Loading today’s stories
Invest1 distinct publisher3 min readPublished
A 23% one-day move rewarded an EPS figure that doubled while operating income stayed flat, and the arithmetic behind it runs through a $2,613M investment mark-up and 141 million shares retired on borrowed money.
The Investor · Invest desk
Compiled by The InvestorSomething wrong?How this is made
Operating income was $2,331M against $2,332M a year earlier, flat on 11% more revenue [9], which means the $2,613M of gains on strategic investments out-earned the entire operating business by about 12% [1]. Strip the gains and non-GAAP net income was $2,767M versus $2,795M, down 1%, with ex-gains EPS of roughly $3.37 against $2.91, and every point of that 16% came from the smaller share count [12]. The operating result was flat and the reported result doubled, and the gap between the two is a mark plus a denominator.
That denominator carries a coupon. The year-over-year increase in interest expense is $406M in a single quarter, or roughly $1.6B annualised if the rate holds, so the $2,613M gain covers about nineteen months of the added interest and nothing after that [2]. A mark can be written back down next quarter; the interest schedule stays where it is. Benioff's own framing on the call was that Slack plus the Anthropic stake "has been like half our value" [13], and the strategic portfolio went from $7.6B to $11.3B in six months, of which this quarter's gain accounts for 71% [14][6]. Most of the six-month move, in other words, landed inside the reporting period that got the 23% [1].
The forward book is the part worth defending, or rather, the part where the bull case is actually made: cRPO growth ran 10%, then 13%, then 14%, and the roughly 14% Q3 guide explicitly excludes Contentful and Fin [6], while Robin Washington said the entire company was compensated this year on net new annual order value outpacing total AOV, which is the mechanical precondition for organic reacceleration in the second half [19]. Two things sit against that. Informatica's contract base is inside cRPO, so part of the acceleration was purchased [16], and management said contract length improved across all segments, which should have pushed value into the noncurrent bucket and did not [17]. Marketing showed "early signs of recovery" with management's own caveat that it is "too soon to call this a sustainable trend" [20]. FY27 guidance went to $46.1B-$46.4B [5].
This is probably wrong in one specific direction, so name it: if net new AOV genuinely is outpacing total AOV, then 6.4% organic [7] is a lagging artefact of deals signed a year or two ago, the market bought the 2027 revenue line rather than the 2026 earnings line, and 23% was cheap rather than credulous. My read is the middle case. The order book is genuinely better, the earnings genuinely are not, and a 103% EPS print [4] that survives no arithmetic at all is a thin thing on which to re-rate a sector.
Ranked by verification strength, evidence, and original report placement.
Salesforce booked $2,613M in gains on strategic investments in the quarter, which added $2.53 to non-GAAP EPS and $2.43 to GAAP EPS, against $0.00 in the prior year.
Diluted share count fell from 962M to 821M, down 14.7%, on a $25B accelerated share repurchase funded with $24.8B of new debt.
Benioff on what drove the gain: "Take the value of Slack. Then take the value of our Anthropic stock. That has been like half our value."
The strategic investment portfolio went from $7.6B to $11.3B in six months.
Interest expense went from $67M to $473M in one year, and noncurrent debt went from $10.4B to $39.3B.
Atlassian got a 35% one-day move on 44% RPO growth three weeks earlier; Salesforce got 23% on 14%, and the market is pricing the order book, with the current-versus-noncurrent split the part that gets read.
Distinct publishers with included, body-backed reporting in this cluster.
Follow any of these and your For You feed starts watching them — no settings page required.
invest
Salesforce's 14% cRPO is the best agent-demand read yet. Organic growth is 6.5%2 distinct publishers
invest
Salesforce bought $27.1B of its own stock in a quarter it grew 13%2 distinct publishers
product
Salesforce's longer-dated backlog grows at half the rate of its cRPO headline1 distinct publisher
product
Salesforce's answer to the SaaSpocalypse: 11% growth, and a $2.6B mark on Anthropic1 distinct publisher
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Reproducible from the issuer's own filings, interpreted by one voice
Nearly every figure that matters - the $2,613M gain, the $2.53 per-share contribution, 962M shares becoming 821M, $67M of interest becoming $473M, Informatica's $440M of subscription revenue - is a disclosed number, and SaaStr is explicit that the organic-growth subtraction took four minutes with the segment table. That makes the arithmetic checkable rather than assertive. What is missing is a second reader: no other outlet in our coverage has run the same bridge, and the valuation behind the Anthropic mark is nowhere examined.
Booked demand strong, agent deployment still shallow
There are two adoption stories and they disagree in scale. The order book is real: $33.5B of current obligations growing 14%, bookings doubling for Agentforce, 3.2 billion work units. The installed footprint is not yet: about $3.9B of combined AI and data ARR against a $45B base, a production base implied to be in the low thousands of accounts, and half of new bookings coming from customers topping up credits rather than new commitments. Part of the 14% arrived by acquisition rather than by sale.
The headline oversells; the order book underlies it
"SaaSpocalypse over" and a doubled EPS line are what the market paid 23% for, and the operating business earned one million dollars less than it did on $1.1B less revenue. That is a gap, and it is large. It is not total: the forward book did accelerate to 14% from 10% two quarters earlier, which is a genuine improvement the arithmetic critique does not erase. The overstatement lives in the framing - a capital-allocation quarter presented as an operating turnaround - not in the underlying disclosure, which Salesforce made complete enough for anyone to catch it.
The issuer marks its own scorecard, twice over
Follow the money in this quarter and it loops back on itself. The gain that supplied 43% of EPS is a revaluation of private holdings the company itself carries. The metric management highlighted as the leading indicator - net new order value outpacing total - is, by Washington's own statement, the metric the whole company was paid on this year. And the buyback that produced the entire residual per-share increase was a board decision funded with debt whose cost lands in later quarters. On the reporting side, SaaStr's franchise runs on finding the number behind the number for a SaaS-executive audience, which cuts toward scepticism rather than toward Salesforce.
Solid on the past quarter, thin on what follows
We would stand behind the backward-looking arithmetic without much hedging - it is issuer-disclosed and it reconciles. Confidence drops on everything forward: whether the twelve-month book converts into second-half organic growth rests on management's framing alone, the private mark could move in either direction next quarter, and with a single publisher in our coverage there is no independent read to triangulate the interpretation against.