Manhattan Associates: 26% Cloud Growth Sits Inside 9.3% Total Revenue Growth, Existing Customers Supplied 75% of Q2 New Cloud Contract Value, and 40x SBC-Adjusted Owner Earnings Leaves No Margin of Safety

The second important divergence is RPO. Remaining performance obligations reached USD 2.47 billion at June 30, about 23% above the prior year and higher than USD 2.35 billion at March 31. More than 99% represents cloud-native subscriptions with terms longer than one year. Management says only approximately 39% will be recognized during the next 24 months, with most of the remainder recognized during the following 36 months. That means roughly USD 0.96 billion is scheduled over the first 24 months and about USD 1.51 billion sits beyond that bucket. Manhattan does not publish a next-12-month current-RPO measure, nor does it disclose a time series for average contract duration.

A more restrained interpretation follows than “RPO +23% means revenue will soon grow 23%.” Record bookings are real: the company explicitly called Q2 its third consecutive quarter of record bookings. New-logo contract value is real as well. Yet the long duration of the backlog mechanically lets a multi-year booking add several years of contractual value to RPO on day one. Because average contract-duration trends are undisclosed, the evidence cannot separate the increase perfectly into bookings volume versus duration. I think genuine demand is the larger explanation because bookings records, new-customer wins and rising RPO all point in the same direction, but the 23% RPO figure is a weaker near-term revenue indicator than a 12-month current-RPO metric would be.

The third divergence is profitability. A cloud mix shift usually creates an expectation of operating leverage. Manhattan did not deliver that in Q2. GAAP operating income fell from USD 73.8 million to USD 66.2 million even as revenue rose 9%; GAAP operating margin contracted from about 27.1% to 22.2%. Adjusted operating income rose only from USD 101.1 million to USD 103.9 million, leaving adjusted margin down from about 37.1% to 34.9%. GAAP EPS fell 8.6%, from USD 0.93 to USD 0.85, while adjusted EPS rose just 6.1%, from USD 1.31 to USD 1.39. The 6.1% figure is the correct year-over-year adjusted-EPS growth rate from the company's numbers.

Several costs explain it. During the first half, the combined cost of cloud subscriptions, maintenance and services rose faster than the associated revenue, including higher personnel, performance compensation and computer-infrastructure expense. Sales and marketing expense rose sharply as Manhattan expanded selling capacity. On June 1 the company then reduced global headcount by about 6%, taking an USD 8.3 million pretax restructuring charge. These are partly contradictory signals: management is simultaneously spending to accelerate sales and cutting elsewhere to refocus the cost structure. Cloud infrastructure itself is becoming a real operating input rather than the near-zero marginal delivery cost associated with perpetual licenses.

Stock compensation is another reason the adjusted figures deserve skepticism. Manhattan recorded USD 111.3 million of equity compensation in 2025, approximately 10.3% of revenue. Q2 2026 stock compensation was roughly USD 29.4 million, close to 9.9% of quarterly revenue, and full-year guidance puts it around 10.1% of revenue. Full-year 2026 guidance calls for GAAP EPS of USD 3.59–3.65 and adjusted EPS of USD 5.44–5.50; roughly USD 1.71 a share of that gap is the stock-compensation adjustment. I therefore use GAAP and SBC-adjusted owner cash flow as the primary valuation anchors and treat adjusted EPS as an operating comparison measure, not as economic earnings available costlessly to shareholders.

The balance sheet is unusually strong. Manhattan finished 2025 with USD 328.7 million of cash, no debt and no credit facility. Cash fell to USD 186.1 million by June 2026, largely while capital was being returned through repurchases. The company bought back roughly USD 275 million of stock in the first half; the share count fell to 58.3 million by late July from 59.65 million used in the March proxy. The repurchases also help absorb employee-equity issuance.

The historical business quality is genuine but needs to be interpreted through the change in business model. Revenue went from roughly USD 604 million in 2016 to USD 1.08 billion in 2025, with a dip through the early cloud-transition and pandemic years before accelerating from 2021 onward. Operating cash flow reached USD 389.5 million in 2025, compared with USD 295.0 million in 2024 and USD 246.2 million in 2023. The old license-plus-maintenance model generated excellent returns, but today's Manhattan earns a growing fraction of revenue through cloud infrastructure and a very large professional-services operation. Historical ROE therefore does not automatically describe the economics of the mature Manhattan Active model.

Competitive quality is stronger than the headline market share data available publicly. Gartner's 2026 WMS assessment again placed Manhattan in the Leaders quadrant; Manhattan says this was its eighteenth time being named a leader. It was also named a leader in Gartner's 2026 TMS evaluation for the eighth consecutive year. Blue Yonder, SAP, Oracle, Infios and others remain material WMS competitors. Precise, current individual vendor market shares by WMS, OMS and TMS are generally locked behind paid industry research, and the free public datasets are inconsistent. I will not turn vendor marketing claims into invented market-share precision.

The threats are concrete. SAP can offer warehouse execution as part of the broader S/4HANA estate. Its embedded EWM can directly use common ERP master data, reducing integration work for SAP-standardized enterprises. SAP itself is in a strong cloud transition, with Q2 2026 cloud revenue up 22% and current cloud backlog up 27%. Blue Yonder is now wholly owned by Panasonic after the 2021 acquisition; New Mountain Capital and Blackstone were former shareholders, rather than current co-owners. Blue Yonder widened its network scope through its 2024 acquisition of One Network Enterprises. Those competitors create two different pressures: SAP offers bundle and ERP integration; Blue Yonder offers broad supply-chain scope and can compete aggressively on package economics.

The physical end market is not in a simple expansion boom. UPS's Q2 U.S. ground average daily volume fell 3.5% and air volume fell 2.3%, partly because of deliberate reductions with its largest customer. C.H. Robinson's truckload volume grew only 0.5%, even while line-haul pricing jumped. Against that, J.B. Hunt reported 10% intermodal volume growth, Expeditors reported 14% airfreight tonnage growth, and Symbotic's latest quarterly revenue rose 22%. This is a selective recovery and network-reconfiguration cycle: freight scarcity, fuel, tariffs, AI infrastructure freight, automation and supply-chain redesign are supporting pockets of investment while broad parcel demand remains soft. Manhattan's booking records are therefore more impressive than they would be in an indiscriminate logistics boom, but enterprise-project deferral risk has not disappeared.

AI is potentially useful but not yet a valuation line item. Manhattan made its AI Agent Workforce commercially available in January 2026, and Agent Foundry is now available within Manhattan Active. Management said AI agents had reached more than 10% of the installed base through pilots or subscriptions and described AI revenue as small. Public pricing is not disclosed, and the company does not isolate AI ARR, bookings or revenue. Its claim that agentic tools can reduce standard implementation timelines by as much as 50% is interesting because the economics cut both ways: better implementations could improve sales and delivery margins, while fewer consulting hours could slow services revenue. I assign no incremental value to AI in the valuation.

At USD 197.39, the market is paying about 56.5 times trailing GAAP earnings, roughly 54.5 times the midpoint of 2026 GAAP EPS guidance and about 36.1 times adjusted EPS guidance. On a stricter owner-cash-flow measure that deducts both estimated maintenance capital spending and the economic cost of stock compensation, I estimate a multiple near 40 times. The U.S. 10-year Treasury yielded about 4.68% on August 14. Manhattan therefore still carries a significant duration premium even after its 2025–26 de-rating.

The share price tells the same story. Manhattan's year-end P/E moved from around 27 times in 2018 to roughly 77 times in 2020, 89 times in 2021 and about 76 times in 2023–24 as investors reclassified the company from mature supply-chain software to a cloud compounder. The multiple fell sharply in 2025 after management initially guided to only 2–3% revenue growth and a decline in adjusted EPS. Q2 2026 then reversed part of that pessimism: the company beat expectations, raised guidance and showed faster cloud/RPO growth, producing a sharp post-earnings share-price reaction. The current roughly 55 times trailing P/E is close to its long-run average but well below its cloud-era peak.

Qualitative portrait: high-quality company in transition. The transition has largely eliminated license revenue but has not yet removed maintenance runoff, reduced services to a mature share of sales, or proven that the cloud model can combine sustained high-teens subscription growth with rising economic margins. The market's central dispute is whether Manhattan is approaching a period when cleaner recurring growth and margin leverage become visible, or whether current cloud growth will decelerate before those benefits reach consolidated revenue and owner earnings.

Company vertical history and financial evolution

Manhattan began in 1990 in Manhattan Beach, California, before later moving its operational center to Atlanta. The founding group included software engineers with experience in manufacturing and distribution systems, including Deepak Raghavan, who had worked at Infosys on apparel-manufacturing systems. The original problem was narrower than today's “supply-chain commerce” language: distribution centers needed software that could control increasingly complicated inventory, picking and fulfillment processes. Its early PkMS warehouse-management product therefore started close to the physical operating floor rather than in corporate finance or ERP.

That origin matters. WMS software fails very visibly. An ERP reporting issue may frustrate a finance team; a warehouse execution failure can stop inventory from moving. Manhattan built its reputation in a domain where software correctness, throughput, integration with automation and operational edge cases matter more than elegant user-interface demos. That helps explain why switching costs later became meaningful: customers accumulated warehouse logic, interfaces, training, operating procedures and implementation knowledge around the system, not just data.

Manhattan incorporated in Georgia in 1998 as part of a reorganization around its predecessor software entity and came public that year. Its IPO sold 3.5 million shares at USD 15, implying gross proceeds of about USD 52.5 million before underwriting costs. The shares opened at USD 26 and closed their first trading day at USD 21.875, 46% above the offer price. The original public-market story was therefore a classic late-1990s enterprise-software growth story, not a recurring-revenue SaaS story.

The first stage, from foundation through the IPO, was product validation. Warehouse management was the center of gravity, particularly in retail, apparel and distribution. The business model was perpetual software license plus implementation and annual maintenance. That model generated attractive upfront economics for software vendors but made revenue timing dependent on license transactions and upgrades.

The second stage, roughly the late 1990s through the 2000s, was suite expansion. Manhattan added capabilities organically and through acquisitions, including Intrepa in 2000 and Logistics.com in 2002, broadening from warehouse execution toward transportation and wider supply-chain management. The strategic logic was simple: the more execution decisions Manhattan could connect, the harder it became for customers to treat WMS as an isolated box.

This period also established a pattern that persists today: Manhattan remained much smaller than ERP giants but could go deeper in the physical execution layer. SAP and Oracle could sell the corporate system of record; Manhattan competed by being better at difficult distribution operations. The company's later omnichannel opportunity grew out of that same niche because online commerce blurred the warehouse, store and order-routing boundaries.

The third stage came under Eddie Capel, who had joined Manhattan in 2000 and became CEO at the start of 2013. Capel inherited a profitable software company rather than a turnaround. His lasting contribution was to move the product strategy toward unified commerce and eventually accept the short-term financial pain of rewriting the software architecture for cloud delivery.

The inflection came in 2017, when Manhattan formally launched Manhattan Active and described the business as becoming cloud-first. Manhattan Active Omni was built around a cloud-native, evergreen, extensible architecture. The strategic decision was more consequential than converting customer contracts from one payment schedule to another: Manhattan had to move from release-based software installed at customer sites toward a continuously updated platform.

That transition temporarily made the financial statements look worse. Perpetual licenses recognize substantial value up front. Cloud subscriptions recognize the economics through time. Existing maintenance also falls as customers leave the old product base. A company can therefore improve its lifetime customer economics while depressing near-term reported revenue. Manhattan experienced exactly that phenomenon.

Revenue history makes the transition visible without requiring every annual data point. Revenue was around USD 604 million in 2016, fell to roughly USD 559 million by 2018, recovered to USD 617 million in 2019, dipped to USD 586 million during 2020 and then accelerated to USD 663 million in 2021, USD 767 million in 2022, USD 929 million in 2023, USD 1.04 billion in 2024 and USD 1.08 billion in 2025. The path resembles a business that first absorbed recognition-model pressure and then benefited as the cloud base became large enough to outrun the old-model runoff.

Historical figures are compiled from reported financial statements; Manhattan's latest 10-K confirms 2025 net income of USD 219.9 million and operating cash flow of USD 389.5 million, versus USD 218.4 million and USD 295.0 million respectively in 2024.

Manhattan Active Warehouse Management then pushed the architectural transition into the company's core franchise. By 2022 the company described new and upgrading WMS customers as receiving Manhattan Active Warehouse Management, a cloud-native, versionless application. The current product runs on a cloud platform, is subscription-based and receives continuing enhancements rather than traditional upgrade releases.

That move changed the competitive problem as well as the revenue model. Under perpetual software, a successful vendor could release major versions every few years and collect maintenance from customers who stayed on older releases. An evergreen service forces Manhattan to keep every major customer closer to the same code base. The reward is lower fragmentation and faster propagation of innovation; the cost is continuous infrastructure, security, reliability and R&D obligations.

The fourth stage, 2021 through 2024, was the financial validation of Manhattan Active. Cloud subscription revenue became large enough to drive consolidated growth, while high demand for implementations also expanded services. Investors responded by paying a materially higher earnings multiple. Manhattan's year-end P/E rose to about 89 times in 2021 and remained around the mid-70s in 2023 and 2024. The multiple was effectively capitalizing a future in which subscription growth would eventually dominate the income statement.

Investors then learned in 2025 that the transition would not move in a straight line. Services are large and project-driven. Manhattan's 2025 annual report says services revenue fell by USD 22.5 million, mainly because customer budget constraints shifted cloud-related professional-services work into later periods. Cloud subscriptions nevertheless rose by USD 70.9 million to USD 408.1 million. Maintenance fell another USD 8.3 million, driven primarily by migration demand in the Americas.

The January 2025 share-price shock was a capital-markets reset rather than an operating collapse. Management's original 2025 outlook pointed to only 2–3% revenue growth and a 4–6% decline in adjusted EPS. Manhattan stock dropped more than 20% after the guidance. The market had been paying a cloud-compounder multiple; a near-flat earnings outlook forced investors to reconsider how much transition friction remained.

The fifth and current stage began with management succession. Eric Clark replaced Eddie Capel as president and CEO on February 12, 2025. Capel became executive vice-chairman, then executive chairman, and moved to non-executive chairman status on January 1, 2026. Clark had run large technology-service organizations before joining Manhattan, giving him more experience with scaled enterprise sales and services than with Manhattan's founder-era product development.

Governance is conventional rather than founder-controlled. The March 2026 proxy showed Vanguard with about 11%, BlackRock 10%, AllianceBernstein 5% and FMR 5%. Directors and executives as a group beneficially owned fewer than 1% of shares; Clark personally was listed with 19,414 shares at that date. The absence of concentrated insider ownership means incentive alignment relies heavily on compensation design rather than founder equity.

That makes the scale of equity compensation relevant. Stock-based compensation rose from USD 71.6 million in 2023 to USD 93.2 million in 2024 and USD 111.3 million in 2025. The unrecognized compensation cost on outstanding RSUs was USD 144.4 million at the end of 2025, scheduled over about 2.3 years. This is recurring compensation, not a historical acquisition accounting artifact.

Buybacks partly counterbalance it. Manhattan repurchased roughly USD 275 million of shares in 2025 and another USD 275 million during the first half of 2026. The 2026 purchases occurred during a period when the stock spent substantial time below today's USD 197.39, so they look economically better ex post than repurchases made near the 2024 valuation peak. The more important long-run test is whether gross share count continues falling after employee issuance; by July 28 it had fallen to about 58.3 million.

The balance-sheet record is a genuine quality marker. Manhattan had USD 328.7 million cash and no debt at the end of 2025. It had no credit facility and said operating cash generation was sufficient to fund the business. Cash dropped to USD 186.1 million by June 2026 after buybacks and normal operating movements, but there is no leverage problem hiding behind the cloud investment program.

Cash generation has also generally exceeded accounting income. Using reported 2021–2025 figures, cumulative operating cash flow was roughly one and a half times cumulative net income. For the latest three full years alone, Manhattan generated USD 930.7 million of operating cash flow. That conversion is attractive, but some of it reflects economically non-free items: deferred subscription collections pull cash forward, while stock compensation is added back in the cash-flow statement even though issuing employees equity costs shareholders. In 2025, deferred revenue added USD 52.1 million to operating cash flow and stock compensation was USD 111.3 million.

Capital expenditure is tiny relative to revenue. The first half of 2026 required only about USD 5.1 million of property-and-equipment investment, versus USD 4.9 million a year earlier. Manhattan does not disclose maintenance versus growth capex. For owner-earnings purposes I conservatively treat all conventional capex as maintenance; the answer barely changes because the amount is so small. Cloud hosting commitments, R&D and labor are the more meaningful recurring investments, and they run through operating expenses rather than capex.

This creates an unusual software cash-flow profile. Conventional FCF looks exceptionally strong because physical capex is negligible. Economic owner cash flow is materially lower once stock compensation is recognized as a shareholder cost. Both statements can be true: Manhattan is highly cash generative and adjusted earnings still flatter its economics.

Historic high ROE, which may have brought Manhattan into quality screens, deserves the same caution. Decades of buybacks reduce book equity, mechanically raising ROE. The old perpetual-license model also had different recognition and infrastructure economics. I put more weight on cash generation, balance-sheet strength, retention/switching economics and incremental operating margins than on the historical ROE percentage itself.

Manhattan's share-price narrative has consequently gone through four identities. Early Manhattan was a specialist enterprise-software growth stock. By the 2010s it was a profitable best-of-breed supply-chain software vendor. From 2020 through 2024 the market increasingly priced it as an elite cloud compounder. The 2025 de-rating and 2026 rebound have produced today's hybrid: investors believe the cloud transition is strategically successful, but no longer assume that 20%-plus cloud growth will automatically produce 20%-plus consolidated earnings growth.

Historical P/E data illustrate the repricing. Year-end multiples were around 30 times in 2016–17, 27 times in 2018, 60 times in 2019, 77 times in 2020, 89 times in 2021, 59 times in 2022, approximately 76 times in 2023–24 and 48 times at the end of 2025. At about 55 times trailing earnings today, Manhattan is near its ten-year mean of roughly 57 times, perhaps around the lower-middle of annual observations, but that statistic hides the architecture change: the 2017 company and the 2026 company do not deserve an identical reference multiple merely because the ticker is the same.

The lasting capability proven across these eras is therefore not a particular revenue model. It is Manhattan's ability to stay technically relevant in operationally difficult supply-chain execution, first through packaged WMS, then broader execution software, omnichannel and finally a cloud rewrite. The cloud transition is the most important test because management deliberately cannibalized a profitable legacy base before a competitor did it for them.

Business model, moat, industry and horizontal competition

Today's Manhattan is formally one company rather than a collection of separately reported profit segments. Economically, however, it has four very different revenue streams: recurring cloud subscriptions, professional services, legacy maintenance, and the nearly immaterial license/hardware tail. The financial statements do not disclose segment operating profit by those revenue categories. More importantly, Manhattan combines the cost of cloud subscriptions, maintenance and services in its cost-of-revenue disclosure. It therefore does not publish standalone cloud gross margin, maintenance gross margin or services gross margin. Exact gross margins for those lines cannot be established from public information.

That disclosure limitation matters. Maintenance should theoretically be very high margin; services are labor-intensive and lower margin; cloud subscription carries hosting and customer-support costs but should produce attractive incremental economics at scale. Combining them makes it difficult to observe whether the cloud margin itself is expanding fast enough to offset maintenance attrition and implementation mix.

We can observe the blended result. Q2 2026 total cost of revenue was about USD 129.5 million against USD 297.8 million of sales, implying a gross margin around 56.5%, compared with roughly 57.3% a year earlier. The gross-margin decline occurred despite the larger cloud mix. During the first half, management attributed higher combined cloud/maintenance/services costs to compensation, performance incentives, computer infrastructure and travel. Cloud mix is therefore improving revenue durability before it has improved reported gross margin.

Services deserve much more attention than they normally receive in a SaaS analysis. At 45% of Q2 revenue, implementation and consulting are larger than subscriptions. Eighty-one percent of first-half professional-services revenue was related to cloud, up from 75% a year earlier. Services are therefore no longer a legacy accessory; they are the installation engine of Manhattan Active.

This creates both an advantage and a valuation complication. A difficult WMS rollout benefits from engineers who understand Manhattan's code and operational edge cases. That expertise improves successful deployment and creates customer intimacy. But professional services scale mainly with people, not with server economics. A business where 35–45% of steady-state revenue remains services cannot converge to the margin structure or revenue growth of a pure subscription SaaS company.

The natural steady state, in my model, is a company where cloud rises from about 43% of revenue now toward the high-50s over five years; maintenance falls into the low-single-digit mix; services decline from about 45% toward the high-30s; license/hardware become residual. Those mix changes would improve recurrence and probably margins, but they would not make total revenue grow at the subscription rate unless services also accelerate sharply.

Manhattan's cost structure reinforces that conclusion. Product development and R&D, enterprise sales capacity, implementation staff, customer support and central infrastructure are the large semi-fixed pools. Cloud computing is more variable with usage. Professional-services labor varies more slowly than projects because Manhattan needs specialized expertise available before bookings convert into deployments. Sales investment also leads revenue: first-half 2026 sales and marketing rose sharply as management expanded selling capacity.

June's headcount reduction illustrates the operating-leverage challenge. A 6% global reduction can improve the expense run rate, but management is simultaneously trying to accelerate bookings and implementation. The question is whether AI tooling, partner capacity and internal productivity let Manhattan serve a larger cloud base without simply rebuilding headcount.

The moat begins with deployment risk. Complex warehouses are tightly connected to conveyors, sorters, robotics, labor systems, carrier processes, ERP master data and order orchestration. Replacing the core execution layer requires re-validating those processes while continuing to ship customer orders. That makes a proven WMS installation sticky for reasons that are more operational than contractual.

The second moat is product depth. Manhattan's WMS franchise has survived multiple technology generations, and current Gartner assessments continue to put the company among the top WMS vendors. Gartner's market does not belong to Manhattan alone: Blue Yonder, SAP, Oracle, Infios and others remain credible alternatives. The evidence therefore supports a moat based on depth and execution credibility rather than monopoly.

The third is architecture. Manhattan Active was designed as a versionless cloud platform, and the company's applications now span warehouse, transportation, order management, store and commerce workflows. A customer that deploys multiple applications can move inventory and order decisions across a shared platform without stitching together as many separate release cycles.

The durable moat is execution depth plus switching cost, not a network effect. One retailer using Manhattan does not directly make the software more useful to another retailer. Scale helps fund R&D and build implementation expertise, but Manhattan lacks the self-reinforcing user network of a marketplace. Its competitive defense therefore depends on remaining technically better enough to justify best-of-breed deployment.

The fourth moat is the installed base. Existing customers represented 75% of Q2 new cloud subscription contract value excluding renewals. That concentration is partly the transition story, but it is also a distribution advantage: Manhattan does not have to win each cloud contract from a cold start. A legacy customer can migrate, add OMS or TMS, or expand the number of sites.

The same installed base is a finite resource. Management commentary indicates that fewer than a quarter of the on-premise base had begun the conversion process by mid-2026. That suggests years of migration opportunity. It also means investors should distinguish conversion bookings from underlying net-new market penetration; once migration is mostly finished, that source of cloud bookings must be replaced by new logos, expansion and pricing.

Industry growth is favorable. Third-party market research estimates the global WMS market at about USD 4.6 billion in 2025 and around USD 10.0 billion by 2030, a roughly 17% compound growth rate. Estimates differ by taxonomy, particularly on whether implementation, adjacent execution modules and smaller warehouses are included, so TAM should be treated as an order-of-magnitude indicator rather than precision.

The strongest secular driver is replacement rather than greenfield creation. Warehouses already exist and companies already manage orders. Growth comes from replacing legacy applications, moving them to cloud infrastructure, handling more complex omnichannel flows, automating distribution, integrating stores into fulfillment and redesigning networks after trade or inventory shocks. Cloud deployment itself was estimated to represent more than half of WMS deployments by 2025 in one industry dataset.

The economic profit pool sits in mission-critical software and recurring subscription/support revenue, while implementation consumes more labor and warehouse hardware carries lower economic returns. Manhattan's strategy is therefore rational: own the decision/execution software, tolerate services as necessary deployment infrastructure, and avoid becoming a capital-intensive robotics manufacturer.

Symbotic illustrates the adjacent layer. Its Q3 fiscal-2026 revenue grew 22% to USD 721 million as large retailers continued funding warehouse automation. Symbotic's machines can make Manhattan's software more valuable because more automated warehouses need sophisticated orchestration; they can also shift customer budgets between physical and software projects. The relationship is more complementary than directly competitive.

The cycle currently offers mixed signals. UPS continues to shrink some package volumes and reconfigure its network. FedEx is likewise emphasizing network efficiency and targets only about 4% annual revenue growth through 2029, although it continues to invest in technology and network redesign. C.H. Robinson says the freight market remains below a conventional mid-cycle level despite sharply higher truckload pricing.

Other signals are better. J.B. Hunt's Q2 intermodal volumes rose 10% as customers responded to fuel prices and tight capacity. Expeditors' airfreight tonnage rose 14%, helped by technology freight and changed trade lanes, even while ocean container volume was flat. This supports a “reconfiguration cycle” rather than a broad demand boom. Supply-chain uncertainty can actually increase the strategic value of Manhattan software, but weakened customer profits can simultaneously delay implementation approvals.

That duality is important for downside analysis. Supply-chain disruption is not automatically bullish for supply-chain software. Tariffs, geopolitical events or transportation scarcity create a need for better planning and execution; severe enough disruption can cut retailer and manufacturer budgets. Manhattan benefits most from uncertainty that is painful enough to force redesign but not painful enough to freeze capital projects.

Regulatory exposure is modest compared with banking, healthcare or defense software. The relevant constraints are enterprise data privacy, cybersecurity, AI governance and contractual obligations around customer data and availability. Geopolitical exposure is primarily indirect through customers and trade networks rather than semiconductor export controls or regulated licenses. The larger external risk is macroeconomic project deferral.

The horizontal comparison is best understood through four competitive archetypes rather than a feature checklist.

Blue Yonder is the nearest broad private competitor. It has deep WMS history and remains a Gartner WMS leader. Panasonic acquired full ownership in 2021 in a transaction valued at about USD 7.1 billion after first acquiring a 20% stake in 2020; New Mountain Capital and Blackstone were the sellers, so describing Blue Yonder in 2026 as jointly private-equity/Panasonic-owned would be stale. Its 2024 acquisition of One Network expanded Blue Yonder toward a multi-enterprise supply-chain network.

Customers can choose Blue Yonder when they want a broader planning-and-execution footprint or a commercially aggressive alternative to Manhattan. Gartner Peer Insights currently shows strong ratings for both vendors, with Blue Yonder's displayed score above Manhattan's, but the review counts differ substantially and voluntary reviews are not market-share data. The meaningful competitive message is that Manhattan has no right to assume customer preference is permanent.

SAP is the structural bundling threat. SAP EWM is designed for complex warehouses and can be embedded in S/4HANA. In an embedded configuration it uses shared master data with the ERP rather than requiring a separate distribution of that data. For a CIO standardizing heavily on SAP, that integration can outweigh Manhattan's best-of-breed depth, particularly in less differentiated facilities.

SAP also has plenty of financial momentum to fund competition. Q2 2026 cloud revenue rose 22%; Cloud ERP Suite revenue rose 25%; current cloud backlog reached EUR 22.9 billion and grew 27% at reported currency. SAP is pushing an enterprise-suite and AI strategy across finance, procurement, HR and supply chain, so Manhattan is defending a specialist profit pool against a vendor whose customer relationship begins much higher in the enterprise architecture.

Oracle poses a similar but somewhat different threat. Fiscal 2026 revenue reached USD 67.4 billion, up 17%, while total cloud revenue grew 39% to USD 34.0 billion. Oracle can bundle database, ERP, cloud infrastructure and applications. Its current growth is heavily influenced by cloud infrastructure and AI capacity, making the consolidated company a poor valuation comparable to Manhattan, but its size and enterprise footprint make it a real competitive alternative in supply-chain applications.

Descartes Systems is the cleaner listed economic comparator. It concentrates on logistics-intensive businesses through routing, customs, transportation, visibility and a large logistics network rather than Manhattan's warehouse/omnichannel center of gravity. In its latest quarter, Descartes revenue rose 15% to USD 193.6 million and adjusted EBITDA reached USD 89.8 million, a 46% margin. Its model is less services-heavy and more acquisition-oriented than Manhattan's.

‡Descartes' model does not map cleanly to Manhattan's cloud-subscription disclosure. §MANH headline FCF before deducting stock compensation; the SBC-adjusted owner-cash-flow multiple is materially higher.

Current prices and multiples are as of 2026-08-14 where available. Manhattan's premium over Descartes is substantial even though Descartes currently has faster consolidated revenue growth and a higher adjusted EBITDA margin. The market is paying Manhattan for WMS franchise quality, the migration runway and expected improvement in revenue mix.

Kinaxis provides another useful industry reference even though its specialty is planning and orchestration rather than warehouse execution. Its Q2 SaaS revenue rose 20%, total revenue rose 16%, ARR rose 19% and RPO rose 18%. The comparison shows that high-teens-to-20% recurring growth is available elsewhere in supply-chain software; Manhattan's 26% cloud rate is exceptional, while its 9% consolidated rate is not.

The niche Manhattan occupies is therefore specific. It is a best-of-breed execution specialist sitting below general ERP and above warehouse automation. It takes economic profit from legacy WMS providers and from suite vendors whenever an enterprise decides that distribution operations are differentiated enough to justify a specialist. SAP takes that profit pool back when architectural simplification is more valuable than best-of-breed functionality. Blue Yonder attacks it with comparable domain breadth. New AI-native entrants would have to overcome the operational switching and implementation barrier before their optimization algorithms alone become a serious substitute.

AI is more likely to extend Manhattan's moat initially than destroy it. The reason is data context and execution rights: an agent embedded in Manhattan Active can observe application state and call approved APIs inside workflows. A generic LLM does not automatically have that trusted operational access. Manhattan describes its agents as grounded in enterprise data and able to act through controlled APIs.

The long-term threat comes if AI lowers the cost of replacing incumbent workflow software. If a new platform can safely infer warehouse logic, generate integrations and migrate custom processes automatically, switching costs could fall. That is speculative today. No public revenue data show AI-native entrants taking material WMS share from Manhattan, and Manhattan itself already has commercially available agents.

Management's AI claims therefore belong in the moat discussion, not the valuation. Commercial availability is proven; adoption pilots are visible; monetization is not isolated. AI receives zero incremental value in my model until Manhattan discloses measurable AI ARR, contract value or revenue.

Current fundamentals and transition mechanics

Q2 2026 was fundamentally better than the headline 9% sales-growth number suggests and less profitable than the 26% cloud-growth number suggests. The correct reading comes from holding both facts simultaneously.

Cloud subscriptions accelerated to USD 126.7 million, up 26.2%. First-half cloud revenue was USD 243.8 million, up 25%. Management lifted full-year cloud guidance to about USD 505.5 million, implying roughly 24% growth from 2025's USD 408.1 million. Total revenue guidance rose to USD 1.160–1.166 billion, or about 7–8% growth.

That guidance spread is the mathematical fingerprint of the transition. At the midpoint, cloud could add almost USD 100 million of 2026 revenue. Consolidated revenue is expected to increase only about USD 82 million. The difference is absorbed by slower services, declining maintenance and license attrition. Some cloud growth is therefore replacing revenue rather than simply stacking on top of the prior base.

Maintenance is not yet trivial. It was USD 61.1 million in the first six months of 2026, down 9% from USD 67.2 million. Q2 alone fell 13%. At roughly 10% of current revenue, even a mid-teens decline takes 1–2 percentage points off consolidated growth each year.

License has reached the point where its percentage growth should be ignored. Q2's USD 1.9 million happened to be above the year-earlier USD 1.5 million, producing an arithmetical 26% growth rate. Economically, the line represents less than 1% of sales and is disappearing. A single contract can move the percentage violently without changing intrinsic value.

Services are the larger offset. First-half services revenue was USD 258.8 million, up 3%. Cloud-related work now represents roughly 81% of professional services, so strong cloud bookings eventually create implementation demand. Yet services depend on project timing, customer budgets and Manhattan's own capacity. The 2025 annual report already showed that customers can move projects between periods when budgets tighten.

This is why I reject the simple thesis that consolidated growth will eventually “converge” to current cloud growth. The maintenance drag should disappear, but cloud growth itself will probably decelerate as the base becomes larger, while services remain a large, slower-growing component. Removing a negative number does not force a weighted average to equal its fastest-growing component.

A simple base-case transition model illustrates the point. It is my model, not company guidance beyond 2026. I start 2026 at roughly USD 505.5 million cloud subscriptions, about USD 520 million services, roughly USD 115 million maintenance and about USD 22 million other revenue. I then assume cloud growth declines gradually from 21% in 2027 toward 14% by 2030, services grows 4–5%, maintenance declines 18–25%, and other legacy revenue fades.

The model produces consolidated growth around 8–10%, despite much faster subscription growth. Its implication is more important than its exact numbers: maintenance runoff becomes less relevant, but services remain too large for reported growth to catch cloud growth. Total revenue reaches low double digits only if cloud stays near or above 20% longer than I assume, services grows faster than mid-single digits, or new products expand the addressable market.

Maintenance also takes longer to disappear than one might infer from the tiny license line. At an average 20% annual decline from an estimated 2026 base around USD 115 million, maintenance would still be roughly USD 45–50 million in 2030. It would become economically small rather than literally zero. Management commentary that less than a quarter of the on-premise base has begun conversion is consistent with a multi-year runoff rather than a transition ending in 2027.

This is bullish in one sense. There is a long runway of customers who can still move into Manhattan Active. It is bearish in another: maintenance drag and migration-heavy cloud bookings will remain part of the reported numbers for years.

The new-logo evidence is improving. In 2025, initial-year revenue from first-time cloud customers rose to 8% of cloud revenue from 2% in 2024. In Q2 2026 about one quarter of new cloud subscription contract value excluding renewals was signed with new customers; over the half, 40% was new. That is enough to show Manhattan is not merely recycling its existing base.

It is not enough to conclude that most of the 26% recognized cloud growth is new business. Existing customers accounted for 75% of Q2 new contract value, and management says conversions alone were more than 40% of new cloud bookings. The cleanest forward KPI would be net-new cloud ARR or bookings split into migration, expansion and new logo. Manhattan does not disclose it.

RPO adds a second layer. USD 2.47 billion sounds enormous relative to roughly USD 505 million of annual cloud revenue, but only 39% is expected to convert during the next two years. The majority of the remaining amount is scheduled during the following three years. That distribution tells us the average signed commitment is long.

RPO is a real demand signal, but it is not a current-RPO signal. Investors should therefore reward a 23% RPO growth rate less aggressively than they would reward 23% growth in next-12-month contracted revenue.

The “third consecutive record bookings quarter” claim nevertheless matters. It means Q4 2025 through Q2 2026 produced a sustained bookings step-up rather than one isolated megadeal, based on management's definition. Manhattan does not publish the bookings dollars, so outsiders cannot independently calculate growth, concentration or average duration.

On balance the evidence says bookings volume has genuinely accelerated. RPO rose; management reported repeated records; new customers represented meaningful contract value; cloud revenue itself accelerated. The remaining uncertainty is how much of the RPO acceleration comes from larger/longer agreements. That uncertainty is material enough that I will not extrapolate 23% into the 2027 revenue line.

Recent quarterly trajectory reinforces the improvement. Late 2025 was marked by weak services and a major investor reset, even while cloud continued growing. Q4 2025 cloud subscription revenue reached USD 108.6 million. Q1 2026 rose to USD 117.1 million, versus USD 94.3 million a year earlier, while total Q1 revenue reached USD 282.2 million, up from USD 262.8 million. Q2 cloud then accelerated again to USD 126.7 million and total revenue to USD 297.8 million.

The market's current narrative has accordingly changed from “cloud transition stall” to “bookings reacceleration.” Following Q2, the shares rose sharply in after-hours trading after adjusted EPS and revenue exceeded consensus and guidance increased. By August 14 the stock had reached USD 197.39, substantially above its April trough.

Margin evidence is less celebratory. Q2 revenue growth of 9.3% produced a roughly 4.9-percentage-point contraction in GAAP operating margin and a roughly 2.2-point contraction in adjusted operating margin. This is not the operating leverage investors normally associate with an increasing subscription mix.

Part of the gap is investment. Manhattan expanded sales capacity; first-half sales and marketing expense rose to about USD 58.5 million from USD 41.0 million. Part is cloud cost: first-half combined cloud/maintenance/services cost grew as infrastructure and personnel spending increased. Part is restructuring. Part is stock compensation.

Full-year guidance implies some recovery. Manhattan expects GAAP operating margin around 24.2–24.4% and adjusted operating margin around 35.0–35.2%. Yet even at those levels, the spread between GAAP and adjusted profitability remains enormous because stock compensation is around a tenth of revenue.

For a long-term owner, the key test is not whether management can hit a 35% adjusted margin. It is whether cloud scale allows economic earnings after employee equity costs to grow faster than revenue. So far in 2026, that proof is incomplete.

Cash flow looks much better than GAAP income. Operating cash flow was USD 90.7 million in Q2, up from USD 74.0 million, and USD 174.7 million in the first half, versus approximately USD 149.3 million. Manhattan's subscription collections and deferred revenue support cash conversion.

Trailing twelve-month operating cash flow works out to roughly USD 414.9 million: 2025's USD 389.5 million less first-half 2025's roughly USD 149.3 million plus first-half 2026's USD 174.7 million. With conventional capex around USD 10 million on a trailing basis, headline FCF is roughly USD 405 million.

That gives a superficially attractive 3.5% FCF yield at the current market value. But trailing stock compensation is roughly USD 114 million. Deducting that economic cost leaves about USD 291 million of SBC-adjusted owner cash flow, or close to USD 5.00 per current share. The resulting owner-earnings yield is only about 2.5%.

Deducting all of it is intentionally conservative. Some equity compensation substitutes for cash salary and the buyback program reduces dilution, but shareholders still surrender value when stock is issued to employees. A valuation that adds all SBC back while treating buybacks as a shareholder return double counts the economics.

AI could improve this equation. Manhattan launched commercially available agents in January and says Agent Foundry can help create and deploy agents within the Active platform. Management reported pilots/subscriptions across more than 10% of the installed base and a high conversion rate from early pilots, though the sample is still small and AI revenue remains minor.

The most interesting AI claim is implementation productivity. Manhattan has said new agentic capabilities can reduce standard implementation timelines by as much as 50% in some cases. If verified at scale, fewer consulting hours per implementation could raise capacity and margins. It could also lower reported services revenue per project. The equity market may celebrate the first effect and initially dislike the second.

The bull/bear dispute can now be stated precisely. Bulls see record bookings, 23% RPO growth, 24–26% subscription growth, an underpenetrated migration base and rising new-logo activity. They expect maintenance drag to diminish while cloud takes a majority of revenue and AI/productivity improve delivery economics.

Bears see 75% of Q2 new cloud contract value coming from existing customers, more than 40% of new cloud bookings tied to conversions, a services business that still represents 45% of sales, contracting margins, SBC near 10% of revenue, and RPO whose majority is long-duration. They expect cloud growth to normalize before consolidated growth materially accelerates.

Both cases are grounded in the same disclosures. The decisive evidence over the next two years will be new-logo contribution and owner-earnings growth, not the absolute size of RPO.

Valuation, risks and tracking framework

The current valuation has de-rated but is not conventionally cheap. At USD 197.39, trailing GAAP P/E is about 56.5 times. Against the midpoint of 2026 GAAP EPS guidance, the multiple is about 54.5 times. Against adjusted EPS guidance it is 36.1 times. Market capitalization is close to 9.9 times 2026 revenue guidance.

Historical context can make that look more benign than it is. Manhattan's ten-year average trailing P/E is around 57 times, while its cloud-era highs reached roughly 75–90 times. Current valuation is therefore around the historical mean and far below the peak. But a historical average is not an intrinsic value model, particularly when the risk-free rate is now around 4.68% and the old comparison window includes extraordinarily low interest rates.

The multiple also depends on which earnings measure is used. Five-year operating-cash-flow conversion is approximately 1.5 times cumulative net income. That looks superb. The difference reflects deferred subscription billing, timing of cash taxes and substantial noncash stock compensation. In 2025 alone, Manhattan recorded USD 111.3 million of SBC and a USD 52.1 million increase in deferred revenue.

For owner earnings I start with trailing operating cash flow around USD 415 million. I deduct about USD 10 million for conventional capex, treating all of it as maintenance because management does not publish a maintenance/growth split. I then deduct approximately USD 114 million of trailing stock compensation as an economic shareholder cost. That gives approximately USD 291 million, or USD 4.99 per current share.

At USD 197.39, that is an owner-earnings multiple around 39.6 times and a yield about 2.5%. The corresponding unadjusted FCF multiple is only about 28.4 times. The gap is large enough that relying solely on headline FCF would materially overstate the margin of safety.

Descartes offers the clearest listed reality check. Its current trailing P/E is around 38 times, forward P/E about 28 times and P/FCF roughly 24 times; its latest revenue growth was 15% and adjusted EBITDA margin 46%. Manhattan deserves some premium for its WMS franchise and current 26% subscription growth, but today's premium leaves little room for that growth to disappoint.

Oracle's roughly 27 times trailing P/E is not directly comparable because its capital intensity, debt and AI infrastructure exposure are radically different. SAP is similarly too broad for a mechanical multiple comparison. Their significance is competitive: lower-multiple giants can subsidize supply-chain applications within much larger enterprise relationships.

The absolute valuation below uses owner earnings. The starting assumptions are approximately USD 5.0 of normalized owner earnings per share, a 9% discount rate and a three-year explicit horizon to 2029. The discount rate is only modestly above the current 4.68% Treasury yield because Manhattan has no debt and high business quality; a higher required return would reduce every value.

The conservative case still gives Manhattan a 29-times terminal owner-earnings multiple. That is not a distressed assumption; it reflects a company retaining category relevance, positive subscription growth and a net-cash balance sheet. Permanent-loss risk appears when the underlying franchise weakens enough that such a premium no longer applies.

The base case requires the cloud business to remain high-teens by the end of the horizon, consolidated revenue to stay around high-single/low-double-digit growth, and owner earnings to outgrow sales through modest operating leverage. It does not assign value to undisclosed AI revenue.

The optimistic case requires Manhattan to sustain roughly 20%-plus cloud growth much longer, increase new-logo contribution, preserve pricing, lower the services share without losing implementation quality and convert the mix shift into economic margin expansion. That combination can justify a 35-times terminal owner-earnings multiple. It is possible, but much of it is already embedded in today's premium.

This is valuation-scenario analysis within a research framework, not investment advice.

The expectation gap is therefore narrow. At USD 197, a conservative outcome produces almost no three-year return. The base case produces a high-single-digit annualized return, only a few points above a risk-free Treasury yield. Equity-style returns require the business to move materially toward the optimistic case.

Terminal multiple is the most fragile assumption. If the base-case 31-times owner-earnings multiple is cut to 70%, or 21.7 times, while the USD 8.00 2029 owner EPS estimate remains intact, present value falls from about USD 192 to approximately USD 134. That sensitivity is the clearest expression of Manhattan's valuation risk.

The flat-earnings test is harsher. If economic earnings do not grow for the next three years and the stock merely finishes at today's owner-earnings multiple, the investor receives approximately a zero annual return because Manhattan pays no regular dividend. Zero is below the approximately 4.68% 10-year Treasury yield. If the multiple normalizes, the return becomes negative. There is no margin of safety at this buy price.

Today's price also sits well above the conservative present value around USD 155, so margin of safety to that scenario is zero. Requiring at least a 20% discount to conservative value produces a buy zone around USD 120–125. That is severe, but Manhattan actually traded near that level during the 2026 de-rating, which shows the zone is not purely theoretical.

Margin-of-safety sufficiency verdict: none.

The permanent-capital-loss risks are concentrated rather than numerous.

First is growth-quality risk, with medium probability and high impact. The observable indicators are new-logo share of signed cloud value, first-time customer revenue, maintenance runoff and cloud growth. If migration continues to represent more than 40% of bookings while new-logo contribution fails to rise, Manhattan can report attractive cloud growth for several years yet enter the post-migration period with only modest organic growth. The transmission path is cloud deceleration, lower terminal growth, then multiple compression.

Second is RPO-duration risk, medium probability and medium-to-high impact. Only 39% of RPO is expected over the next 24 months, and no next-12-month measure is disclosed. If future RPO growth is increasingly driven by contract duration rather than annualized contract value, the stock could be pricing acceleration that does not reach near-term revenue. Watch the 24-month recognition percentage and any future disclosure of contract duration.

Third is competitive bundling, medium probability and high impact. SAP EWM can sit inside the S/4HANA environment and share ERP master data; Blue Yonder has comparable WMS credibility and a wider network strategy. The risk would first appear in new-logo win rates, pricing or a slowing new-customer share rather than immediate churn of installed warehouses. The ultimate transmission path is lower subscription growth and a lower premium multiple.

Fourth is economic-margin risk, medium-to-high probability and high impact to valuation. Q2 showed that cloud mix alone is not sufficient to expand margins. Sales investment, infrastructure, services labor and SBC can absorb the benefit. SBC around 10% of revenue is particularly important because adjusted EPS removes most of it. The observable indicator is owner cash flow after SBC, not adjusted operating margin.

Fifth is valuation-duration risk, high probability and high potential share-price impact, although it does not necessarily damage the business. Manhattan's owner-earnings yield is around 2.5% while the ten-year Treasury yields about 4.68%. A move from roughly 40 times owner earnings toward 25–30 times could wipe out several years of operational progress.

Sixth is customer-budget risk, medium probability and medium impact. UPS parcel volumes remain weak, freight indicators are uneven and Manhattan already acknowledged 2025 customer budget constraints that postponed services work. A deep economic slowdown can delay multi-year supply-chain programs despite their positive return on investment.

Financial distress is low probability. Manhattan has no debt, generates substantial cash, carries no large physical inventory and requires little conventional capex. The more realistic financial risk is capital-allocation leakage through overly expensive buybacks or persistent equity compensation, not insolvency.

Governance risk is also moderate rather than acute. The new CEO has only been in the role since February 2025, insider ownership is below 1% and compensation relies heavily on equity. The counterweight is Capel's continued presence as non-executive chairman and the company's strong operating history.

Positive catalysts over the next 12 months are straightforward: cloud revenue remains at or above the revised 24% full-year trajectory; RPO stays near 20% growth while the 24-month recognition mix does not lengthen; new-logo contract share rises; the 6% headcount reduction and implementation tooling restore operating leverage; or Manhattan begins disclosing paid AI adoption large enough to matter financially.

Negative catalysts are equally measurable: cloud growth slips below 20%; RPO growth drops into the low teens; services contract without corresponding margin improvement; new-logo mix falls below 20%; GAAP operating margin remains around 22% despite growing subscription mix; or SAP/Blue Yonder wins begin showing up in management commentary.

†Manhattan had not announced a Q3 reporting date on its IR events page as of the research date. MarketBeat's schedule-based estimate was October 27, 2026, while another historical-calendar service estimated the prior week; investors should replace the estimate with the company's announcement when it appears.

The dashboard is designed to distinguish transition success from optical cloud growth. Cloud growth and RPO describe demand; new-logo share describes the quality of demand; maintenance shows how much old-model drag remains; services and GAAP margin show whether deployments are economically scaling; SBC captures the difference between adjusted earnings and shareholder economics; interest rates determine how expensive the same earnings stream looks in capital markets.

Cross-synthesis, research conclusion, sources and uncertainties

Vertically, Manhattan has proven one capability more convincingly than any other: it can reinvent the software architecture beneath a mission-critical franchise without losing the franchise. The company began with packaged warehouse software, broadened into supply-chain execution, built omnichannel functionality and then cannibalized its own license-plus-maintenance model to move into a cloud-native platform. Many mature enterprise-software vendors choose incremental hosting conversions because a complete rewrite risks both revenue and customers. Manhattan accepted that risk.

Its success was not merely an era tailwind. E-commerce, omnichannel complexity and cloud adoption were substantial tailwinds, but they were available to Blue Yonder, SAP, Oracle and numerous smaller vendors. Manhattan retained a strong WMS competitive position through those shifts. The evidence of capability is the combination of Gartner positioning, a surviving installed base, rising cloud contract value and a transition that has almost eliminated license revenue without destroying cash generation.

The historical shareholder experience did contain a major multiple tailwind. Investors moved from valuing Manhattan around 25–30 times earnings in the late 2010s to 70–90 times during the cloud re-rating. A meaningful portion of the 2020–24 share-price advance therefore came from paying more for each dollar of earnings, not just from business growth. The 2025 collapse showed what happens when that willingness reverses.

Those facts matter now because the multiple still prespends success. About 56 times GAAP earnings and about 40 times estimated owner earnings require more than “Manhattan remains a good WMS company.” They require sustained subscription growth, cleaner new-logo penetration, successful conversion of the remaining installed base, margin improvement and continued category leadership.

Horizontally, Manhattan's real advantage over SAP is depth in supply-chain execution. Its disadvantage is that SAP can make WMS one part of a broader architectural decision. That structural asymmetry will never disappear. Manhattan must remain sufficiently better that customers accept another enterprise platform in the stack. SAP only has to become good enough for an SAP-centric organization to choose integration simplicity.

Against Blue Yonder the competition is more direct. Both have long WMS pedigrees. Blue Yonder's Panasonic ownership supplies a patient strategic parent and its One Network acquisition broadens multi-enterprise scope. Manhattan's answer is the unified Manhattan Active architecture, deployment expertise and execution reputation. Neither competitor has obviously eliminated the other's market.

Descartes shows why Manhattan's valuation deserves discipline. Descartes has a different logistics niche, but its current revenue growth and cash profitability are strong while its valuation is materially below Manhattan's. Investors are paying specifically for Manhattan's execution moat and cloud transition, not simply for exposure to supply-chain software.

The market is most likely misjudging one of two things. The optimistic misjudgment is that consolidated growth looks weaker than the underlying franchise because maintenance attrition still masks subscription expansion. As that drag shrinks, economic quality improves even if reported growth rises only modestly.

Its pessimistic counterpart is more subtle: investors may expect reported growth eventually to approach today's 24–26% cloud growth. My transition model does not support that. Cloud should become a majority of sales, but services remain large and cloud itself is likely to slow with scale. The most plausible steady-state outcome is roughly high-single-digit to low-double-digit consolidated growth, not 20%-plus.

That does not make Manhattan a poor company. A 9–11% revenue grower with a strong mission-critical franchise, net cash, high recurring revenue and economic margin expansion can compound shareholder value for a long time. It simply cannot be valued as though 26% is the permanent growth rate of the whole enterprise.

One-year variables are bookings, RPO quality and margins. The three-year variables are new-logo penetration, migration economics and the services share. The five-year variable is whether Manhattan remains the preferred specialist execution layer once SAP, Blue Yonder and AI-enabled architectures have had another technology cycle to attack the position.

Manhattan's conversion runway is useful but not permanent. It acts like an internal sales channel into known customers. Once that pool diminishes, Manhattan must show that its cloud platform can grow organically through net-new logos and cross-sell at a rate high enough to support the valuation. The Q2 new-customer figure of 25% is therefore one of the most important disclosures in the entire filing.

RPO needs the same discipline. USD 2.47 billion and 23% growth sound extremely bullish, but the recognition schedule spans years. Because the company does not publish a one-year current-RPO number or average contract duration, investors cannot yet prove how much of the divergence from revenue reflects acceleration versus longer commitments. Record bookings tilt the interpretation bullish; disclosure quality prevents certainty.

Margins are where the next phase must prove itself. Cloud is already 43% of quarterly revenue. If cloud passes 50% and Manhattan still cannot improve GAAP or SBC-adjusted owner margins, the thesis that subscription mix automatically creates economic leverage will have failed. The company then becomes a durable but services-heavy vertical-software business deserving a lower terminal multiple.

AI is an option, not the thesis. Commercial products exist; adoption is beginning; management says revenue remains small. AI could increase win rates, reduce implementation cost and deepen product integration. It could also reduce services billings and eventually lower switching barriers industry-wide. Assigning a separate AI valuation today would turn an unmeasured operating initiative into a financial asset without evidence.

The external cycle is better than raw package volumes suggest but weaker than a broad boom. Physical logistics companies are simultaneously cutting capacity, raising prices, redesigning networks and investing selectively. Manhattan's record bookings amid that environment strengthen the case that supply-chain modernization is strategic rather than purely cyclical. A recession still matters because large WMS programs can be delayed even when their eventual ROI is attractive.

The stock price currently sits close to what I regard as a base-case hold range. It is neither at the 2024 cloud-euphoria multiple nor at the panic valuation reached in spring 2026. That leaves existing holders with a credible path to modest returns if execution continues, while offering a weak margin of safety to a new buyer.

Core bull reasons

• Cloud subscription revenue grew 26% in Q2 and full-year cloud guidance is about 24%, showing that the Manhattan Active transition has substantial momentum despite only 7–8% consolidated growth.
• RPO reached USD 2.47 billion, up about 23%, and management reported a third consecutive bookings record, supporting multi-year revenue visibility.
• Roughly three quarters of the on-premise customer base had not yet begun migration according to management commentary, leaving a long internal conversion runway.
• Manhattan remains among Gartner's WMS leaders while also holding a leadership position in TMS, supporting the claim that the franchise survived the cloud rewrite.
• The company has no debt, strong operating cash flow and very low conventional capital requirements.

Core bear reasons

• Existing customers represented 75% of Q2 new cloud contract value and conversions represented more than 40% of new cloud bookings, so migration still accounts for a large part of the cloud-growth engine.
• Only about 39% of RPO is expected to be recognized during the next 24 months, making 23% RPO growth a less powerful near-term indicator than the headline suggests.
• Q2 adjusted operating margin contracted by about 220 basis points and GAAP operating margin by nearly 490 basis points even with 26% cloud growth.
• Stock compensation runs around 10% of revenue, creating a large gap between adjusted EPS and economic shareholder earnings.
• At roughly 40 times SBC-adjusted owner earnings, Manhattan requires sustained premium growth while the ten-year Treasury yields about 4.68%.

The first pre-mortem is competitive. Imagine that by 2028 SAP makes EWM sufficiently seamless inside its cloud ERP estate and Blue Yonder discounts broad execution packages. Manhattan's new-logo share falls below 15%, cloud growth drops toward 12% and services growth turns negative as fewer greenfield implementations enter the funnel. Economic operating margin fails to improve. Owner earnings remain around USD 5–6 a share and investors reduce the multiple from roughly 40 times to 22 times. A USD 110–130 stock would represent a decline of roughly 35–45% from today's price. The operating business would still be profitable; the loss would come from simultaneous growth and multiple compression.

The second pre-mortem is an RPO/macro script. By 2027 it becomes clear that part of the 2025–26 RPO acceleration came from long contract duration. Retail and logistics customers defer deployment work during a downturn. RPO growth falls to single digits, total revenue growth drops toward 3–4%, services contracts and SBC remains close to 10% of sales. GAAP operating margin falls toward 20%. If GAAP EPS settles around USD 4 and the market pays 25 times rather than today's 50-plus, the share price approaches USD 100. The permanent-loss path is therefore entirely plausible without bankruptcy or technological obsolescence.

The original judgment should be overturned positively if Manhattan proves three things together: net-new customers become roughly a third or more of cloud contract value on a sustained basis; cloud remains above 18–20% after migration runoff materially shrinks; and owner earnings after SBC grow at least in the low teens while GAAP margin expands. That combination would justify treating Manhattan as a durable cloud compounder rather than a transition story.

The judgment should be overturned negatively if cloud growth falls below 15% while a material maintenance base remains, new-logo contract value drops below 20% for several quarters, RPO growth becomes single-digit, or GAAP margin fails to recover above the mid-20s despite cloud crossing half of revenue. Those outcomes would imply that the premium valuation had been capitalizing a transition benefit that never fully reached shareholder economics.

Manhattan is a company I would be comfortable calling high quality at the operating level. The warehouse franchise has survived three decades of technology change, the cloud rewrite appears strategically successful, the balance sheet is clean, bookings are strong and the company is winning genuinely new customers. The weakness lies in translating those advantages into the valuation denominator: total revenue still grows only high single digits, services consume nearly half the mix, economic margins have yet to show the expected cloud leverage and SBC remains large.

At USD 197.39, the stock prices a substantial amount of future success. My base present value is around USD 192, which makes today's quote defensible for an existing long-term holder but unattractive as a margin-of-safety entry. The conservative value is closer to USD 155; an entry with at least a 20% discount to that outcome lands around USD 120–125. A buyer waiting for that price risks missing continued compounding and another cloud-driven re-rating, but the opportunity cost is preferable to accepting a roughly 2.5% current owner-earnings yield with a 4.68% Treasury alternative.

【Company-profile scores】

• Fundamental quality: high
• Growth: medium
• Moat: strong
• Financial soundness: strong
• Management credibility: medium
• Valuation attractiveness: low
• Risk level: medium
• Suitable investor type: long-term growth

【Investment rating】

• Rating: Hold
• One-line thesis: Cloud bookings and RPO support durable growth, but 56x GAAP earnings and roughly 10% stock compensation leave little margin of safety.
• Ideal buy price:

【Ideal Buy Price】120-125 USD

Basis: at least a 20% margin of safety below the roughly USD 155 value implied by the conservative owner-earnings scenario.

• Acceptable hold price: 170-215 USD
• Clearly overvalued price: 292-310 USD
• Current-price classification: acceptable hold
• Whether to wait for a better price: yes. For new capital, I would require roughly USD 125 or below while cloud growth remains at least 18%, RPO growth remains at least mid-teens and competitive evidence does not show material WMS share loss. The opportunity cost is that record bookings could keep earnings compounding and prevent the stock from revisiting that level.
• Target holding horizon: 3–5 years
• Expected annualized return: conservative about 0.5%; base about 7.9%; optimistic about 19.0% over the three-year explicit valuation horizon, before taxes and assuming no dividend.
• Max-loss risk: roughly 45–55% in the pre-mortem case where cloud growth falls toward low teens, economic margins stagnate and the owner-earnings multiple compresses to the low-20s.
• Reassessment-trigger signals: cloud growth below 18% for two consecutive quarters; new-customer share of new cloud contract value below 20% for two quarters; RPO growth below 15% without a shorter recognition profile; GAAP operating margin below 22% for two quarters; SBC above 11% of revenue without proportionately faster owner-earnings growth.

【Valuation Range】

• current: 197.39 (close as of 2026-08-14)
• bear (conservative · ideal buy zone): [120, 125]
• base (fair · acceptable hold zone): [170, 215]
• bull (optimistic · above the clearly-overvalued line): [292, 310]

The principal research uncertainties are unusually important because they map directly to the multiple. First, Manhattan does not disclose recognized cloud revenue split between legacy migrations, expansion of existing cloud customers and true new logos. Second, it does not publish next-12-month current RPO or average contract duration, preventing a clean separation of bookings acceleration from duration. Third, cloud, maintenance and services gross margins are combined in the filing, so standalone cloud unit economics cannot be measured. Fourth, exact current vendor market shares by WMS, OMS and TMS are not reliably available in public industry data; paid ARC/Gartner-style datasets contain more granularity than the open sources. Fifth, AI pricing, ARR and attributable revenue are not disclosed, so AI has been assigned no incremental valuation.

Source hierarchy for this report puts Manhattan's June 2026 10-Q, 2025 10-K, 2026 proxy and Q2 earnings release first. Those establish revenue mix, RPO, customer mix, cash flow, SBC, governance and guidance. Historical company materials and Manhattan's corporate timeline establish the cloud transition and product history. SAP, Oracle, Descartes, Kinaxis, UPS, C.H. Robinson, J.B. Hunt, Expeditors and Symbotic disclosures are used for competitive and end-market cross-checks. Gartner-related vendor disclosures and independent industry reports are used only to establish competitive positioning and market structure, not fabricated precise market shares. Price, historical valuation and Treasury data are dated to the latest available trading day.

Other tickers mentioned

• DSGX.US — Descartes Systems is the closest listed logistics-software valuation reference and currently trades at lower earnings and FCF multiples than Manhattan.
• SAP.US — SAP's S/4HANA and embedded EWM strategy is Manhattan's clearest ERP-bundling threat.
• ORCL.US — Oracle provides a second large-suite competitive threat across enterprise cloud and supply-chain applications.
• KXS.TO — Kinaxis is a supply-chain planning and orchestration specialist whose SaaS and RPO growth provide an adjacent vertical-software benchmark.
• 6752.TSE — Panasonic wholly owns Blue Yonder, Manhattan's closest broad private WMS competitor.
• SYM.US — Symbotic provides the physical warehouse-automation counterpart to Manhattan's execution-software layer.
• CHRW.US — C.H. Robinson's latest truckload data help gauge the state of freight demand among Manhattan's logistics customers.
• UPS.US — UPS's declining package volumes show that Manhattan's bookings acceleration is occurring without a broad parcel-volume boom.
• FDX.US — FedEx's network redesign and moderate long-term revenue target illustrate customers prioritizing technology-driven efficiency over simple capacity expansion.
• EXPD.US — Expeditors' airfreight growth and flat ocean volume illustrate the current selective, trade-lane-driven logistics recovery.
• JBHT.US — J.B. Hunt's 10% intermodal-volume growth provides one of the stronger current freight-cycle signals.

This report is based on public information and does not constitute investment advice. Markets carry risk; invest with caution.