SAP SE (NYSE: SAP) — Europe’s Software Champion, Halved on a Backlog Wobble While the Engine Reaccelerates

Independent equity research. Report date: 2026-06-13. All figures in EUR under IFRS unless noted; ADR prices in USD. SAP is a German foreign private issuer (files Form 20-F / 6-K, not 10-K/10-Q). Current ADR ~$164 (€143 equiv.); ~1,167.6M shares outstanding; market cap ~$191B (~€166B); enterprise value ~€164B (net cash).

⚡ Claude’s Take

This block is the author’s own independent opinion and is general information only — not investment advice and not a recommendation to buy or sell any security. The analysis that follows it takes no position and carries no price target; do your own research and consult a licensed adviser before investing.

Verdict: BUY / accumulate on weakness. Conviction: medium-high. Directional zone: I’d accumulate the ADR below ~$175 (≈18–19x 2026E non-IFRS EPS / ~16x forward FCF), add aggressively below ~$150 (~16x), and consider the position rich only back above ~$230 (~26x). The analysis below takes no position; this is the one place I do.

Tag: “Europe’s best software franchise on its deepest sale since 2022.” SAP — the company that, eleven months ago, was the most valuable in Europe — has lost ~46% of its value from its July-2025 peak. The trigger was not a fundamental break but a second-derivative one: at the January 2026 FY-results, management guided that current-cloud-backlog (CCB) growth would “slightly decelerate” from 25%, and the stock had its worst day since 2020. The market then extrapolated that wobble, layered on a global software/AI de-rating and a fashionable “AI agents will gut seat-based SaaS” narrative, and re-priced SAP as if growth were over. Meanwhile the actual P&L is inflecting up: 2025 revenue grew 7.7% (cloud +27% in Q1’26), IFRS operating margin jumped from 22.8% to 26.1% and hit 30% in Q1’26, FCF is guided to ~€10B in 2026 (+21%), and management doubled the buyback to €10B straight into the sell-off. This is a quality-compounder-at-a-price set-up, not a falling knife in the fundamentals — though the tape (below all moving averages, deeply negative momentum) is precisely what makes it a contrarian call rather than a comfortable one. The factor read agrees: SAP loads as German + low-volatility + software, with Momentum zeroed — the trend is broken even as the business is not.

What the market is mispricing: at ~13x forward EV/EBIT and ~16x forward FCF, the multiple embeds low-single-digit long-term FCF growth for a franchise guiding +23–25% cloud growth into a forced 2027 ERP-migration wave with ~73% gross margins and genuine ERP switching-cost captivity. What the bears get right: CCB is decelerating, the AI-on-SaaS question is real and unresolved, and “sovereign cloud / government deal timing” is a convenient excuse that could mask demand softening. Bull-flip trigger: CCB stabilizes ≥~20% and Joule begins showing up as consumption revenue. Bear-flip trigger: a 2026 cloud-revenue guide cut, or CCB growth breaking below ~20% with no AI offset. I think the risk/reward at ~$164 is asymmetric to the upside; you are paid to wait via a 1.5% dividend and a 5%-of-cap buyback.

1. Executive Summary

SAP SE is the world’s largest enterprise-application-software company and the dominant vendor of enterprise resource planning (ERP) software — the financial and operational “system of record” for a large share of the global economy. After a decade-long, often-painful pivot from perpetual licenses to cloud subscriptions, SAP has reached the point where the transition is paying off: cloud is now the majority of revenue and growing ~25–27%, the legacy-license drag is small and shrinking, total revenue growth is re-accelerating, and the margin J-curve has turned decisively upward (IFRS operating margin 22.8% → 26.1% in 2025; 30% in Q1 2026).

Despite this, the ADR has fallen ~46% from its July-2025 high to ~$164, erasing more than €130B of market value. The proximate cause was the January-2026 guidance that current cloud backlog (CCB) growth — the metric the market had fixated on — would “slightly decelerate” from 25%. That, amplified by a broad de-rating of software/AI names and fears that agentic AI will erode seat-based SaaS economics, compressed SAP’s multiple from ~30x EV/EBITDA (2024) to ~17x (~13x forward).

The investment tension is sharp and worth stating plainly. Bull: a wide-moat, mission-critical incumbent with ~73% gross margins, ~30% operating margins, ~€10B of 2026 FCF, a forced multi-year 2027 ERP-migration tailwind, and a doubled €10B buyback — now at a multiple that prices stagnation. Bear: a maturing growth company whose key forward metric is rolling over, facing a genuine (if early) threat to the per-seat SaaS model from AI agents, with a goodwill-laden balance sheet (€29B) and a history of optical earnings volatility. This memo argues the franchise quality and the embedded-expectations gap are real; it also takes the bear’s AI and deceleration concerns seriously rather than waving them away.

No recommendation or price target appears below this summary (see Claude’s Take above for the single, fenced-off exception). The body evaluates SAP as embedded expectations and scenarios.

2. Business Overview

What SAP does. Founded in Walldorf, Germany in 1972, SAP builds the software that runs the back office of large enterprises: ERP (finance, procurement, manufacturing, supply chain, asset management), plus a suite of adjacent applications. Its installed base spans >180 countries and includes a very large share of the world’s biggest companies; SAP routinely notes that a substantial majority of global commerce, by transaction value, touches an SAP system. The company employs ~108,000 people.

Product architecture. The portfolio is organized around a cloud ERP core and a set of “line-of-business” (LoB) suites:

• SAP S/4HANA / SAP Cloud ERP — the next-generation ERP suite, delivered in two principal cloud models: Cloud ERP Private (the offering formerly marketed as RISE with SAP, a managed single-tenant private cloud for complex existing customers) and Cloud ERP Public (GROW with SAP, a standardized multi-tenant SaaS aimed at faster deployments and newer/smaller customers). This is the strategic center of gravity.
• LoB suites: SuccessFactors (human capital management/HCM), Ariba / Fieldglass / Concur (spend management — procurement, contingent workforce, travel & expense), Customer Experience (CRM/commerce), and Business Technology Platform (BTP) — the development, integration, data, and AI layer (SAP HANA database, SAP Build, and the Business Data Cloud).
• AI: Joule, SAP’s generative-AI copilot and emerging agent framework, embedded across the suite.
• Acquired capabilities: Signavio (process mining), LeanIX (enterprise architecture), WalkMe (digital adoption, acquired Nov 2024), SmartRecruiters (talent acquisition, Sept 2025), Taulia (working-capital/supply-chain finance), and most recently Reltio (master-data management, referenced in the Q1 2026 call).

The platform/data layer (why BTP matters). Increasingly the strategic linchpin is not any single application but the Business Technology Platform and the Business Data Cloud — the layer that unifies data and process across SAP’s (and third-party) applications and exposes it to AI. This matters for three reasons: (1) it is where AI/Joule is monetized; (2) it deepens lock-in (once a customer builds extensions and integrations on BTP, leaving is even harder); and (3) it is SAP’s answer to the hyperscalers and best-of-breed players — keep the customer’s data and process logic inside the SAP estate. The HANA in-memory database underpins this. In Greenwald terms, BTP converts SAP’s application captivity into platform captivity — a structurally stronger position if it holds.

How it makes money — revenue model and mix. SAP reports three revenue streams:

1. Cloud — subscription (and increasingly consumption) revenue for SaaS/PaaS. ~€21.0B in 2025, ~57% of revenue, growing ~25–27%. This is the future of the model and the swing factor for the whole thesis.
2. Software licenses & support — the legacy on-premise business. Support is a high-margin (>90% gross margin), extremely sticky maintenance annuity on the installed base; new licenses are in deliberate, rapid runoff (down ~33% YoY in Q1 2026) as SAP pushes customers to cloud.
3. Services — consulting/implementation; lower-margin, lumpier, and the source of a 2026 “setback” management has flagged as a one-off.

The key structural fact: SAP’s revenue is now highly recurring and predictable. Cloud subscriptions plus on-premise support together constitute the large majority of revenue (the prior internal work pegged “predictable revenue” at ~83% of total in 2024, up from ~72% in 2020), and “current cloud backlog” (contracted cloud revenue due in the next 12 months) gives ~12-month forward visibility. The business has shifted from selling a product (lumpy licenses) to renting a utility (recurring cloud + support).

Revenue composition (EUR, approximate). The mix shift is the whole story of the franchise:

The arithmetic of the transition: cloud has roughly tripled while the legacy license-and-support block has shrunk — so total revenue grew more slowly than cloud, which is exactly why the headline growth rate looked uninspiring for years even as the high-quality recurring base compounded underneath. As the legacy block becomes a smaller share, total growth mechanically converges upward toward the cloud growth rate — the “re-acceleration” management is guiding toward through 2027.

Geographic footprint. Truly global — >180 countries, ~108k employees, ~100 development locations. In Q1 2026 management called out the US as “particularly strong,” alongside India, South Korea, Switzerland, and the UK — notable because US public-sector caution had been a feared 2026 headwind. EMEA (its home region) and the Americas are the largest revenue blocks; Asia-Pacific is the fastest-growing.

Verdict: A genuine franchise business model — recurring, mission-critical, deeply embedded — now past the hardest part of a cloud transition that mechanically depressed reported growth and margins for years. The model is structurally improving, not deteriorating.

3. Industry Dynamics

Market structure. Enterprise application software is one of the most attractive sub-sectors in all of technology: large (>$300B and growing high-single-to-low-double digits), recurring, sticky, and structurally oligopolistic at the high end. ERP specifically — the system of record for large enterprises — is effectively a two-horse race between SAP and Oracle at the top of the market, with a long tail of mid-market players (Microsoft Dynamics, Infor, Workday Financials, NetSuite, Unit4) and best-of-breed specialists in adjacent LoBs.

The profit pool is deep because the product is (a) mission-critical (an ERP outage stops a company from invoicing, paying, manufacturing, or closing its books), (b) deeply integrated into customer processes and data, and © extraordinarily expensive and risky to replace. These properties create the high barriers to entry and customer captivity that Greenwald & Kahn identify as the source of durable competitive advantage.

The anatomy of the switching cost. It is worth being concrete about why ERP is so sticky, because the entire moat rests on it. A large enterprise’s SAP system typically encodes: years of custom configuration and code (ABAP), thousands of integrations to upstream/downstream systems, master data for customers/vendors/materials/employees, regulatory and tax logic across dozens of jurisdictions, and the trained muscle memory of thousands of employees and the entire ecosystem of consultants who support them. Replacing it means re-implementing all of that while keeping the business running — a project measured in years and hundreds of millions of euros, with a real risk of botching the financial close, payroll, or order-to-cash in the process. The expected value of switching is almost always negative even when the incumbent raises prices or ships mediocre features. This is demand-side captivity in its purest Greenwald form, and it is why SAP can run a >90%-gross-margin support annuity and push a forced cloud migration without mass defection.

The 2027 catalyst. SAP has set end-2027 as the end of mainstream maintenance for legacy ECC (SAP Business Suite 7), with paid extended maintenance to 2030 and limited leeway to 2033. This is the single most important industry dynamic for SAP’s medium term: it is a forced migration event for a very large installed base that must move to S/4HANA (cloud or on-premise) or run unsupported. SAPinsider survey data indicate adoption is accelerating (e.g., ~30% of surveyed customers fully live on S/4HANA Cloud Private, up from ~19% a year earlier), and — notably — that SAP’s AI roadmap has become the #1 external factor in customers’ ERP decisions (cited by ~43%), now ahead of the maintenance deadline itself (~39%). A looming resource bottleneck (too few experienced integration consultants for the migration wave) is a real friction but also a demand signal.

AI as both threat and tailwind. This is the central industry debate. The bear case is that agentic AI compresses the value of per-seat software: if an AI agent does the work, you need fewer seats. The bull case is that enterprise AI is only as good as the structured, governed business data and process context it runs on — which is exactly what SAP owns — and that AI raises switching costs further by embedding intelligence in the system of record. SAP’s own framing (see the Competitive Position section) is that agents “often don’t yet have the full understanding of business data and processes to deliver highly accurate outcomes,” and that SAP’s data/process layer is the prerequisite for reliable enterprise AI.

Regulation & geopolitics. Enterprise software is lightly regulated relative to, say, healthcare or banking, but two themes matter: (1) data sovereignty — European and other governments increasingly demand “sovereign cloud” (data residency, operational control), which SAP is positioning as a differentiator but which lengthens government/defense sales cycles; and (2) macro/tariff/IT-budget cyclicality — enterprise software is relatively defensive but not immune; deal timing slips when CFOs get cautious.

The capital-cycle lens (Marathon). The enterprise-software/SaaS sector is a textbook study in Chancellor’s capital cycle. The 2020–2021 era saw an enormous supply of capital rush into software — record VC funding, SPACs, sky-high SaaS multiples (SAP itself reached ~30x EV/EBITDA and ~8x sales in 2024) — the classic signal of a sector attracting too much capital, which mean-reverts. The 2025–2026 de-rating is that reversion: capital and enthusiasm withdrawing, multiples compressing across the group. The Marathon insight is that this is precisely when a high-quality, low-capital-intensity incumbent becomes interesting — the supply-side response (capital fleeing the sector, weaker private players starved of funding, AI-native start-ups burning cash) tends to strengthen the entrenched, cash-generative incumbent’s position even as its multiple falls. SAP is not adding capacity into a glut; it is harvesting an installed base and returning cash. That asymmetry — falling multiple, strengthening competitive position — is the capital-cycle setup the framework prizes.

Profit pools and the duopoly. The deepest profit pool in enterprise software is the system-of-record layer — ERP and its high-margin support annuity — precisely because it is the hardest to displace. SAP and Oracle have controlled this layer for decades; their combined share of large-enterprise ERP has been remarkably stable, the share-stability test Greenwald uses to confirm a moat. The adjacent LoB pools (HCM, CRM, ITSM, spend) are more contestable and is where best-of-breed challengers have taken share — but those are thinner, more competitive pools than the ERP core SAP anchors.

Verdict: Structurally good industry — high barriers, recurring revenue, deep profit pools, an oligopolistic high end, and a rare forced-migration catalyst. The AI question is the genuine swing variable: it could be the next leg of the moat or the first crack in the model. On current evidence it looks more like a tailwind for the incumbent that owns the data, but the verdict is not yet provable.

4. Competitive Position

The moat, named. In Greenwald’s taxonomy, SAP’s advantage is primarily customer captivity (demand-side switching costs), reinforced by economies of scale and intangibles (process/industry IP and data):

• Switching costs / captivity (dominant). ERP is the “central nervous system” of an enterprise — every order, invoice, payroll run, and financial close flows through it. Ripping it out and replacing it is a multi-year, multi-hundred-million-euro, business-risking project. Customers stay through pricing actions and product friction because the alternative is worse. This shows up financially as >90% support gross margins on a decades-old installed base and very high gross/net retention in cloud.
• Scale economies. SAP’s installed base funds an R&D budget (~€6.6B in 2025) and a partner ecosystem (SIs, ISVs) that few can match. The partner channel — accounting for a large share of order entry — is itself a scale advantage: thousands of consultants whose careers are built on SAP create distribution and lock-in.
• Intangibles. Fifty years of industry-specific process templates, data models, and configuration IP are not easily replicated. This is the asset that makes the AI argument credible: SAP’s structured business data and process semantics are the substrate enterprise AI needs.

Greenwald tests. (1) Market-share stability: the SAP/Oracle ERP duopoly at the high end has been stable for years; best-of-breed players (Salesforce in CRM, Workday in HCM, ServiceNow in workflow) have peeled off adjacent LoBs but have not displaced the ERP core. (2) ROIC test: SAP earns mid-teens ROE (15.4% in 2025) and ~12% reported ROIC — depressed by a goodwill-heavy balance sheet (€29B goodwill); on tangible invested capital the returns are far higher, consistent with a real moat. The 2024 trough (7.1% ROE) was a restructuring artifact, not a deterioration in economics.

Direct competitive read:

• Oracle — the closest peer; comparable ERP installed base and a more aggressive cloud-infrastructure (OCI) story. Oracle has out-rated SAP on the “AI infrastructure” narrative. In ERP applications, the two are roughly co-leaders (each ~6.5% of a fragmented global ERP market on one 2025 estimate).
• Workday / ServiceNow / Salesforce — best-of-breed in HCM / workflow / CRM respectively; they compete with SAP’s LoB suites (SuccessFactors, etc.), not its ERP core, and trade at materially higher multiples (NOW ~50x, WDAY/CRM ~22–25x forward earnings vs SAP ~19–20x).
• Microsoft — Dynamics competes in the mid-market and, more importantly, Azure + Copilot make Microsoft both a partner (SAP runs on Azure) and a long-term platform rival.
• Hyperscalers (AWS/Azure/GCP) — partners today (RISE runs on their infrastructure) and potential up-stack threats over time.

SAP vs. Oracle — the instructive head-to-head. The two ERP co-leaders have diverged in market perception even as their application businesses remain comparable. Oracle re-rated hard in 2024–2025 on its OCI / AI-infrastructure story (training-cluster capacity, RPO bookings from AI customers), while SAP de-rated on cloud-application-growth concerns. The contrast is informative: Oracle is being rewarded for a capital-intensive infrastructure bet (heavy capex, debt-funded data centers), while SAP is being penalized despite a capital-light applications model that throws off far more free cash per euro of revenue (SAP capex ~2% of revenue vs. Oracle’s surging data-center capex). On the core ERP application franchise — the durable, high-margin profit pool — the two are peers; SAP arguably has the stronger pure-applications position and the cleaner cash model, yet trades at a similar-to-lower multiple. For an investor focused on cash returns rather than AI-capex narratives, that relative positioning favors SAP.

The best-of-breed dynamic, quantified. Workday, ServiceNow, and Salesforce have built large franchises ($ multi-billion revenue each) by out-executing SAP in single LoBs (HCM, workflow/ITSM, CRM respectively). This is a real, ongoing share loss at the edges — SuccessFactors and SAP CX are not the category leaders in their niches. But two facts contain the damage: (1) these are adjacent pools, not the ERP core (a customer can run Workday for HR and still run SAP for finance/supply chain); and (2) SAP’s strategy is to win on integration — the value of having finance, supply chain, procurement, and HR on one data model — rather than on best-of-breed depth in each. Whether “integrated suite” beats “best-of-breed + integration tools” is the perennial enterprise-software debate; SAP’s bet is that AI raises the value of the unified data model, tilting the debate back toward the suite.

The AI question, examined as a moat issue. The bear’s sharpest argument is that AI breaks the moat by collapsing the value of seats. Three pieces of evidence push back, none conclusive:

1. Pricing is already mostly not seat-based. Management states <40% of 2025 cloud revenue is tied to named/seat users; the rest is priced on consumption, revenue processed, memory used, and other value metrics. If an agent does the work, value-based pricing can capture that productivity rather than lose it — and management explicitly expects “an increasing share of consumption-related cloud revenue” as AI scales.
2. The data/process layer is the constraint on enterprise AI. SAP’s own candid framing on the Q1 2026 call: agents “often don’t yet have the full understanding of business data and processes to deliver highly accurate outcomes” for mission-critical work. SAP owns that structured business data and the process semantics (the Business Data Cloud strategy) — the substrate a trustworthy enterprise agent needs. This is a moat-deepening argument: AI raises the value of being the system of record.
3. Adoption is real but early. Customers cite SAP’s AI roadmap as the #1 external factor in ERP decisions (~43%), yet only ~3% run Joule in production — meaning the monetization is almost entirely ahead, not in the base. That is optionality, not a current crutch. The honest counter: this is SAP’s narrative, and the technology is moving fast; an AI-native competitor that re-imagines ERP workflows without the legacy baggage is a low-probability-but-high-impact tail risk. The moat is strong today; the AI platform shift is the one vector that could erode it over a 5–10 year horizon, and it cannot yet be adjudicated.

Verdict: A durable, financially-validated moat, anchored in ERP switching costs and reinforced by scale and data/process IP — among the widest in software. The vulnerability is at the edges (best-of-breed LoB erosion) and in the next platform shift (AI), not at the ERP core, which remains defensible.

5. Growth History and Forward Opportunities

The transition, in numbers. SAP’s reported growth was deliberately suppressed for years by the model shift: moving a customer from a big up-front license + maintenance to a ratable cloud subscription lowers near-term reported revenue even as lifetime value rises (the SaaS “trough”). That trough is now behind it:

The story the table tells: cloud revenue has roughly 2.6x’d in five years while total revenue grew more slowly (legacy license/support runoff is the offset), and operating margin troughed in 2022–2023 (peak transition + restructuring) before inflecting sharply upward as cloud scales and the 2024 restructuring lands.

Current momentum (Q1 2026). Cloud revenue +27% (to ~€6B); Cloud ERP Suite +30%, now ~87% of total cloud growth; public-cloud order entry >70% of volume and accelerating; total revenue +12%. This is acceleration, not deceleration, in the revenue line — the deceleration debate is purely about backlog growth (a leading indicator), not realized growth.

Forward opportunities:

1. The 2027 migration wave. The largest near-term driver: a forced move of the ECC base to S/4HANA, each migration typically expanding the customer’s spend (more modules, cloud premium, AI add-ons).
2. Cloud ERP penetration of the installed base. Even after years of cloud growth, a large share of SAP’s core ERP base is still on-premise — a multi-year conversion runway.
3. AI / Joule monetization. Today nascent (only ~3% of customers run Joule in production), but the optionality is large: premium AI SKUs, consumption-based revenue, and the Business Data Cloud as a new data-platform revenue stream. Management explicitly expects “an increasing share of consumption-related cloud revenue” over the coming years.
4. LoB cross-sell and Business Network (Ariba/Business Network B2B commerce) into the ERP base.
5. Margin-driven earnings growth — even at decelerating revenue growth, operating leverage (mix shift to high-margin cloud + cost discipline post-restructuring) drives faster profit and FCF growth than revenue (2026 guide: revenue +12–13% cc, non-IFRS op profit +14–18%, FCF +21%).

The migration economics — why a forced move helps SAP. A typical ECC-to-S/4HANA migration is not a like-for-like swap; it is an upsell event. Customers moving to RISE/GROW generally (a) adopt the cloud subscription (recurring, often higher annual value than the old license+maintenance), (b) take additional modules and the BTP platform, and © become candidates for Joule/AI add-ons and Business Network. SAP has repeatedly signalled that migrated customers expand their footprint. So even if the number of migrating customers is finite, the revenue per migration is rising — which is how a maturing logo-growth story can still produce mid-teens-plus cloud-revenue growth. The 2027 deadline (extended maintenance to 2030/2033) effectively converts a discretionary upgrade into a deadline-driven purchase for a very large base — a multi-year, visibility-enhancing demand pull.

2026 guidance as the forward bridge. Management’s maintained FY2026 outlook (constant currency) quantifies the trajectory: cloud revenue €25.8–26.2B (+23–25%), cloud-and-software revenue €36.3–36.8B (+12–13%), non-IFRS operating profit €11.9–12.3B (+14–18%), and FCF ~€10B (+~21%). The shape is the thesis in miniature — revenue growth in the low-double-digits, profit growth faster (operating leverage), and FCF growth faster still (low capex + working-capital tailwind). Note that management felt compelled to fold the inorganic Reltio acquisition into the range “to protect” it against the services setback — an honest disclosure that the organic path has a wider error band this year, and a fair point for bears.

Verdict: High-quality growth. It is recurring, margin-accretive, partly contracted (backlog), and supported by a structural forced-migration catalyst. The honest caveat: the rate of cloud growth is maturing from ~25% toward the low-20s/high-teens, and the bull case depends on AI/consumption revenue arriving to extend the runway before the migration wave crests.

6. Financial Quality

Margins and operating leverage. SAP runs a classic high-quality software P&L: ~73% gross margin, with cloud gross margin ~75% (IFRS 74.6%, non-IFRS 75.2% in Q1 2026) and support gross margin >90%. The key 2024→2025 event is the operating-margin inflection: IFRS operating margin rose from 22.8% to 26.1%, and reached 30% in Q1 2026 (+2.9pp YoY). EBITDA margin rose from 26.6% to 29.7%. This is the cloud-scale-plus-cost-discipline thesis showing up in the numbers, and it is the cleanest evidence that the transition economics are working.

Earnings — read non-IFRS for the trend, but watch the gap. IFRS net income is optically volatile because of large below-the-line and one-time items:

• 2024 IFRS net income €3.12B (margin 9.1%, ROE 7.1%) was depressed by the ~€2.5–3.1B restructuring charge (the ~10,000-role program) and high SBC; it was not a deterioration in the business.
• 2025 IFRS net income €7.16B (margin 19.5%, ROE 15.4%, EPS €6.14) is the normalized picture as restructuring rolled off.
• Non-IFRS operating profit €10.42B (2025) is the metric management guides and the market models; the IFRS-to-non-IFRS bridge is mainly SBC and acquisition-related amortization/restructuring.

Quality-of-earnings flags to keep honest: (1) SBC is real and sizeable (€1.70B in 2025, down from €2.39B in 2024) — non-IFRS adds it back, so non-IFRS overstates economic profit; (2) acquisition-related amortization recurs because SAP keeps acquiring; (3) 2024 also carried large Sapphire Ventures / equity-investment gains in non-operating income that flattered pre-tax income. The corrective is to anchor on cash flow, which is harder to dress up.

Cash flow — the anchor. This is where the franchise shows best:

FCF inflected hard in 2025 to €8.42B (company figure €8.24B) and is guided to ~€10B in 2026 (+~21%). Capital intensity is very low (~2% of revenue) — SAP runs largely on hyperscaler infrastructure for cloud, so it scales without heavy capex, a structurally superior cash model versus infrastructure-owning peers. FCF conversion of net income is high (>1x).

Balance sheet. Fortress-like. Net cash of €2.43B at end-2025 (cash €9.8B incl. STI vs. total debt €7.5B), down from net debt years earlier as the company de-levered. Total assets €70.4B are dominated by goodwill (€29.0B) + intangibles (€2.3B) from a long M&A history — this inflates the equity base and depresses reported ROIC/ROE (i.e., the moat is understated by accounting). Current ratio ~1.16; negative cash-conversion cycle (customers/deferred revenue fund the business). Deferred revenue (€6.6B short-term) is a feature, not a liability, of the subscription model.

The cloud gross-margin J-curve — the underappreciated lever. A subtle but important point: cloud gross margin (~75%) is structurally lower than the on-premise support gross margin (>90%). So as the mix shifts from support to cloud, blended gross margin faces a mechanical headwind — which is why SAP’s overall gross margin has been roughly flat at ~73% for years despite cloud scaling. The bull’s insight is that cloud gross margin is itself still climbing as SAP optimizes hyperscaler costs, multi-tenant efficiency, and (prospectively) AI-driven delivery productivity. Management cited using Joule for ABAP development and third-party tools (Claude Code, GitHub Copilot) lifting developer productivity >30%, and AI cutting service-delivery timelines up to 30%. If cloud gross margin grinds from ~75% toward the high-70s/low-80s while operating expense leverage continues, the operating-margin path to management’s mid-30s ambition is credible — and that, not revenue growth, may be the larger driver of earnings over the next three years.

Working capital and cash conversion. SAP runs a negative cash-conversion cycle (−21 days in 2025): customers pre-pay subscriptions and support (€6.6B short-term deferred revenue), funding the business interest-free. DSO is well-managed; the 2025 OCF surge (€9.16B vs €5.21B in 2024) reflects both the earnings recovery and the unwind of restructuring cash outflows. The €408M Teradata litigation settlement paid in Q1 2026 is a one-time cash item, not a recurring drag.

Returns on capital. ROE 15.4% (2025), ROIC ~12.4% reported — solid but not spectacular as reported; adjusted for the goodwill drag and the restructuring trough, the underlying cash-on-tangible-capital returns are far higher, consistent with a wide-moat compounder. Note the trajectory: ROE collapsed to 7.1% in the 2024 restructuring year and snapped back to 15.4% in 2025 — a vivid illustration of why single-year IFRS ratios mislead here and why the multi-year, cash-based read matters.

Verdict: Economics improve with scale — emphatically. The margin inflection, the FCF step-change, the asset-light cloud model, and the fortress balance sheet are all hallmarks of a high-quality business. The one discipline required of the analyst is to read cash flow and non-IFRS trends rather than noisy IFRS net income, while not letting non-IFRS launder away real SBC.

7. Capital Allocation

Philosophy. SAP’s capital allocation has matured markedly under CFO Dominik Asam: fund organic R&D, make targeted bolt-on/tuck-in acquisitions in data/AI/process, return the rest via a growing dividend and — increasingly — large buybacks. The 2025 dividend payout was ~36% of IFRS earnings, leaving ample room for repurchases.

Capital returns — and the signal in the timing. The standout capital-allocation fact of this report:

• A new €10B share-buyback program was announced on 29 January 2026 (term through end-2027), on the same day the stock fell ~11% on the CCB-deceleration guide. This is on top of the prior €5B program launched in January 2024. A €10B authorization is ~5% of the current market cap.
• Dividend raised to €2.50/share for FY2025 (+6.4% YoY from €2.35), continuing a steady upward march (€1.58 → €1.85 → €2.05 → €2.20 → €2.35 → €2.50 over recent years).
• Actual repurchases: €1.94B (2025), €2.11B (2024). Combined with dividends (~€2.7B in 2025), SAP returned ~€4.6B to shareholders in 2025 and is set to return materially more in 2026–2027 under the enlarged authorization.

Management doubling the buyback into a 46% drawdown is the clearest possible statement that the board views the shares as cheap and the FCF outlook as durable. It is also the right textbook move: buy back stock when the implied return is highest. We treat this as a strong (if self-interested) bullish signal, not as proof.

M&A track record — mixed, improving. SAP’s history includes both value-destructive mega-deals and sensible bolt-ons:

• Cautionary tale: the Qualtrics saga — acquired for ~$8B in 2019, partially re-IPO’d, then taken private/sold (Silver Lake) in 2023 — was an expensive round-trip that the market rightly criticized. Older large deals (Concur, SuccessFactors, Ariba) were strategically sound but richly priced.
• Recent discipline: WalkMe (~$1.5B, 2024), SmartRecruiters (2025), Signavio, LeanIX, Taulia, and Reltio (2026) are smaller, capability-focused tuck-ins that fit the data/AI/process strategy. The 2025 cash-for-acquisitions line was a modest €0.7B.

The concern: a €29B goodwill balance is the accumulated evidence of a company that has historically bought growth at high prices. The reassurance: recent deal sizes are disciplined and the cash-return mix has shifted decisively toward buybacks/dividends.

Reading the goodwill honestly. €29B of goodwill against €45B of equity is a large number, and it depresses reported ROE/ROIC (the denominator carries the full acquisition price while the numerator carries only the operating returns). Two interpretations: the bear sees a serial acquirer that has overpaid and may impair; the bull notes that SAP’s major acquisitions (SuccessFactors, Ariba, Concur, Qualtrics) were bought years ago, are now deeply embedded and cash-generative, and have not required large impairments — i.e., the goodwill mostly reflects genuinely-integrated, revenue-producing assets rather than a ticking write-down. The Qualtrics round-trip was the clearest misstep (bought high, exited at a loss of strategic value), but it was also corrected — management recognized the error and divested. On balance the M&A record is “rich but not reckless,” and the recent shift toward small tuck-ins plus large buybacks suggests the lesson was learned. The single most important capital-allocation question going forward is whether management resists the temptation to make another large, expensive “AI” acquisition with the FCF that would otherwise fund the buyback.

The shareholder-yield math. Combining the two return channels: a ~1.5% dividend yield plus a €10B buyback authorization (~5% of the ~€166B market cap) over roughly two years (~2.5%/yr) implies a ~4% all-in shareholder yield before any growth — and the buyback is far more accretive executed at ~16x FCF than it would have been at the 2024 peak of ~47x. If management executes the authorization near current prices, it retires stock at a ~6% FCF yield, a genuinely value-additive use of cash. This is the capital-allocation expression of the entire thesis: the same FCF that the market is capitalizing cheaply, management is using to shrink the share count cheaply.

R&D and reinvestment. ~€6.6B R&D (2025), ~18% of revenue — appropriate and necessary to fund the S/4HANA + BTP + Joule roadmap. SBC (€1.7B) is the labor-retention cost of competing for engineers; it is declining, which is healthy. Critically, SAP funds this R&D and expands margins simultaneously — the hallmark of a scale business where incremental revenue carries very high contribution margin (incremental operating margin ran ~60–69% in 2024–2025 per ROIC).

Incentive alignment. Management compensation is tied substantially to cloud revenue, current cloud backlog, non-IFRS operating profit, and TSR — i.e., to the metrics that drive the thesis. (The risk: CCB-linked incentives can encourage backlog-optimizing behavior; investors should watch revenue realization against backlog.)

Verdict: Capital allocation has shifted from “buy growth richly” to “compound the core and return cash” — a clear improvement. The doubled buyback into weakness is well-timed and shareholder-friendly. The historical M&A record (Qualtrics, the goodwill pile) tempers the grade from “excellent” to “good and improving.”

8. Changes and Headwinds — Last Two Years

Strategic & leadership changes:

• Hasso Plattner retired (May 2024) as Supervisory Board chairman, ending the 52-year founder era; Pekka Ala-Pietilä chairs the board, Christian Klein remains sole CEO (since 2021), Dominik Asam is CFO. Founder departure is a governance watch-item but the transition has been orderly.
• 2024 restructuring (~10,000 roles, ~€2.5–3.1B charge): a deliberate reallocation toward AI/cloud talent; headcount ended ~flat as SAP rehired into strategic areas. This depressed 2024 IFRS earnings and is the main reason 2024 optics look weak.
• Portfolio reshaping: Qualtrics divested (2023); a string of AI/data/process tuck-ins acquired (WalkMe, LeanIX, Signavio, SmartRecruiters, Reltio).
• Product repositioning: RISE → “Cloud ERP Private,” GROW → “Cloud ERP Public”; launch of Business Data Cloud and the Joule agent framework as the AI layer.

Headwinds (the de-rating drivers):

1. CCB deceleration guide (Jan 2026) — the proximate catalyst. 25% CCB growth (~26% expected) plus guidance for “slight deceleration” → worst day since 2020.
2. Sovereign-cloud / government deal timing — geopolitical tension (management referenced a conflict escalation and government “firefighting”) lengthened public-sector/defense sales cycles; some sovereign/defense deals carry termination clauses that keep them out of reported backlog, distorting CCB lower even where demand exists.
3. A 2026 “services setback” — management flagged a one-off services-revenue shortfall, partially offset by pulling the Reltio acquisition into guidance to “protect the range.”
4. Global software/AI de-rating — a sector-wide multiple compression in expensive software through late-2025/2026.
5. AI-disruption narrative — the fear that agentic AI erodes per-seat SaaS economics (addressed, and partly rebutted, in the Variant Perception section).
6. FX — SAP reports in EUR; USD weakness/EUR strength is a reported headwind (much of revenue is non-EUR), which is why SAP guides at constant currency.

The geopolitical/scenario overlay. Management was unusually explicit that the maintained 2026 outlook rests on a scenario — a near-term de-escalation of a geopolitical conflict (referenced obliquely; the Q&A touched on Iran) and no sharp deterioration in the macro/tariff backdrop. In the immediate aftermath of the escalation, governments and affected industries shifted into “immediate firefighting,” delaying deal activity. This cuts both ways for the analyst: it is a plausible, externally-verifiable reason for CCB softness (i.e., not a demand-quality problem), but it also means guidance carries more conditionality than usual. SAP’s exposure here is mostly timing (deals slip right) rather than loss (deals cancel) — sovereign/defense demand is, if anything, structurally rising — but the timing can be lumpy and is genuinely harder to forecast, which is why management widened the range of CCB outcomes.

The counter-signal: at Q1 2026 (April), management maintained full-year guidance, CFO Asam publicly pushed back on the bear narrative, the buyback was doubled, and the actual numbers (cloud +27%, op margin 30%, non-IFRS EPS +20%) were strong. The Teradata litigation was settled (€408M payout in Q1 2026), removing an overhang. The market’s response — continued drift lower into June — suggests investors are extending no benefit of the doubt; sentiment, not fundamentals, is doing the work in the tape.

Verdict: The last two years contain one genuine negative (growth-rate maturation, signaled by CCB) wrapped in several transient or self-inflicted optics (restructuring, services one-off, FX, sovereign-deal timing) and a sector de-rating. On balance the operational changes strengthen the thesis (margins, FCF, capital return) while the narrative changes (CCB, AI fear) are what moved the stock. Whether the growth maturation is benign or the start of something worse is the crux — and is not yet resolved.

9. Risk Analysis (Risk Matrix)

How the risks interact. The risks are not independent — they share a common root in the growth-durability question. If cloud/CCB growth stabilizes (Risk 2 benign), then the AI threat (Risk 1) is likely net-positive (consumption upside), the migration converts (Risk 3 benign), and the de-rating reverses. If instead growth rolls over, it will probably be because AI is compressing demand (Risks 1+2 compounding), the migration disappoints (Risk 3), and the multiple compresses further. In other words, the bull and bear scenarios are internally coherent clusters, and the single observable that most cleanly separates them is realized cloud revenue growth versus the guided range over the next 2–4 quarters. That is the variable to monitor above all others. The lower-impact risks (FX, governance, SBC) are manageable and largely cosmetic; the M&A risk (Risk 6) is the one self-inflicted danger — a return to large, richly-priced deals would simultaneously consume the buyback firepower and re-open the capital-allocation critique.

Risk verdict: The dominant risks are thesis risks (AI on SaaS; growth maturation), not survival risks. SAP’s balance sheet, recurring revenue, and mission-criticality make a permanent capital loss from business failure highly improbable; the realistic downside is a de-rating extension if growth disappoints, not impairment.

10. Valuation Discussion (Embedded Expectations)

No price target or recommendation. This section frames what the current price implies and runs scenarios.

Where the multiple sits — and where it came from. SAP has de-rated dramatically:

*EV/EBIT on non-IFRS operating profit is ~13x forward; on IFRS ~17x.

On forward 2026 guidance at the current ~€164B EV: EV/EBIT (non-IFRS €11.9–12.3B) ≈ ~13.4x; EV/FCF (~€10B) ≈ ~16.4x; forward non-IFRS P/E ≈ ~18–20x. The AZI own-history valuation_index corroborates: P/S in the ~51st percentile and P/B ~42nd percentile of SAP’s own ~10-year range — i.e., middling-to-cheap on its own history, despite the best margin/FCF profile in that history. (The P/E percentile, ~27th, overstates cheapness because of GAAP EPS distortion; lean on P/S, P/FCF, and EV/EBIT.)

Peer context. Among large-cap application software, SAP at ~18–20x forward non-IFRS earnings sits in the value tier. Approximate forward-earnings multiples for the comparable set (directional, mid-2026):

The point: SAP trades alongside the slowest-growing, most-challenged names (Adobe, IBM) despite pairing top-tier margins (30% operating, rising), accelerating cloud (+27%), and a forced-migration tailwind — a combination that historically commands a premium to the group, not a discount. Only the AI-disruption-fear cohort (Adobe) is comparably cheap, and Adobe lacks SAP’s switching-cost captivity and migration catalyst.

Embedded-expectations / reverse-DCF logic. At ~€164B EV with 2026E FCF of ~€10B (and FCF guided to grow ~21% in 2026 alone), the current price embeds low-single-digit long-term FCF growth under any normal discount rate. Work it explicitly: at an ~8.5% WACC, a simple two-stage model where FCF grows ~12% for five years then fades to a 3.5% terminal growth rate produces an intrinsic EV comfortably above €230–250B — i.e., ~40–50% above the current ~€164B. To justify the current ~€164B EV on a perpetuity, one needs FCF to grow only ~4–5% per year forever (≈ €10B × 1.045 / (0.085 − 0.045) ≈ €261B… and even discounting near-term, the implied long-run growth the market is underwriting is mid-single-digit at best). That is well below the +14–18% non-IFRS profit and +23–25% cloud-revenue growth management has guided for the current year alone, and below the structural runway implied by the 2027 migration and AI optionality.

Put differently: the de-rating has taken SAP from pricing in perpetual ~10%+ growth (the 2024 peak at 30x EBITDA) to pricing in near-stagnation. The truth is somewhere in between — but the current price requires the bear’s structural-maturation thesis to be not just right but severe, while the bull case requires only that SAP grow FCF at, say, 8–10% for a few years (less than half its guided 2026 rate) and that the multiple normalize toward the group. The market is pricing maturation-to-stagnation; management is delivering acceleration. That gap is the variant perception.

Scenario analysis (illustrative, 3-year horizon; directional, not a target):

• Bear: CCB decel accelerates, cloud growth fades to ~mid-teens then low-double-digits, AI pressures seats, multiple stays ~16x FCF or compresses further. FCF ~€11–12B in 3 yrs; modest or no re-rating → limited upside, possible further downside. The de-rating proves justified.
• Base: cloud growth normalizes to high-teens/low-20s, margins continue to expand, FCF compounds to ~€13–14B by 2028, multiple normalizes toward the mid-20s EV/EBIT / ~20x FCF as fears fade → meaningful upside from a combination of FCF growth and partial re-rating.
• Bull: the 2027 migration wave + Joule/consumption revenue extend ~20% cloud growth, margins reach mid-30s, FCF approaches ~€15B, and SAP re-rates toward its own historical premium → substantial upside.

Verdict: SAP is priced as a mature, ex-growth software laggard while operating as an accelerating, margin-inflecting compounder. The valuation embeds expectations that the company’s own current-year guidance contradicts. The risk is that guidance proves optimistic; the asymmetry is that even the base case requires only “fears fade and FCF compounds,” not heroics.

11. Variant Perception (Consensus, Bull, Bear)

Consensus view (mid-2026). SAP is a quality franchise whose growth is maturing; the cloud-transition payoff is largely “in the numbers”; CCB deceleration plus AI uncertainty cap the multiple; the stock is “dead money” until backlog growth stabilizes or AI monetization proves out. The tape reflects this: deeply negative 6/12-month momentum, below all moving averages, Momentum factor loading zeroed, and a -46% drawdown — capitulation, not enthusiasm.

The factor-positioning read (FactorsToday). SAP loads as Country: Germany (1.12), LowVolatility (0.53), Software/Cloud (~0.37/0.33), with Momentum and Value both absent (zeroed) in the all-factors model and a modest Market beta (~0.86–1.05 depending on model). Risk-adjusted track record has turned sharply negative short-term (1yr Sharpe −1.36, 6m return −54%) against a positive long-run record (10yr return ~9%/yr). Translation: this is a broken-trend, low-beta quality name — not a crowded momentum trade unwinding, and not yet a statistically “cheap” value screen. The absence of a Value loading despite the −46% move tells you the de-rating has only brought a richly-valued name back to fair, not to distressed — which is consistent with the embedded-expectations math (priced for maturation, not for collapse).

Strongest bull case. A wide-moat, mission-critical incumbent is being sold as a mature laggard precisely when its P&L inflects: cloud +27%, operating margin to 30%, FCF to ~€10B (+21%), a forced 2027 migration wave dead ahead, AI optionality (Business Data Cloud + Joule) barely monetized, and a €10B buyback retiring ~5% of the cap into the weakness — all at ~13x forward EV/EBIT and ~16x FCF. The market extrapolated one leading metric (CCB) and a narrative fear (AI) and mispriced the cash machine underneath.

Strongest bear case. CCB deceleration is the first honest signal that SAP’s growth is structurally maturing as the easy cloud-conversion is done; agentic AI genuinely threatens per-seat economics across SaaS and SAP is not immune; “sovereign cloud / government timing / services one-off” are convenient excuses that could be masking broader demand softening; the €29B goodwill pile is evidence of a serial acquirer that buys growth; and at ~20x earnings the stock is “only fair,” not cheap, so the de-rating may simply be the market correcting a 2024 bubble (when it briefly hit 30x EBITDA and 8x sales). In this read, there is no margin of safety, only a fairly-priced maturing franchise.

The 3–5 assumptions that matter most:

1. Does cloud/CCB growth stabilize in the high-teens/low-20s, or keep falling? (The whole thesis hinges here.)
2. Is AI a net tailwind (data-layer moat, consumption upside) or a net threat (seat compression) to SAP specifically?
3. Does the 2027 migration wave convert to revenue on schedule, or get bottlenecked/deferred?
4. Do margins/FCF continue to inflect (validating the quality-compounder framing)?
5. Will management resist value-destructive M&A and keep returning cash?

Falsification: The bull case is falsified by a 2026 cloud-revenue guide cut, CCB growth breaking below ~20% without an AI/consumption offset, or margin/FCF stalling. The bear case is falsified by CCB re-stabilizing ≥~20%, Joule/consumption revenue becoming visible in the P&L, and the 2027 migration converting — at which point ~13x forward EV/EBIT for a 30%-margin accelerating compounder looks like a clear mispricing.

Why the mispricing can persist (and why that is the opportunity). A fair challenge to any “it’s cheap” thesis: if it’s so obvious, why hasn’t it corrected? Several reasons the gap can stay open for a while — each of which is also why the entry is attractive: (1) momentum/quant flows mechanically sell a −46% name (Momentum loading zeroed; trend-followers and risk-parity sheets reduce); (2) narrative overhang — “AI kills SaaS” is an easy, un-falsifiable-in-the-short-run story that suppresses the multiple until disproven by data; (3) show-me dynamics — after a guide-down, investors demand two or three quarters of CCB stabilization before re-engaging, so even good results are discounted; (4) European-discount — SAP, as the rare European tech mega-cap, carries a structural valuation discount to US software that widens in risk-off tape. None of these is a fundamental reason; all are sentiment/flow reasons — which is precisely the kind of dislocation a patient, valuation-disciplined buyer is paid to exploit.

Verdict: Consensus is pricing the bear’s maturation premise as fact while ignoring the bull’s margin/FCF/capital-return reality and the embedded-expectations gap. The variant perception is that SAP’s de-rating reflects extrapolated narrative risk, not delivered fundamental deterioration — and that the burden of proof has shifted to the bears, who need the deceleration to continue against a backdrop of accelerating revenue, expanding margins, and a forced migration catalyst.

12. Fact vs. Interpretation Table

13. Open Questions

1. What is the exit rate of CCB growth by end-2026, and does the “wider range of outcomes” resolve up or down? This is the single most important unknown.
2. How much of the CCB deceleration is genuinely “sovereign/government deal timing + services one-off” vs. underlying demand softening? Management’s framing needs validation against realized cloud revenue over the next 2–3 quarters.
3. What is the actual monetization trajectory of Joule / Business Data Cloud? When does “consumption-related cloud revenue” become a visible P&L line?
4. Will AI agents expand or compress SAP’s revenue per customer? The <40% seat-based figure is reassuring but management-sourced; what is the trend?
5. What is non-IFRS EPS net of a fair SBC charge, and how fast is share count actually falling under the €10B buyback?
6. How disciplined will M&A remain? Any return to large, richly-priced deals would re-open the capital-allocation concern.
7. Does the consultant-capacity bottleneck push 2027 migration revenue right, smoothing/extending growth, or cap it?

14. What Must Be True (Bull and Bear, with Falsification Tests)

Bull case — what must be true:

1. Cloud growth stabilizes in the high-teens/low-20s (no cliff); CCB decel is gentle, not a roll-over.
2. Operating margin and FCF continue to inflect (toward mid-30s margin, FCF €13–15B over 3 yrs).
3. AI is net-neutral-to-positive for SAP’s economics (data-layer moat holds; consumption offsets any seat softness).
4. The 2027 migration converts to expanding per-customer revenue on roughly the expected timeline.
5. Capital discipline holds (buyback executed; no value-destructive mega-M&A).

Falsification of the bull: a 2026 cloud-revenue guidance cut; CCB growth printing <20% without an AI/consumption offset; or operating margin/FCF stalling for two+ quarters. Any of these breaks the “accelerating compounder mispriced as a laggard” thesis.

Bear case — what must be true:

1. CCB deceleration is the leading edge of structural growth maturation (the easy cloud conversion is done).
2. Agentic AI compresses seat-based SaaS value faster than SAP can monetize data/consumption.
3. “Sovereign/services” explanations are masking broader demand weakness.
4. ~20x earnings is merely fair for a maturing franchise — no margin of safety; the de-rating was a bubble correction.

Falsification of the bear: CCB re-stabilizing ≥~20%; Joule/consumption revenue becoming a visible, growing P&L contributor; the 2027 migration converting on schedule; and the multiple re-rating as fears fade — at which point ~13x forward EV/EBIT for a 30%-margin, +20%-cloud business is a clear mispricing.

The two cases share one fulcrum: the durability of cloud/CCB growth and the direction of AI’s effect on SAP. Everything else (margins, FCF, balance sheet, capital return, catalyst) currently favors the bull; the bear’s entire weight rests on the forward growth metric and the AI question — which is exactly why those are the items to monitor.

15. Source Appendix

See the Source Appendix below for the full, dated, URL-level citation list. Primary sources: SAP SE Form 20-F (FY2025) and 6-K interim filings (EDGAR, CIK 0001000184); SAP Q1 2026 quarterly statement and earnings call (23-Apr-2026); SAP Investor Relations (capital-return, dividend, guidance disclosures). Quantitative data: ROIC.ai (statements, ratios, EV, multiples, transcript), AZI price history and valuation-index percentiles, FactorsToday factor model. Qualitative/industry: SAPinsider, erp.today, Reuters/Yahoo Finance, CNBC, CIO, PRNewswire.

APPENDIX A — Standard Diligence Questionnaire

SAP SE (NYSE: SAP) — Standard Diligence Questionnaire Appendix

Supplemental to the research memo. Report date: 2026-06-13. EUR/IFRS unless noted. Fact / Interpretation / Assumption labels applied where it matters.

General

What thoughtful questions have other investors asked about this company? The questions clustering in 2026 sell-side notes and the Q1 2026 call Q&A: (1) Is current-cloud-backlog (CCB) deceleration the start of structural growth maturation, or transient (sovereign-deal timing + a services one-off)? (2) Does agentic AI threaten SAP’s seat-based cloud revenue? (3) How much of 2026 guidance is “protected” by the inorganic Reltio acquisition rather than organic strength? (4) Will the 2027 ECC end-of-maintenance migration convert to revenue on schedule, or get bottlenecked by consultant capacity? (5) Why doesn’t a 30%-operating-margin, +27%-cloud, ~€10B-FCF business deserve more than ~13x forward EV/EBIT? KeyBanc (Jackson Ader) explicitly pressed management on how much of guidance depends on Reltio and on the macro/geopolitical scenario assumed.

Cyclicality & Earnings Nature

Are earnings at a cyclical high or low? Neither extreme. Margins/FCF are inflecting up off a 2022–2024 transition+restructuring trough (IFRS operating margin troughed ~19% in 2023, now 26% and rising toward management’s mid-30s ambition). So profitability is early-cycle in its own structural ramp, not at a cyclical peak. (Interpretation.)

Driven by external environment or internal actions? Predominantly internal — the cloud mix shift and post-restructuring cost discipline drive the margin/FCF inflection. External factors (FX, enterprise IT-budget caution, geopolitics) affect timing and reported (vs constant-currency) optics, not the structural trajectory.

How stable are revenues? Very stable and increasingly so: cloud subscriptions + on-premise support (a >90%-gross-margin annuity) form the large majority of revenue; CCB provides ~12-month forward visibility. The lumpy/declining pieces (new licenses −33% YoY; services) are shrinking shares. (Fact.)

Outlook for products/services? Cloud ERP (S/4HANA public+private) is the growth engine; licenses run off by design; services are lower-quality and the source of a 2026 one-off setback. AI (Joule, Business Data Cloud) is early-stage optionality.

How big is the market — growing/shrinking, domestic/international? Enterprise application software is a >$300B, high-single/low-double-digit-growth global market; ERP at the high end is a SAP/Oracle duopoly. SAP is truly global (>180 countries; US was “particularly strong” in Q1 2026 alongside India, South Korea, Switzerland, UK). Market is growing.

Business Quality & Competitive Moat

Is the industry getting more or less competitive? Stable-to-intensifying at the edges (best-of-breed LoB players; hyperscalers up-stack; AI-native entrants) but the ERP core remains a defensible duopoly. The next competitive battleground is AI/data platforms.

How profitable is the business (ROIC, ROE)? ROE 15.4%, reported ROIC ~12.4% (2025) — solid; understated by a €29B goodwill balance. Underlying cash-on-tangible-capital returns are far higher. Gross margin ~73%; operating margin 26% IFRS (30% Q1’26). (Fact + Interpretation on the goodwill adjustment.)

How profitable is the industry / barriers to entry? Among the most profitable in tech; barriers are very high (switching costs, integration depth, mission-criticality, scale R&D, partner ecosystems). Few competitors at the high end.

Can the business be easily understood? The model (recurring cloud + support annuity) is understandable; the accounting requires care (IFRS vs non-IFRS, transition optics, SBC add-backs, goodwill). Use cash flow as the anchor.

Can it be undermined by foreign low-cost labor? No — this is IP/network/switching-cost-protected software, not labor-arbitrage-exposed. (Labor cost matters in the services line, not the core.)

Do brands matter? Yes, but as trust/installed-base rather than consumer branding — “nobody gets fired for buying SAP” for mission-critical ERP. The brand is the captive installed base and partner ecosystem.

Nature of competition / switching costs? Competition is on platform breadth, AI roadmap, and total cost of migration. Switching costs are extreme for the ERP core (multi-year, multi-€100M, business-risking replacements) — the heart of the moat.

Financial Condition & Balance Sheet

Assets not fully recognized on the balance sheet? Yes — the installed-base relationships, the partner ecosystem, the process/data IP, and brand are not capitalized; these are the real moat assets and are invisible on the balance sheet, which instead shows acquired goodwill (€29B). (Interpretation.)

Off-balance-sheet liabilities? Operating commitments and leases are capitalized under IFRS 16 (capital leases €1.68B). No unusual off-balance-sheet exposure identified; deferred revenue (€6.6B ST) is a favorable liability (cash received ahead of recognition).

How conservative is the accounting? Mixed. IFRS is conservative and the cash flows are clean; but management guides and is judged on non-IFRS, which adds back real SBC (€1.7B) and acquisition amortization — so non-IFRS flatters economic profit. 2024 IFRS earnings were depressed by restructuring; 2024 also had non-operating equity-investment (Sapphire) gains. Read cash flow. (Interpretation.)

How CapEx-hungry? Very capital-light: capex ~€0.74B (~2% of revenue), as cloud runs largely on hyperscaler infrastructure. This is a structurally superior cash model vs infrastructure-owning peers. (Fact.)

Capital Allocation & Management

How much FCF, and how is it used? ~€8.4B FCF (2025), guided ~€10B (2026). Used for: a growing dividend (€2.50/share, ~36% payout), large buybacks (new €10B program through 2027, atop a prior €5B), bolt-on M&A (€0.7B in 2025), and debt reduction (now net cash). (Fact.)

Significant acquisitions recently? WalkMe (~$1.5B, 2024), SmartRecruiters (2025), Reltio (2026), plus Signavio/LeanIX/Taulia earlier — disciplined, capability-focused tuck-ins. Historical caution: Qualtrics (~$8B, 2019) round-trip and a €29B goodwill pile from serial acquisition. (Fact + Interpretation.)

Buying back shares? Yes — €1.94B (2025), €2.11B (2024); a €10B authorization (2026–2027) ~5% of market cap, announced into the drawdown. (Fact.)

Issuing large amounts of stock to insiders? SBC €1.7B (2025), declining from €2.39B (2024) — material but shrinking; buybacks more than offset dilution (share count roughly flat-to-down at ~1,167M). (Fact.)

Compensation / incentive structure? Tied to cloud revenue, current cloud backlog, non-IFRS operating profit, and TSR. Alignment is good; the watch-item is that CCB-linked pay could encourage backlog optimization — validate revenue realization vs backlog. (Interpretation.)

Motivations of management? CEO Christian Klein (sole CEO since 2021), CFO Dominik Asam. Founder Hasso Plattner retired (May 2024); chair Pekka Ala-Pietilä. Management is executing a credible quality-over-optics strategy (margin/FCF/capital return) and has publicly defended the stock and doubled the buyback into weakness — a constructive signal. (Fact + Interpretation.)

Valuation & Market Data

ADR, MLP, or K-1 issuer? ADR — German foreign private issuer; 1 ADR = 1 ordinary share; files Form 20-F/6-K; reports EUR/IFRS. NYSE: SAP; home line SAP.DE (XETRA), ISIN US8030542042 (ADR). No K-1; standard 1099 dividend treatment (German withholding tax applies — a consideration for US holders). (Fact.)

Dividend policy? Progressive; €2.50/share FY2025 (+6.4%), ~36% payout, ~1.5% yield at ~$164. (Fact.)

How profitable is the business? Very — see above (73% gross, 30% Q1 operating margin, ~€10B FCF).

Is net income diverging from cash from operations? OCF (€9.16B) exceeds IFRS net income (€7.16B) in 2025 — favorable divergence (high cash conversion, deferred-revenue tailwind, non-cash SBC/amortization). In 2024 the divergence was extreme (OCF €5.21B vs NI €3.12B) due to restructuring/non-cash items. Cash flow is the cleaner read. (Fact.)

Risks & Downside

What would cause the stock to decline (further)? A 2026 cloud-revenue guidance cut; CCB growth breaking <20% without an AI offset; evidence AI is compressing seats; a value-destructive large acquisition; a broader software de-rating or macro downturn; margin/FCF stall.

Risk of catastrophic loss? Low. Net-cash balance sheet, mission-critical product, ~108k-customer diversification, recurring revenue. The realistic downside is multiple de-rating extension on disappointing growth, not solvency impairment. (Interpretation.)

Chance of total loss? Very low — would require simultaneous loss of the ERP franchise and the support annuity, implausible on any near/medium horizon.

Recent News & Events

Has the business environment changed recently? Yes — the dominant 2026 development is the ~46% ADR de-rating triggered by Jan-2026 CCB-deceleration guidance, amplified by the software/AI sector sell-off and AI-on-SaaS fears; partly offset by a strong Q1 2026 (cloud +27%, margin 30%, EPS +20%), maintained guidance, a doubled buyback, and the Teradata litigation settlement (€408M Q1 payout). (Fact.) (Note: AZI’s curated news feed returned no SAP coverage — foreign ADR; the timeline here is built from SAP IR, filings, and trade press.)

Significant acquisitions? Reltio (master-data, 2026); SmartRecruiters (2025); WalkMe (2024).

Change in accounting policies? None material identified; ongoing IFRS-vs-non-IFRS reporting cadence unchanged.

Recent changes — new markets/facilities/management? Founder Plattner retired (2024); product rebranding (RISE→Cloud ERP Private, GROW→Cloud ERP Public); launch of Business Data Cloud and Joule agent framework; ~10,000-role 2024 restructuring reallocating toward AI/cloud talent.

APPENDIX B — Source Appendix

SAP SE (NYSE: SAP) — Source Appendix

Report date: 2026-06-13. Primary sources first. Quantitative figures reconciled to SAP filings; third-party aggregators (ROIC.ai, AZI, FactorsToday) used as cross-checks and labeled as such.

Primary — SAP filings & disclosures

• SAP SE Form 20-F, FY2025 (annual report; filed ~2026-02-26), EDGAR CIK 0001000184. Mirrored locally to output/SAP/sources/20-F/. https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001000184&type=20-F
• SAP SE Form 6-K interim filings (FY2021–FY2026 quarterly statements & press releases), EDGAR CIK 0001000184. 40 6-Ks mirrored to output/SAP/sources/6-K/.
• SAP Quarterly Statement Q1 2026 (released 2026-04-23). https://www.sap.com/docs/download/investors/2026/sap-2026-q1-statement.pdf ; PRNewswire: https://www.prnewswire.com/news-releases/sap-quarterly-statement-q1-2026-302752280.html
• SAP Q1 2026 earnings call transcript (2026-04-23; Christian Klein, Dominik Asam, Alexandra Steiger) — via ROIC.ai transcript tool. Key data: CCB €21.9B +25%; cloud revenue +27%; Cloud ERP Suite +30%; total revenue €9.6B +12%; operating margin 30%; non-IFRS operating profit +24% to €2.9B; FCF €3.2B (incl. €408M Teradata settlement); IFRS EPS €1.66 (+9%), non-IFRS EPS €1.72 (+20%); FY2026 outlook maintained; “<40% of 2025 cloud revenue tied to named users.”
• SAP Investor Relations — capital return, dividends, guidance. https://www.sap.com/investors/en/stock/dividends.html ; share buy-back: https://www.sap.com/investors/en/stock/share-buy-back/2024.html
• SAP Supervisory Board dividend proposals: FY2024 €2.35 — https://news.sap.com/2025/02/sap-proposes-dividend-for-fiscal-year-2024/ ; FY2025 €2.50 — https://news.sap.com/2026/02/sap-proposes-dividend-of-e2-50-per-share/
• SAP FY2025 results / 2026 guidance (29-Jan-2026) and new €10B buyback (term through 2027-12-31).

Quantitative aggregators (cross-checks, reconciled to filings)

• ROIC.ai MCP — income statement, balance sheet, cash flow, profitability ratios, enterprise value, valuation multiples (FY2020–FY2025), company profile, Q1 2026 transcript. Third-party aggregated; EDGAR/20-F primary where they differ.
• AZI price history — https://azitrading.com/controls/download-data.php?t=SAP (adjusted OHLCV 1995→2026-06-12; 52w high $306.60 on 2025-07-09; 6/12/2026 close $164.18; 200-EMA $208; beta 0.95; alpha −0.10). AZI valuation_index percentiles (own ~10y history): composite ~39.7, P/E ~26.9, P/B ~41.5, P/S ~50.9.
• FactorsToday factor model — factorstoday.com/api: stock-loadings (Country: Germany 1.12, LowVolatility 0.53, Software 0.37, Cloud 0.33; Momentum & Value zeroed), leaderboard (1yr return −44.7%, Sharpe −1.36; 10yr return ~9%), related-stocks (SAPGF 0.96, NOW 0.75, BSY, HUBS, DT, RELX, CRM), specific vol (~22.9% annual).

Industry / qualitative

• SAP 2026 cloud outlook & share drop: Reuters/Yahoo Finance — “SAP’s 2026 cloud forecasts disappoint, shares endure biggest daily loss since 2020.” https://finance.yahoo.com/news/sap-meets-q4-revenue-forecasts-051512891.html
• S/4HANA 2027 maintenance deadline / RISE migration / Joule adoption: SAPinsider — https://sapinsider.org/blogs/sap-s4hana-erp-migration-trends-2026/ and https://sapinsider.org/map/2026-sap-s4hana-migration-cloud-ai-bottleneck/ ; erp.today — https://erp.today/rise-with-sap-cloud-erp-adoption-accelerates-ahead-of-2027-deadline-sapinsider-benchmark-report/ and https://erp.today/sap-q1-2026-earnings-cloud-ai-growth/ (Joule: 40+ agents, 2,400 skills, ~3% in production; AI #1 ERP-decision factor at 43% vs deadline 39%; ~30% live on S/4HANA Cloud Private up from 19%).
• 2024 restructuring (~10,000 roles, ~€2.2–3.1B): CNBC — https://www.cnbc.com/2024/01/23/sap-plans-job-changes-or-buyouts-for-8000-employees-in-restructuring-plan.html ; CIO — https://www.cio.com/article/3477211/sap-restructuring-to-impact-more-jobs-than-expected.html ; The Stack — https://www.thestack.technology/sap-layoffs-cost-3-billion-licence-deadline/
• Hasso Plattner retirement (May 2024): TechTarget — https://www.techtarget.com/searchsap/feature/SAP-leadership-turmoil-needs-to-settle
• Valuation / forward view: TIKR — https://www.tikr.com/blog/sap-stock-is-down-38-in-6-months-heres-what-could-drive-18-annual-returns ; Simply Wall St — https://simplywall.st/stocks/us/software/nyse-sap/sap/future
• Q1 2026 reaction / cloud detail: Futurum — https://futurumgroup.com/insights/sap-q1-fy-2026-earnings-show-cloud-erp-suite-acceleration/ ; StockTitan — https://www.stocktitan.net/news/SAP/

Cross-reference set

• Peer/valuation cross-reference set: Oracle (ORCL), Salesforce (CRM), ServiceNow (NOW), Workday (WDAY), Microsoft (MSFT), IBM, Adobe (ADBE), Intuit (INTU) — public filings and market data.