SAP Cloud Backlog Growth Hits 26% as AI Push Offsets Profit Trim

SAP delivered a second quarter that underscored the accelerating shift to cloud-based AI software, even as acquisition costs forced a cut to its full-year profit target.

Cloud revenue jumped 24% to €6.3 billion and the all-important current cloud backlog (CCB) grew 26% to nearly €23 billion, a pace that reversed two quarters of relative sluggishness and beat analyst expectations. Total revenue rose 11% to €9.9 billion. But non-IFRS operating profit grew only 9% to €2.7 billion, decelerating from the previous quarter, and SAP lowered its 2026 operating profit guidance to a range of €11.8 billion to €11.2 billion, down €0.08 billion from its prior outlook, to reflect the dilutive impact of the Dremio and Prior Labs acquisitions.

“Q2 was an outstanding quarter,” Chief Executive Christian Klein said on a conference call. “Current Cloud Backlog grew 26%, an acceleration compared to Q1. After two quarters where our CCB was lagging behind cloud revenue growth, it was a welcome trend reversal.”

The results were powered by the May Sapphire event, where SAP launched its Business AI Platform, Joule Work, and a suite of autonomous agents. Klein said more than 90% of the 50 largest deals in the quarter embedded AI and SAP Business Data Cloud, evidence that enterprises are linking AI adoption to data modernization.

MetricQ2 2026YoY Growth (cc)Total Revenue€9.9B+11%Cloud Revenue€6.3B+24%Software LicensesN/A-32%Non-IFRS Op Profit€2.7B+9%Current Cloud Backlog~€23B+26%Cloud ERP Suite (share)88% of cloud+27%

CFO Dominik Asam noted that the Reltio acquisition, which closed on May 7, contributed less than one percentage point to CCB growth. The company maintained its top-line guidance, but adjusted operating profit sparked questions from analysts. Morgan Stanley’s Adam Wood pressed Asam on why the profit number missed expectations after a recent run of beat-and-raise cycles. Asam called Q2 “a little bit of a special quarter” with an unusual concentration of factors: heavy R&D investment, marketing spend around Sapphire, lower cloud revenue growth comps from the prior year, and the dilutive impact of Reltio. “Don’t read too much into that,” he said. As for the full-year cut, he explained that the company chose not to absorb the organic upside to offset M&A dilution, opting instead to preserve room for aggressive growth investment.

Klein emphasized that AI investments are already paying off. Developer productivity gains have reached 30%, and hiring plans for the rest of the year are being scaled back. “There is nothing which now stops us with the ability to switch models to always choose the best model for an ideal price outcome,” Klein said, referring to SAP’s strategy of flexibly using large language models from providers like Anthropic, Cohere, and Mistral AI. The company is also internally rolling out Joule Work Desktop to 4,000 test users, with a broad launch expected in Q3.

AI Platform and the “Autonomous Enterprise”

The Business AI Platform, built on three pillars—build, context and reason, and run and govern—emerged as the quarter’s signature innovation. Klein detailed how the recent acquisitions fit in: Dremio gives agents real-time access to non-SAP data without copying, Reltio governs master data for non-SAP sources, and Prior Labs’ tabular AI will power predictions that are “more accurate than any other agent in the industry” because it is trained on SAP’s vast proprietary tabular datasets. The agents will be monetized through value-based pricing, not as a standalone model, offering what Klein called “a unique chance for SAP to do a reset and really price outcome value-based.”

The RISE and GROW with SAP migration programs also saw strong uptake. New wins included Shell, Morgan Stanley, Eli Lilly, Electrolux, and Samsonite Group. A new AI migration toolchain cut ERP migration costs by up to 30%, and a case study cited Dexco eliminating 97% of legacy customizations and achieving a 75% faster accounting close.

Analysts Press on Margins and Outlook

The Q&A session exposed lingering concerns. BNP Paribas’s Ben Castillo-Bernaus noted that H1 cloud revenue was running ahead of guidance, yet the implied H2 deceleration seemed drastic. Klein attributed the caution to macro uncertainty, especially in the Middle East, but said the pipeline post-Sapphire was stronger than a year ago. “We are more and more confident that we can hit the guidance,” he said.

UBS’s Michael Briest asked about the 14% jump in R&D costs versus a 3% headcount increase. Klein explained that token consumption in AI development and the hiring of higher-cost AI experts drove the gap, but he expects headcount growth to moderate as productivity gains offset spending. The company is also shifting its development backlog from traditional SaaS features to AI agents—a transition that will take time but will ultimately improve efficiency.

Bank of America’s Frederic Boulan questioned whether the 0.7 percentage point decline in cloud gross margin (to 74.6%) was structural. Klein said it was temporary, driven by one-time investments in test environments, Joule Work, and sovereignty infrastructure. He reiterated the commitment to an 80–90% expense-to-revenue ratio.

Evercore ISI’s Kirk Materne pressed on the monetization strategy for Prior Labs, and Klein stressed that the open-source tabular AI module would be embedded into agents and priced based on the outcomes delivered—a departure from traditional software licensing. Asam added that Prior Labs’ training on SAP’s proprietary tabular data positions the company uniquely, as “that is not public data.”

Geographic Performance

RegionCloud Growth HighlightAPJOutstanding, led by India, South KoreaEMEAOutstanding, led by Germany, France, Italy, SpainAmericasSolid; U.S. particularly strongNotable MarketsBrazil also delivered standout growth
Outlook

For the second half, SAP expects a slight deceleration in CCB growth due to geopolitical risks and the fluid Middle East situation, but management emphasized that the second half typically accounts for the lion’s share of bookings. AI monetization is expected to start showing in H2 as agents and Joule Work launch. Asam closed with a broad defense of SAP’s model: “We are more convinced than ever that our strategy not to be locked into any generic large language frontier model … is the right one. Customers do value the resilience of this model delivered by a provider actually headquartered in Germany.”

As the enterprise AI race intensifies, SAP is betting that its combination of mission-critical data, industry know-how, and open model flexibility will win over customers wary of runaway token costs and vendor lock-in.