock had already sprinted. It closed Thursday at $107.02, which means buyers pushed it up 11.3% before the earnings release. Good entry quality would mean the business improves before the price celebrates. Here, the party started while the bartender still carried in the ice.

Then the earnings reaction arrived. Rubrik traded at $101.00 at 7:07 a.m. Eastern Friday, down 5.6% from Thursday's close. That means the market liked the quarter less than the headline numbers suggest, probably because expectations had become ridiculous. Good looks like a reset that gives long-term owners a fair bill. A six-percent haircut does not do enough when the stock has gained more than 200% from its $32 IPO price.

The real reason to care sits beyond one quarter. Ransomware recovery is no longer just an insurance policy, and companies now need guardrails for AI agents that can touch production systems. Rubrik already owns the recovery relationship inside large enterprises. If it turns that trust into an AI security platform, this can become a bigger category. I do not know yet whether the new AI products will create material revenue. Management did not give us that number.

The numbers, translated

• Revenue reached $427.3 million, up 38%. Revenue is the money Rubrik earned before costs. Good for a young subscription security company is growth above 25%, so this clears the bar with room to spare.
• Subscription revenue reached $407.2 million, up 37%, and supplied 95% of sales. That means nearly every dollar came from repeat contracts rather than one-time hardware or services. Good looks like more than 90% recurring, so the sales base is sturdy.
• Subscription annual recurring revenue reached $1.66 billion, up 33%. This number annualizes active subscription contracts, so it shows the size and speed of the repeat-revenue engine. Good at Rubrik's scale is growth above 25%, and 33% says demand still runs hot.
• Cloud annual recurring revenue reached $1.48 billion, up 39%. This is the repeat value of Rubrik's cloud subscriptions, the products that should carry the future. Good looks like cloud growth beating total subscription growth, and 39% versus 33% passes that test.
• Customers spending at least $100,000 a year rose to 3,084, up 23%. These are large accounts that justify an expensive enterprise sales force. Good would be growth above 20%, so Rubrik keeps landing and expanding serious customers.
• Gross margin was 78.4%, down from 79.5%. Gross margin is what Rubrik kept after the direct cost of delivering its products, before sales, research, and office bills. Good software economics usually keep at least 75 cents of each dollar, so the level is strong, but the decline deserves watching.
• The standard-accounting operating loss was $71.9 million, equal to 16.8% of revenue. Standard accounting includes share pay and other real owner costs. Good would be a loss below 10% of revenue on the way to profit, so Rubrik still burns too much on the official scorecard.
• Free cash flow was $65.7 million, up 14%, for a 15% margin. Free cash flow is cash left after running the company and funding the equipment and internal software it needs. Good for a company growing this fast is a margin above 15% and rising, so Rubrik hits the floor but slipped from 19% a year ago.
• Stock-based compensation was $101.0 million, equal to 23.6% of revenue. This is employee pay handed out in shares, which dilutes owners even when cash never leaves the bank. Good would be below 15% of revenue and falling, so this bill is still ugly.
• Cash and short-term investments totaled $1.75 billion against $1.13 billion of convertible notes. That leaves roughly $615 million more cash than debt, which means the balance sheet can fund growth without begging lenders. Good is positive net cash, and Rubrik clears it.
• Management now expects full-year revenue of $1.685 billion to $1.693 billion and free cash flow of $323 million to $333 million. The midpoint means about $1.689 billion of sales and $328 million of leftover cash. Good looks like management raising both growth and cash together, and that is exactly what happened.
• At $101.00, the fully diluted equity value is about $23.0 billion and the operating business costs about $22.4 billion after net cash. Using management's full-year midpoint, that equals roughly 13.3 times sales and 68 times free cash flow. Good for a roughly 30% grower would be closer to 10 times sales or below 50 times cash flow, so the valuation still demands near-perfect execution.

What I like

First, Rubrik has a real category and a real customer problem. Companies do not cancel cyber recovery because the quarterly budget feels moody. The product protects the thing every executive suddenly remembers after an attack, the ability to reopen the doors.

Second, the growth quality looks good. Subscription sales supplied 95% of revenue, cloud annual recurring revenue grew 39%, and large customers grew 23%. Those numbers mean repeat contracts, modern products, and enterprise adoption all moved in the same direction. Good companies do not need one metric to distract you from the other five.

Third, the model now produces cash. The company generated $65.7 million of free cash flow in the quarter and expects about $328 million for the year. That means Rubrik can fund expansion from customers instead of selling hope every time payroll arrives. Good cash durability starts with positive cash and improves toward a 20% margin.

Fourth, founder and chief executive Bipul Sinha still runs the place. Founder leadership can keep a company aggressive when a category changes fast. Good founder leadership pairs vision with discipline, and the raised growth and cash outlook suggests the discipline has started to catch up.

What worries me

The valuation is the obvious bruise. At roughly 13.3 times expected sales, investors already pay for a future winner. Good asymmetry means the upside can surprise while the downside has padding. This price offers ambition, not padding.

Share pay is the quieter problem. Stock compensation consumed 23.6 cents of every revenue dollar, and the basic weighted share count rose 5.8% from a year ago. That means employees earned a bigger slice while existing owners got diluted. Good would push share pay below 15% of revenue and share-count growth below 3%.

The standard-accounting loss still matters. Rubrik lost $71.9 million from operations even though the adjusted version showed a profit. Adjusted profit removes costs management calls unusual, including share pay that somehow shows up every quarter wearing the same jacket. Good means the official operating loss shrinks below 10% of sales without sacrificing growth.

Cash growth also trailed revenue growth. Free cash flow rose 14% while revenue rose 38%, and the cash margin fell from 19% to 15%. That means sales expanded faster than leftover cash. Good would reverse that pattern and move the margin above 20%.

Finally, the AI-agent story may be real, but the revenue proof is not here yet. Good would mean management discloses customer adoption, contract value, or annual recurring revenue tied to agent security. Until then, I treat AI as optional upside, not the reason to pay the bill.

What would make this a buy

I want price and proof, not one or the other.

The price door opens around $80 to $85. At $80, the operating business would cost about 10.4 times the midpoint of this year's sales and 54 times expected free cash flow. That is still expensive, but good entry quality gives a fast grower some room for an ordinary quarter.

The proof door opens if subscription annual recurring revenue stays above 30% growth, the free-cash-flow margin reaches 20% or better, stock compensation falls below 20% of revenue, and the standard-accounting operating loss drops below 10% of sales. Those targets mean growth stays elite while owner economics stop acting like a charity raffle. I would pay more after Rubrik proves that combination for two quarters.

What makes me walk away

If subscription annual recurring revenue growth falls below 25%, the premium-growth engine has cooled. Good needs the repeat contract base to grow faster than the broader security market.

If management cuts full-year free-cash-flow guidance below $300 million, cash discipline has cracked. Good means keeping cash above the current $323 million to $333 million range, not explaining the miss with new adjectives.

If stock compensation stays above 25% of revenue while the basic share count keeps growing above 5%, owners are funding the payroll through dilution. Good means both numbers fall. One quarter can wobble, a pattern gets this thrown out.

If the new AI products fail to produce disclosed customer or contract evidence by next fiscal year, I remove AI from the valuation story. Good product stories eventually show up in dollars. PowerPoint cannot pay rent.

IPO Watch

• Siyata PTT (private before listing), expected September 8 under proposed ticker PTT on the Nasdaq Capital Market. The filing sets no price range because this is a direct listing, which means existing holders may sell into the market instead of the company selling a fixed block through an underwriter. The company makes rugged push-to-talk phones and vehicle devices that run over cellular networks. I write it up only after the market establishes a sane price and the company proves that carrier approvals turn into durable sales. The filing shows private transactions from $0.29 to $7.00 a share, which means price discovery could resemble a food fight. Good would be transparent volume, a stable float, and audited growth after listing.

Earnings On Deck

• Rubrik (RBRK), reported August 27 after the close. I watched subscription annual recurring revenue, cloud growth, cash, and share pay. Good meant recurring revenue above 25% growth and positive free cash flow, which Rubrik delivered. Bad meant cash margin slipping and dilution staying heavy, which also happened.
• MongoDB (MDB), September 1 after the close. Watch cloud-database growth, large-customer spending, and standard-accounting profit. Good means customers spend more without sales costs swallowing the gain. Bad means AI database demand sounds loud while billed business slows.
• Zscaler (ZS), September 3 after the close. Watch contracted backlog, large deals, and free cash flow. Good means future signed business grows faster than current revenue while cash stays strong. Bad means competition lengthens sales cycles and management buys growth with discounts.

The verdict

Score: 72 out of 100, watch. Rubrik earned points for category leadership, 33% recurring-revenue growth, 39% cloud growth, a deep enterprise customer base, founder leadership, positive cash generation, and an expanding AI-security opportunity. It lost points for a 13.3-times-sales valuation, a 16.8% standard-accounting operating loss, share pay equal to 23.6% of revenue, dilution, and a falling cash margin.

The bucket is watch. Candidate status starts at 80, and Rubrik does not get there because the business is excellent while the transaction is lousy. I want this company. I want somebody else to pay today's cover charge.

The DBTech read

Stock-based compensation pays employees with ownership instead of cash. It helps a young company hire talent, but every new share makes your existing slice smaller. Compare share pay with revenue because $101 million sounds different when you learn it consumed 23.6 cents of every sales dollar. Good means the percentage and the total share count both fall as the company matures. If management excludes share pay from profit forever, put it back in yourself.

Sources and as-of date

Research completed August 28, 2026 at 7:07 a.m. Eastern.

The $101.00 premarket price came from Nasdaq's real-time New York Stock Exchange feed at 7:07 a.m. Eastern on August 28, 2026. Yahoo Finance independently showed $101.00 at 7:07:28 a.m. Eastern. The $107.02 regular-session close came from Nasdaq at 4:00 p.m. Eastern on August 27, 2026. Alpaca's IEX broker feed showed a $107.05 last trade at 3:59:53 p.m. Eastern on August 27, a three-cent difference caused by the narrower IEX feed.

Quarterly financial statements, annual recurring revenue, customer counts, guidance, cash, debt, share compensation, and product launches came from Rubrik's earnings exhibit filed with the U.S. Securities and Exchange Commission on August 27, 2026: https://www.sec.gov/Archives/edgar/data/1943896/000194389626000055/rubrikinc-991pressrelease7.htm

The $32 IPO price came from Rubrik's final prospectus filed with the SEC on April 26, 2024: https://www.sec.gov/Archives/edgar/data/1943896/000119312524118478/d359771d424b4.htm

The IPO calendar came from Nasdaq's September 2026 calendar, last updated August 27, 2026. Siyata PTT's direct-listing structure, proposed ticker, business description, and private transaction range came from its amended Form F-1 filed with the SEC on August 25, 2026: https://www.sec.gov/Archives/edgar/data/2110025/000121390026093558/ea0303098-f1a1_siyata.htm

The earnings calendar came from Nasdaq for August 31 through September 4, 2026. Community research covered Reddit, GitHub, Hacker News, and YouTube from July 29 through August 28, 2026. The Rubrik-specific social sample leaned heavily on small YouTube channels and thin practitioner discussion, so I used it to test sentiment, not to prove the business.