Claudeforce and a 12% Pop: What Salesforce's Q2 Actually Tells Revenue Teams
Salesforce reported second quarter results on August 26 and announced an expanded partnership with Anthropic on the same day. The market reaction was immediate and large. For those of us who spend our time inside quote-to-cash, the interesting part is not the stock move — it's what the announcement says about where Salesforce is putting its engineering effort over the next two years.
The numbers, and what's actually in them
Revenue for the quarter ending July 31 was $11.345 billion, up roughly 11% year over year and about in line with expectations. Shares rose about 12.7% after hours. Underneath that, three things are worth separating.
The headline EPS beat is mostly an investment gain. Non-GAAP earnings came in at $5.90 per share against guidance of $3.25 to $3.27, and net income of $3.53 billion was up 87% from $1.89 billion a year ago. Salesforce attributed much of that to a $2.6 billion gain on strategic investments, tied in part to its own equity stake in Anthropic. That is a mark-to-market revaluation, not quoting volume. An 80% earnings beat built on it should be read accordingly.
The guidance raise is mostly acquisitions. Full-year fiscal 2027 revenue guidance moved up $300 million in constant currency to $46.1 to $46.4 billion. Roughly $100 million of that is organic strength in Agentforce, Data 360, and Slack; the other $200 million is attributed to the pending Contentful and Fin deals. Non-GAAP operating margin guidance did not move, holding near 34.3%.
The numbers that do reflect the business are the bookings and the AI ARR. Current remaining performance obligation reached $33.5 billion, up 14% in constant currency, which CFO Robin Washington characterized as the strongest quarter for net new annual order value in four years, with attrition near record lows. Agentforce ARR passed $1.5 billion, up 240%. Those are the figures to watch next quarter.
What Claudeforce is
Announced alongside earnings, Claudeforce runs in three directions at once. Claude becomes a default reasoning model inside Agentforce and the default model in Slack. And a plugin called Salesforce in Claude ships with 37 prebuilt sales skills — meeting prep, deal health review, pipeline review — letting a seller reason over live CRM data and push governed updates back without opening Salesforce. It's with pilot customers now, with an open beta planned for September.
Benioff's pitch on the earnings call was that AI models are only as useful as the enterprise data sitting behind them, and that Salesforce owns that data. The practical version of that claim: an agent is only as good as the data model it reasons over, and Salesforce is betting that owning the catalog, the pricing rules, the contract terms, and the billing schedule is a more durable position than owning the model.
Why this lands on the CPQ migration desk
If you are running legacy Salesforce CPQ and weighing your options, this announcement changes the calculus in one specific way, and leaves the rest untouched.
- It confirms where the investment is going. Agentforce Revenue Management is the platform receiving AI development. Legacy CPQ is in maintenance. Nothing announced this week suggests that changes.
- Agent value depends on your data model. This is the part most teams underrate. Legacy CPQ runs on managed-package custom objects. Agentforce Revenue Management runs on native Salesforce objects. Agents reason over native data. Staying on CPQ does not just freeze your feature set — it puts your revenue data outside the surface where the new capabilities operate.
- It does not create a deadline. Salesforce still has not published an end-of-support date for CPQ. A strong quarter is not a countdown clock. Anyone using this week's news to manufacture urgency in your buying process is selling, not advising.
The honest read
A good quarter is evidence about Salesforce's business. It is weak evidence about whether Agentforce Revenue Management is ready for your catalog complexity, your amendment logic, or your billing edge cases. Those questions get answered by a scoping exercise against your actual org, not by a guidance raise.
What we would take from this week: the destination is settled, the timeline is still yours to set, and the argument for moving is about capability access rather than fear of a shutoff. That is a much better position to plan from than the one most teams thought they were in eighteen months ago.
Sources
Quartz via Yahoo Finance, Q2 FY27 results · Salesforce newsroom, Claudeforce announcement
Planning your Revenue Cloud move?
We staff the architects and consultants who scope and run CPQ migrations — people who have done them before.