Salesforce Stock Q2 Earnings Analysis: CRM Jumped, Is “SaaS Apocalypse” Cancelled?

Salesforce beat revenue expectations by barely $15 million, or roughly one-tenth of 1%, yet CRM stock jumped about 13% after hours. 

The market was not celebrating a small headline beat. It was reacting to something more important: the first convincing evidence that AI may increase customer spending on Salesforce instead of making its software irrelevant.

Let's break down what changed in Salesforce’s Q2 FY27 earnings, why Agentforce’s refill data mattered more than its token count, how much growth came from Informatica, and whether the feared “SaaS apocalypse” is actually over.

Salesforce Q2 FY2027 Earnings Results: Revenue, EPS and Key Numbers

Salesforce reported its fiscal second-quarter results for the three months ended July 31, 2026.

The financial results come from the Salesforce Q2 FY27 earnings release. Consensus estimates are based on FactSet and Visible Alpha figures reported by The Wall Street Journal and other financial publications.

CRM stock had closed virtually flat at about $206 on August 26. It then climbed roughly 13% to around $232 in after-hours trading. According to MarketWatch, it was the stock’s largest post-earnings increase since August 2024.

Why Is Salesforce Stock Rising After Earnings?

The simple answer is that Salesforce beat expectations and raised guidance. The more useful answer is that several indicators moved in the right direction together.

Before earnings, options traders were pricing a move of about 7% in either direction. The actual after-hours gain of roughly 13% was almost twice that amount.

That suggests positioning also played a role. CRM stock was down about 22% for the year before the report, partly because investors feared that AI agents would replace traditional software subscriptions. When the results provided evidence against that theory, investors had to quickly reprice the risk.

Salesforce Agentforce Growth: Why Credit Refills Mattered Most

The strongest part of Salesforce’s quarter was the movement of Agentforce customers from experimentation to paid consumption. Agentforce is Salesforce’s platform for AI agents that can complete tasks such as answering customer questions, updating sales records or managing support requests.

The company sells some of this usage through Flex Credits. Customers buy credits, use them as their AI agents complete tasks, and purchase more when their credits run out.

Think of it like a prepaid mobile plan. Selling the first recharge proves that a customer is curious. Regular recharges prove that the customer actually needs the service.

Salesforce disclosed the following Agentforce adoption funnel during its Q2 earnings call:

An Agentic Work Unit, or AWU, represents a task completed by an AI agent. It is more useful than a token count because tokens measure how much information an AI model processes, while AWUs measure completed work.

Even AWUs do not automatically equal revenue. The best evidence comes when the sequence is complete:

Deployment leads to usage, usage consumes credits, and consumed credits lead to another purchase.

Salesforce said half of Agentforce bookings came from customers “refilling the tank.” That is probably the quarter’s most valuable disclosure. Agentforce’s annual recurring revenue also exceeded $1.5 billion, up from $1.2 billion in Q1. That suggests sequential growth of at least 25%.

However, investors should treat that comparison carefully. Salesforce expanded the Agentforce ARR definition in Q2 to include additional AI offerings, Slackbot and Headless 360. Part of the increase may therefore come from a broader definition rather than pure like-for-like growth.

Salesforce’s 11% Revenue Growth Needs an Acquisition Adjustment

Reported revenue rose by $1.11 billion, from $10.24 billion to $11.35 billion. However, Salesforce disclosed that Informatica contributed $456 million of Q2 revenue.

Removing that acquisition provides a rough view of underlying Salesforce growth.

This is not an official organic growth figure because it does not adjust for every currency, product or acquisition effect. However, it shows why the headline 11% growth rate should not be treated as entirely organic.

The same calculation can be applied to subscription and support revenue.

Salesforce’s core business is therefore still growing at a mid-to-high single-digit rate. It has not suddenly returned to double-digit organic growth.

That does not make the quarter weak. The important improvement was in leading indicators such as bookings, cRPO, attrition and contract duration. These can influence reported revenue over future quarters.

Salesforce cRPO Growth: Why the 14% Increase Mattered

Current remaining performance obligation, or cRPO, represents contracted revenue Salesforce expects to recognise over the next 12 months.

Think of Salesforce as a restaurant that already has reservations and deposits for future meals. Revenue records the meals served during the quarter. cRPO shows part of the business already booked for the coming year.

Salesforce’s cRPO reached $33.5 billion, up 14% both on a reported and constant-currency basis. This was one percentage point above management’s constant-currency guidance and slightly ahead of analyst expectations.

Management attributed the performance to Slack, Agentforce and Data 360. It also said:

• Net new annual order value growth was the strongest in four years.
• Net new order growth significantly exceeded total order-value growth.
• Customer attrition remained near historic lows.
• Contract duration improved across new business and renewals.
• Seats in Agentforce Sales, Service and Slack grew year-on-year.

These points directly challenge the “SaaS apocalypse” theory, under which companies reduce software seats because AI agents can do more work with fewer employees.

There is still a disclosure limitation. Salesforce did not separately report Informatica’s contribution to Q2 cRPO. Acquisitions can increase cRPO, so investors should not automatically assume the entire 14% growth rate was organic.

Still, the combination of stronger bookings, longer contracts and low attrition is much harder to dismiss than a simple revenue beat.

Salesforce Q2 EPS: Why the $5.90 Figure Overstates Core Profit

At first glance, Salesforce crushed the adjusted EPS estimate of $3.27 by reporting $5.90. That comparison is misleading.

Salesforce recorded a $2.6 billion gain on its strategic investments, largely associated with the increased value of its Anthropic investment. According to Reuters, this added approximately $2.53 per share to adjusted EPS.

After removing the investment gain, Salesforce beat the consensus estimate by about 10 cents, or 3%. That is a healthy beat, but nowhere close to the 80% headline difference between $5.90 and $3.27.

There is one more adjustment investors need to understand: Salesforce’s diluted share count fell from 962 million to 821 million, a decline of 14.7%.

Using simplified calculations:

Underlying adjusted EPS increased roughly 16%, but total adjusted profit excluding the investment gain was approximately flat. This happened because the same profit was divided across far fewer shares.

Imagine a pizza that remains almost the same size but is divided into fewer slices. Every remaining slice becomes larger even though the pizza itself has not grown.

The reduced share count is valuable to continuing shareholders, but it should not be confused with operating profit growth.

Salesforce FY2027 Guidance: Revenue, cRPO and Margin Outlook

The guidance bridge is important:

• Salesforce added $100 million for better organic performance.
• It added $200 million from the pending Contentful and Fin acquisitions.
• A stronger dollar removed $100 million.
• The net reported increase was therefore $200 million.

The organic increase is more encouraging than the overall $200 million headline. It indicates that management sees stronger demand in Agentforce, Data 360 and Slack even before the pending acquisitions contribute.

However, Q3 revenue growth guidance of 11%-12% still includes slightly more than four percentage points from Informatica. That implies an underlying growth rate closer to 7%-8%, based on a simple subtraction.

Salesforce has promised that organic revenue growth will begin accelerating in Q3. The guidance supports that direction, but the expected acceleration remains gradual.

What Marc Benioff Said About the “SaaS Apocalypse”

Salesforce CEO Marc Benioff used the earnings call to directly challenge concerns that AI will destroy traditional software companies.

Benioff said it was time for the “SaaS apocalypse” argument to stop. His case was built around four observations:

1. Salesforce seats did not collapse.
2. Customer attrition remained near historic lows.
3. Premium Agentforce bookings more than doubled.
4. AI models increasingly accessed Salesforce data and workflows.

Management also said nine of the ten largest AI companies use Salesforce and Slack, with their combined spending increasing 435% year-on-year.

The strongest part of Benioff’s argument is not that Salesforce’s traditional interface will remain unchanged. It probably will not. The stronger argument is that AI models still require trusted customer data, permissions, business rules and workflows before they can safely act inside a company.

This is where the new Anthropic partnership becomes strategically important.

What Is Claudeforce and Why Does It Matter?

Salesforce and Anthropic announced Claudeforce alongside the earnings results.

Its first product, Salesforce in Claude, places Salesforce’s data and business actions directly inside Anthropic’s Claude interface. It launches with 37 prebuilt sales skills covering tasks such as meeting preparation, deal reviews and pipeline updates, according to the official Claudeforce announcement.

This reveals Salesforce’s strategy for surviving AI disruption.

Salesforce is accepting that Claude or another AI assistant may become the interface employees use. Its goal is to remain the trusted engine underneath that interface.

A simple analogy is a restaurant and a food-delivery app. The customer may interact with the delivery app, but the restaurant still prepares the meal. Salesforce wants Claude to take the order while Salesforce supplies the data, permissions and business process that complete the work.

This can defend Salesforce’s relevance even if employees spend less time clicking through its screens. It may also create more API calls, data usage and AI consumption.

The risk is that the company controlling the interface can gain greater influence over customer relationships and pricing. If Claude becomes the place where work happens, Salesforce could become valuable but less visible infrastructure.

Claudeforce is therefore both a defence and an admission. Salesforce is defending its underlying platform while admitting that the traditional software interface may no longer be the centre of enterprise work.

Salesforce’s Buyback Was Well Timed, but Debt Has a Cost

Salesforce financed a $25 billion accelerated share repurchase and expects the programme to retire at least 14% of its outstanding shares.

Management said shares purchased so far had an average cost of approximately $176. The post-earnings price of around $232 was roughly 32% above that cost, making the timing appear favourable.

The buyback has already increased EPS per remaining share. But it was financed primarily with debt.

Interest expense increased from $67 million to $473 million and consumed approximately 20% of Q2 GAAP operating profit.

This explains why Salesforce continues to guide for only 4%-5% full-year free-cash-flow growth despite reporting an 81% quarterly increase. The quarterly figure benefited from timing, while the annual outlook reflects the cost of the additional debt.

The buyback appears to have created value because Salesforce purchased shares below the current market price. However, future analysis should account for interest costs, total debt and free cash flow rather than looking only at EPS.

What Went Right for Salesforce?

What Still Needs Improvement for Salesforce?

The quarter was better, but it did not resolve every concern.

First, estimated revenue growth excluding Informatica was only about 6.4%. Salesforce still needs to show that stronger bookings can push core revenue growth towards sustainable double digits.

Second, the faster-growing Data 360, Headless Platform and Other category includes Informatica. It also combines Data 360 with Tableau, MuleSoft and other products, making it difficult to identify which businesses are accelerating.

Management acknowledged continued licence-revenue headwinds and volatility in analytics and integration. It also said Marketing showed early signs of recovery but that it was too soon to call the improvement sustainable.

Third, Salesforce expanded the definition of Agentforce ARR during the quarter. Changing definitions can make sequential comparisons appear stronger, so investors should place greater weight on production deployments, consumption and credit refills.

Finally, adjusted operating margin fell slightly to 34.1%, while GAAP operating margin dropped from 22.8% to 20.5%. Acquisitions, amortisation, stock compensation and interest costs remain meaningful.

How Did Salesforce Answer Our Pre-Earnings Questions?

The quarter therefore moved Salesforce from an “AI promise” story towards an “AI monetisation” story. It has not yet completed the transition into an organic double-digit growth story.

What Should CRM Stock Investors Watch Next?

The next few quarters should be judged using five indicators.

1. Organic revenue growth: Reported growth will remain influenced by Informatica, Fin and Contentful. Investors need clear disclosure showing whether the original Salesforce business is accelerating.

2. Agentforce refill rate: A rising share of bookings from customers purchasing additional credits would show that Agentforce is becoming a repeat-use product.

3. Consistently consuming customers: Salesforce added 2,000 paying production customers, but said only hundreds had reached the consistently consuming stage. Closing that gap will determine how much ARR becomes recognised revenue.

4. Operating margin and AI costs: AI usage can create infrastructure and model costs. Salesforce must demonstrate that consumption growth does not weaken its 34.3% adjusted margin target.

5. Debt and interest expense: The buyback has reduced the share count, but debt has increased sharply. Free cash flow should increasingly be evaluated after considering interest costs and future acquisition spending.

Salesforce Stock Q2 Earnings Analysis: Final Take

Salesforce’s Q2 results were not a blockbuster because revenue exceeded estimates by $15 million or because adjusted EPS reached $5.90. Those headlines exaggerate what happened.

The real strength came from the operating evidence underneath them: cRPO accelerated, customer attrition remained low, contract duration improved, net new orders recorded their strongest growth in four years, and half of Agentforce bookings came from customers purchasing more credits.

This was Salesforce’s strongest evidence so far that AI can expand customer spending rather than simply replace traditional software licences.

Our view is that the market was justified in reducing the “SaaS apocalypse” discount attached to CRM stock. However, the 13% rally should not be interpreted as proof that Salesforce has already returned to high organic growth.

Estimated revenue growth excluding Informatica was about 6.4%. Underlying adjusted profit was approximately flat after removing the investment gain and share-count effect. Full-year free-cash-flow growth is still expected to be only 4%-5%.

The Salesforce story has therefore improved, but it remains a transition.

The next phase depends on whether the Agentforce consumption flywheel can move from deployments to regular use, from regular use to credit refills, and from credit refills to double-digit organic revenue growth. Q2 proved that the flywheel has started turning. It has not yet proved how fast it can spin.