This is investment research, not personal financial advice.
A $38 billion day, with three different engines
Salesforce, Inc. (NYSE:CRM) closed at $252.05 on 27 August, up 22.58%. The move followed a fiscal second-quarter result with 11% revenue growth, 14% growth in current remaining performance obligations and a higher full-year outlook. The SEC's company record confirms the listed legal name and ticker. Applied to CNBC's 819 million-share snapshot, the $46.43 gain added about $38.0 billion of equity value and left the company with a $206.4 billion market value (SEC 2026; CNBC 2026).
The quarter gave the market a reason to reject the cleanest version of the software-disruption argument. Subscription and support revenue rose 12%. Agentforce annual recurring revenue, or ARR, reached $1.5 billion, up more than 240%. Agentforce plus Data 360 ARR approached $3.8 billion. An independent same-day report described investors reading the result as evidence against an imminent collapse in subscription software economics (Salesforce 2026a; Motley Fool 2026).
But the 11% headline joined three different engines. Informatica contributed $456 million of quarterly revenue, equal to four percentage points of growth. Foreign exchange added less than one point. The old business therefore grew about 7%. The cRPO figure has the same issue: 14% reported growth becomes about 9% after removing the five-point Informatica contribution, with currency having little effect (Salesforce 2026a; Salesforce 2026b).
Then there is the capital structure. In March, Salesforce issued $25.0 billion of senior notes and used the $24.8 billion of net proceeds to fund an accelerated share repurchase. It paid $25.0 billion up front and received 103 million shares at an initial average of $198.34. That cut the quarterly basic share count by 14% year on year, but lifted debt from $14.4 billion in January to $39.3 billion in July (Salesforce 2026b).
The reaction was directionally justified. Organic bookings accelerated, the AI products are becoming material, and fewer shares give each remaining share a larger claim on future cash. The full 22.6% move looks roughly proportionate only if owner cash can grow through the higher interest bill and if Agentforce keeps the core business above its recent mid-single-digit pace. The market rewarded an operating improvement and a financial-engineering step at the same time. They should not be confused.
The 11% growth rate was 7% before the deal
Salesforce has one operating segment, but its economics come from two revenue streams. Subscription and support supplied 95% of first-half FY2027 revenue. Customers pay to use cloud applications and related support, usually through annual invoices. Professional services supplies the smaller balance and helps with implementation, although its revenue fell 4% in Q2 as demand for large, multi-year transformation work softened (Salesforce 2026b).
The product presentation changed this year. Salesforce now groups subscription revenue into Agentforce Apps and a second group called Data 360, Headless Platform and Other. Agentforce Apps includes the sales, service, marketing, commerce and Slack applications with the agent layer embedded. The data group includes Data 360, Informatica, MuleSoft, Tableau and other platform products. The new labels make strategic sense, but they also reduce continuity with the old Sales, Service, Platform, Marketing and Integration disclosures.
Agentforce Apps produced $7.19 billion of Q2 subscription revenue, up 8%. Data 360, Headless Platform and Other produced $3.62 billion, up 20%. The latter contains Informatica, so its growth is partly acquired. The filing discloses $456 million of Informatica revenue for the quarter and $900 million for the first half, but not the acquired business's operating profit or cash contribution. Expense contributions are no longer separately identifiable because the operations have been integrated (Salesforce 2026b).
Subtracting Informatica from both periods leaves Q2 revenue of roughly $10.89 billion, up about 6.4% from the prior year's $10.24 billion. The company's rounded release describes the organic rate as 7%. That is better than a stalled franchise. It is not the same as an 11% acceleration generated by Agentforce.
Bookings carry more encouragement. cRPO covers contracted revenue expected within 12 months. It rose 14%, or 9% organically, against management's prior expectation of 9% to 9.5% organic growth. Total remaining performance obligation reached $72.4 billion from $66.3 billion at January. That is a large contracted base, though the 10-Q warns that RPO depends on contract duration, renewals, seasonality, currency and acquisitions. It is not a cash receipt and does not guarantee the same growth rate in recognised revenue.
The distinction shapes the commissioning question. The event says the core business improved, but the reported rates overstate the size of that improvement. Q1 had already established the gap between reported and organic growth as Informatica entered the comparison (Salesforce 2026c). A second quarter of at least 8% organic cRPO growth would support a new underlying pace. A drop back below that level would make Q2 look more like a good renewal window combined with acquired revenue.
Salesforce's moat is broad workflow history, not the word "agent"
The durable part of Salesforce's position is the customer information and workflow history held inside its applications. Sales teams record opportunities, service teams handle cases, marketers define audiences and commerce systems process transactions. Replacing one cloud can disturb data models, integrations, permissions, reports and employee routines across several departments. Those costs support renewal even when a competing product has a lower sticker price.
A second loop comes from breadth. Data 360 can connect customer information; MuleSoft links outside systems; Tableau analyses the result; Slack carries the conversation; the applications contain the workflow. Agentforce attempts to turn that data and those permissions into automated work. If the products operate together, an added module can improve the value of the installed set and reduce the number of vendors a customer manages.
There is financial support for that loop. Subscription revenue remains 95% of the group. Organic cRPO grew 9%. Agentforce ARR reached $1.5 billion and the combined Agentforce and Data 360 figure approached $3.8 billion. Yet Agentforce remains about 3% of the company's $46.1 billion to $46.4 billion FY2027 revenue range. Its growth can matter without being large enough to carry the group by itself (Salesforce 2026a).
The counter-evidence sits beside the growth numbers. Pricing was not a significant driver of either quarterly or first-half revenue growth. Subscription cost rose faster than subscription revenue because of service delivery, hosting, generative AI and acquired intangible amortisation. Professional services demand weakened. A strong moat can support renewal and cross-selling without giving unlimited price power or cost-free AI delivery (Salesforce 2026b).
Peer results also limit the exclusivity claim. ServiceNow reported $3.99 billion of Q2 revenue, up 24%, and $29.0 billion of total RPO at June. Its workflow platform gives large enterprises another path for data, automation and agents. Microsoft, Oracle and specialist application vendors add other routes. Salesforce has distribution and embedded data; it does not own enterprise automation as a category (ServiceNow 2026).
The moat is stable in customer workflow and distribution, widening where the data and application layers connect, and less clear where product breadth is purchased. Informatica can improve Data 360, but a $900 million acquired revenue contribution does not prove an attractive return. The integration needs to create faster organic bookings, better retention or lower cost. Until those effects appear, acquisition breadth is capability bought with capital rather than evidence of a stronger loop.
Five years changed the operating return
The longer history gives the quarter useful scale. Revenue rose from $26.49 billion in FY2022 to $41.53 billion in FY2026. Operating cash flow rose from $6.00 billion to $15.00 billion. GAAP net income moved from $1.44 billion to $7.46 billion as Salesforce cut costs and lifted operating margin. The annual rows below come from the original 10-K filings; the Q2 FY2027 row is the filed six-month period, not a full year (Salesforce 2022; Salesforce 2023; Salesforce 2024; Salesforce 2025; Salesforce 2026d; Salesforce 2026b).
| USD millions unless stated | FY2022 | FY2023 | FY2024 | FY2025 | FY2026 | Q2 FY2027, six months |
|---|---|---|---|---|---|---|
| Revenue | 26,492 | 31,352 | 34,857 | 37,895 | 41,525 | 22,478 |
| Net income | 1,444 | 208 | 4,136 | 6,197 | 7,457 | 5,633 |
| Operating cash flow | 6,000 | 7,111 | 10,234 | 13,092 | 14,996 | 7,970 |
| Author-computed ROIC | 1.0% | 0.5% | 6.8% | 9.9% | 10.3% | 10.7% annualised |
| Author-computed net debt / (net cash) | 5,132 | 3,585 | 954 | (415) | 7,112 | 30,978 |
| Author-computed interest cover | 2.5x | 3.4x | 17.7x | 26.5x | 25.7x | 5.9x |
The ROIC calculation uses after-tax GAAP operating income divided by average invested capital. Invested capital is stockholders' equity plus debt minus cash. The tax rate is the filed tax provision divided by pretax income for each period. Marketable securities remain in the denominator because the filed annual tags are not consistent enough across all five original reports for a clean like-for-like subtraction. The Q2 ratio doubles six-month NOPAT and is only an annualised proxy.
Interest cover is also author-computed, using GAAP operating income divided by interest expense on debt. The Q2 ratio uses the filed six-month amounts without annualising either input. Net debt is debt less cash; it is shown for the balance-sheet change but omitted from the machine-readable history because it is a calculation rather than a directly reported line.
The progression matters more than the decimal. Salesforce earned about 1% on this broad capital base in FY2022 and 0.5% in FY2023. By FY2026, the proxy reached 10.3%. Operating margin, not faster sales growth, did most of the work. The incremental return from FY2023 to FY2026 was high because operating profit rose by $7.30 billion while end invested capital increased only $4.31 billion. That interval also contains restructuring and changing tax rates, so it is evidence of a margin reset rather than a steady-state reinvestment rate.
Q2's reported net income needs one adjustment before it joins that trend. The quarter included $2.61 billion of net gains on strategic investments, primarily a $2.7 billion unrealised Anthropic gain. Interest expense was $473 million, up from $67 million. Net income rose to $3.53 billion, but the investment mark was larger than the year-on-year increase in profit. The operating line was almost flat at $2.33 billion despite 11% revenue growth because gross margin and acquired-cost pressure absorbed the added sales (Salesforce 2026b).
That makes the current return picture mixed. Five-year margin work is real. Q2's profit headline contains a large market-value gain, and the new debt raises the recurring interest charge. The next full year will show whether the improved ROIC survives both items.
The buyback transferred timing risk to the balance sheet
The March transaction was not an ordinary use of surplus cash. Salesforce sold $25.0 billion of senior notes with maturities from 2028 to 2066, received $24.8 billion after issuance costs, and directed those proceeds to the accelerated share repurchase. The initial 103 million shares arrived at $198.34. Another part of the contract remains unsettled; the final number depends on Salesforce's share price during the measurement period and is expected in Q3 (Salesforce 2026b).
That structure has a clear arithmetic benefit. Q2 basic shares averaged 820 million, down from 956 million a year earlier. Diluted shares averaged 821 million. At the $252.05 close, each retired share represents a claim worth more than the initial ASR price. The company also repurchased 11 million shares in the open market during Q1 for $2.15 billion at an average $192.00.
The cost is less visible in per-share profit. Debt reached $39.29 billion at July, against $8.31 billion of cash. Principal due within the contractual table rose to $39.5 billion from $14.5 billion at January. Interest expense reached $790 million in the first half, up $655 million. Annualising the Q2 run rate produces roughly $1.9 billion, before refinancing or cash-balance effects. Salesforce had a $5.0 billion undrawn revolving facility, so current liquidity is adequate. The constraint is the recurring claim on operating cash, not near-term survival.
There is also a capital-allocation comparison. The ASR's initial delivery price of $198.34 was below the post-results close, but above the severe-downside and bear scenario ranges in this article. A repurchase creates value per remaining share only when the retirement price is below the long-run value of the cash claim surrendered. Borrowing makes that test harder because interest continues even if the share price falls.
Management still had $22.9 billion of repurchase authority at July. It also paid $747 million of dividends in the first half and has $8.32 billion of future stock-compensation expense attached to outstanding awards. Those awards are not all issued shares yet, but they show why a 14% cut in the count is not a permanent one-for-one reduction. Q2 excluded 42 million shares linked to the unsettled ASR and 40 million employee awards from diluted EPS because they were antidilutive (Salesforce 2026b).
The anti-thesis to the capital-allocation concern is simple: a company with recurring revenue, $15 billion of prior-year operating cash and growing owner cash can carry the debt, and the shares were retired at less than the new market price. That may prove correct. The burden of proof has shifted. Future per-share cash must exceed the interest cost and employee-equity replenishment, not merely show a lower denominator this quarter.
Cash flow still pays an equity bill
Salesforce generated $14.996 billion of operating cash in FY2026 and spent $594 million on property and equipment. The simple cash measure after capex was $14.402 billion. Stock compensation was $3.509 billion. Subtracting all three leaves $10.893 billion of author-computed owner cash. This is a stricter measure than company-reported free cash flow because it treats employee shares as compensation paid by owners.
| Owner-cash bridge, USD millions | FY2022 | FY2023 | FY2024 | FY2025 | FY2026 | H1 FY2027 |
|---|---|---|---|---|---|---|
| Operating cash flow | 6,000 | 7,111 | 10,234 | 13,092 | 14,996 | 7,970 |
| Less capex | (717) | (798) | (736) | (658) | (594) | (316) |
| Less stock compensation | (2,779) | (3,279) | (2,787) | (3,183) | (3,509) | (1,763) |
| Author-computed owner cash | 2,504 | 3,034 | 6,711 | 9,251 | 10,893 | 5,891 |
The bridge shows a substantial improvement. Owner cash more than quadrupled in four years because operating cash rose faster than capex and stock compensation. First-half FY2027 owner cash reached $5.89 billion, though cash collection is seasonal and should not be doubled mechanically.
Management's FY2027 range implies another test. Operating cash flow growth of 4% to 5% would place the full-year figure around $15.6 billion to $15.7 billion. If capex stays near $700 million and stock compensation remains around 9% of the $46.1 billion to $46.4 billion revenue range, owner cash would be roughly $10.7 billion to $10.9 billion. That is approximately flat against FY2026 before allowing for a lower share count.
Per-share owner cash could still rise by about 15% if the year-end diluted count remains near 820 million. That is the intended transaction outcome. Yet the group would have reached it through financial structure rather than faster aggregate cash generation. The distinction affects duration. Operating growth can compound; debt-funded share retirement happens once and leaves interest behind.
What $252 requires from the next three years
Salesforce's $206.4 billion equity value equals about 19.0 times FY2026 owner cash of $10.893 billion. Because that cash measure is after interest and stock compensation, the valuation can be built directly on the equity cash stream without adding cash or subtracting debt again. The market's 819 million shares also align closely with the post-ASR Q2 count (CNBC 2026; Salesforce 2026b).
A reverse cash-flow calculation gives the first cross-check. Starting with $10.893 billion, a ten-year model with a 9% discount rate and 2% terminal growth reaches about $225 per share if owner cash grows 4% a year, and $261 if it grows 6%. The current close therefore embeds roughly 5.5% to 6% annual owner-cash growth for a decade, with no new dilution beyond the current share base.
The discount rate is not generous. The 10-year US Treasury constant maturity rate was 4.66% on 26 August. A 9% equity rate allows roughly 4.3 percentage points for business and equity risk above that government benchmark (Federal Reserve 2026). The largest sensitivities are owner-cash growth and the terminal multiple; a one-point change in either can move the equity value by tens of dollars per share.
The four frontmatter scenarios use a simpler three-year exit model. Each takes FY2030 owner cash, divides by the estimated diluted share count, applies a terminal multiple, then discounts three years at 9%.
| Case | FY2030 owner cash | Diluted shares | Terminal multiple | Present value per share |
|---|---|---|---|---|
| Severe downside | $8.5B-$9.5B | 830M-850M | 12x-14x | $93-$124 |
| Bear | $10.5B-$11.5B | 815M-830M | 15x-17x | $147-$185 |
| Base | $13.5B-$14.5B | 790M-810M | 17x-19x | $219-$269 |
| Bull | $16.5B-$18.0B | 760M-790M | 20x-22x | $323-$402 |
The severe-downside range assumes acquisitions fail to lift the core, organic growth falls below 5%, interest absorbs more cash and the share count rebuilds. The bear range assumes owner cash stays near the FY2026 level and the current repurchase mostly offsets later grants. The base range requires low-to-mid-single-digit owner-cash growth plus some further share retirement. It includes the $252 close, but near the upper half. The bull range needs Agentforce and Data 360 to restore high-single-digit organic growth, operating costs to scale and the share count to keep falling without another large debt step.
The price does not require a return to 20% group growth. It does require the 9% organic cRPO result to become a floor rather than a peak, and owner cash to grow after the interest bill. That is more measured than the stock's 22.6% one-day move suggests, but it still leaves little room for the acquired revenue to mask a slowing core.
The anti-thesis is a slower, better software company
The strongest case against this article's caution is not that Salesforce will become a hyper-growth company again. It is that the company can compound per-share cash at a useful rate while group revenue grows only 7% to 9%. The installed base is large, recurring revenue dominates, operating margin has already improved and the ASR reduced the denominator. Agentforce does not need to replace every application; it only needs to deepen enough customer relationships to keep organic bookings above revenue growth.
Q2 supports part of that argument. Organic cRPO reached 9%. Agentforce ARR grew above 240%. Data 360-related revenue grew faster than the application group. ServiceNow's 24% quarterly growth also indicates that enterprise workflow spending has not disappeared. The popular "software ends" thesis is too broad for the evidence (Salesforce 2026a; ServiceNow 2026).
The harder objection is cost. Data 360 and Agentforce delivery requires hosting and model consumption. Informatica added amortisation and integration expense. Interest rose by $406 million year on year in Q2. Research and development expense increased 14% for the quarter. Salesforce expects R&D to remain broadly steady as a percentage of revenue while it invests in AI and Informatica integration (Salesforce 2026b).
There is a management issue as well. The quarter's new product language arrived alongside a recast revenue presentation, acquired growth and an unrealised investment gain. None is improper. Together they make the headline cleaner than the economics. A careful reading needs organic growth, operating profit before investment marks and cash after stock compensation. The company now supplies enough disclosure to calculate each, but management's preferred narrative does not lead with them.
That leaves the reaction verdict near the middle. The market was right to revise the probability of a severe software disruption downward. It was also right to recognise the per-share effect of retiring 103 million shares. The closing value assumes those two facts become recurring owner-cash growth. Q2 alone proves an improved bookings quarter and a smaller share count, not the full duration.
Three reporting dates will settle the argument
The first date is the Q3 FY2027 report. It should disclose the final ASR settlement, the resulting share count and another organic cRPO comparison. An organic rate of at least 8% would keep the acceleration intact as the Informatica anniversary approaches. A lower rate would separate the strong headline from the continuing core.
The second date is the FY2027 10-K. It will contain a full year of interest expense, operating cash, capex and stock compensation. Owner cash above $11 billion would show that the business absorbed the new debt without losing the FY2026 cash base. A figure below that threshold would mean per-share improvement came almost entirely from the denominator.
The third window runs through FY2028, when Agentforce has a larger comparison base and Informatica should be integrated. Agentforce ARR below $2.0 billion at the FY2027 year end would leave it too small to explain a durable group step-up. Organic Data 360 growth, subscription gross margin and interest coverage will show whether the combined platform created operating return rather than only revenue.
At $252.05, the market is pricing Salesforce as a slower company whose margin work, AI cross-selling and lower share count can deliver roughly 6% owner-cash growth for a decade. The next two filings will show whether that cash comes from the operating system or from a transaction whose interest bill has only begun.
Source notes
Confidence is high on the filed history, Q2 acquisition contribution, debt, ASR terms and closing market data. The SEC filings and company releases were downloaded and read during this run, and the closing quote was reconciled to the Finance API sidecar. The Finance API's event list stopped at May, so the 26 August 8-K and 27 August 10-Q were taken from EDGAR rather than inferred from the normalized packet.
The main missing information is product-level profit and cash flow. Salesforce does not disclose the operating contribution of Informatica, Agentforce or Data 360, and it no longer presents the old cloud categories. The scenario ranges therefore treat group owner cash as the decisive output instead of assigning unsupported margins to each product. The media source independently confirms the reaction and its timing; all event numbers come from the filed release and 10-Q.
References
- Federal Reserve (2026), 10-year US Treasury constant maturity rate, 26 August.
- Motley Fool (2026), Here's Why Salesforce Stock Soared Today, carried by Yahoo Finance, 27 August.
- CNBC (2026), Quote snapshot for Salesforce, Inc. (CRM), 27 August.
- Salesforce (2022), Form 10-K for the year ended 31 January 2022.
- Salesforce (2023), Form 10-K for the year ended 31 January 2023.
- Salesforce (2024), Form 10-K for the year ended 31 January 2024.
- Salesforce (2025), Form 10-K for the year ended 31 January 2025.
- Salesforce (2026a), Q2 FY2027 results release, 26 August.
- Salesforce (2026b), Form 10-Q for the quarter ended 31 July 2026, filed 27 August.
- Salesforce (2026c), Q1 FY2027 results release, 27 May.
- Salesforce (2026d), Form 10-K for the year ended 31 January 2026.
- SEC (2026), Company filings record for Salesforce, Inc. (CRM).
- ServiceNow (2026), Form 10-Q for the quarter ended 30 June 2026.