Cognite

1.1 Corporate Identity, Executive Leadership, and Governance Architecture

Cognite operates as a global software-as-a-service provider delivering industrial data operations and generative artificial intelligence capabilities to asset-intensive sectors including energy, manufacturing, and utilities. Originally founded in Oslo, Norway, the company established its global corporate headquarters in Arizona while maintaining expanding development hubs in Norway and India alongside commercial branch offices in Houston, Tokyo, and Abu Dhabi. Executive leadership is directed by Chairman and Chief Executive Officer Girish Rishi, who succeeded co-founder and former Chief Executive Officer Dr. John Markus Lervik. Governance is anchored by industrial holding group Aker as the controlling majority stakeholder alongside growth equity partner TCV, whose general partner Jake Reynolds holds a seat on Cognite's board of directors following their landmark 2021 funding round. In 2025, Cognite undertook an executive management scale-up by recruiting Chirayu Shah as Chief Product Officer, Sandy Joung as Chief Marketing Officer, and Bill Owen as Chief Services Officer to manage global expansion across core platform and AI agent offerings.[CO001, CO002, CO003, CO004, CO005, CO006, CO007]

1.2 Capitalization Trajectory, Operating Scale, and Financial Performance

Cognite attained unicorn status in May 2021 after securing a $150 million funding round led by growth equity firm TCV, valuing the enterprise at $1.6 billion post-money. This financing supplemented earlier institutional venture backing from Accel and majority capitalization from industrial parent Aker. Operating scale expanded substantially through fiscal year 2025, with management announcing that Cognite surpassed $170 million in annual revenue ending FY2025, accompanied by 36% year-over-year expansion in annual recurring revenue bookings and a net revenue retention rate of 150% among newer customer cohorts. Global associate headcount increased by 21% year over year to exceed 800 employees across the Americas, EMEA, Asia, and Japan. Despite strong operational momentum and self-reported performance indicators, third-party databases report conflicting historical launch dates and uncorroborated later-stage valuations reaching $3.1 billion, while primary audited balance sheets and equity cap-table allocations remain undisclosed under private ownership.[CO008, CO009, CO010, CO011, CO012, CO013, CO014, CO015]

1.3 Product Evolution, Industrial AI Deployment, and Milestone Trajectory

Cognite's technology platform centers on Cognite Data Fusion, an industrial data operations and contextualization engine deployed across enterprise accounts including bp, Mitsubishi, and Saudi Aramco. The company accelerated its product roadmap through quarterly enhancements, culminating in the launch of Cognite Atlas AI, an industrial agent workbench, and Cognite Flows, an action layer integrating operational workflows and agentic AI. Customer adoption metrics reported for fiscal year 2025 show the industrial knowledge graph surpassing 80 trillion time-series data points in context, representing 48% growth, while AI token usage increased nearly 500% and monthly active users grew 26%. However, customer proofs remain concentrated in heavy industry, and customer count disclosures remain largely private. Cognite committed to a Moonshot objective targeting $100 billion in cumulative customer business value by 2035, though audited validation of financial return metrics across the complete customer base represents an ongoing diligence gap.[CO001, CO009, CO010, CO016]

1.4 Exhibits

2.1 Market Boundary, Category Scope, and Industry Adjacencies

Cognite defines its primary addressable category within the industrial DataOps and operational intelligence software landscape, specifically targeting asset-intensive industries including energy, chemicals, power, utilities, and pharmaceuticals [CM001]. The platform's architectural core focuses on unifying unstructured, operational, and engineering data into contextualized knowledge graphs, distinguishing it from generic enterprise data lakes or horizontal cloud storage infrastructure [CM002]. Adjacent market segments include historical time-series historians, specialized operational analytics engines, and enterprise AI orchestration suites [CM003]. Excluded from this boundary are transaction-oriented enterprise resource planning systems, edge hardware controllers, and generic business intelligence dashboards that lack domain-specific asset contextualization [CM004]. Cognite's deliberate boundary strategy positions Cognite Data Fusion and Atlas AI directly where asset owners seek to transition from passive descriptive telemetry toward autonomous agentic workflows [CM005].

2.2 Market Sizing Lenses and Addressable Spending Layers

Evaluating the addressable market for industrial DataOps and operational AI requires multi-tier sizing lenses rather than a single monolithic top-down projection [CM006]. The broadest lens encompasses global industrial software and enterprise operational technology modernization, representing a vast total addressable market spanning global energy and discrete manufacturing facilities [CM007]. Within this macro spend, the serviceable addressable market isolates industrial intelligence, contextualization platforms, and DataOps layers deployed by global asset operators to bridge information silos [CM008]. Cognite's serviceable obtainable market reflects late-stage growth enterprise deployments, evidenced by $170 million in 2025 revenue and an established footprint across heavy asset enterprises [CM009]. However, comprehensive third-party independent consensus regarding the precise dollar ceiling of pure-play industrial DataOps remains an evidence gap given overlapping definitions across analyst categories [CM010].

2.3 Buyer Segmentation, Procurement Paths, and Commercial Friction

Enterprise buyer personas within asset-heavy organizations are divided across corporate IT leadership, operational technology heads, and plant engineering directors [CM011]. Digital transformation initiatives and enterprise industrial AI architectures are typically championed by Chief Digital Officers and Chief Information Officers seeking high-ROI operational use cases and centralized data governance [CM012]. In contrast, departmental plant managers and operations engineers prioritize uptime, process safety, and direct operational workflow enhancements, often relying on partner systems integrators for deployment execution [CM013]. Commercial friction remains significant due to enterprise subscription packaging; Cognite Data Fusion pricing requires custom quotes, master service agreements, and mandatory professional services or implementation accelerators, preventing transparent self-serve adoption and prolonging procurement cycles [CM014]. Furthermore, external category evaluations note that Cognite commands modest mindshare in broader AI data analysis rankings relative to general enterprise data management platforms [CM015].

2.4 Exhibits

3.1 Competitive Landscape and Structural Alternatives

Industrial DataOps represents an operational middleware layer designed to extract, contextualize, model, and govern disparate telemetry across operational technology (OT), enterprise information technology (IT), and engineering technology (ET). In this category, Cognite confronts four distinct competitive vectors: legacy industrial automation conglomerates, cloud-native point solutions, horizontal enterprise data fabrics, and in-house do-it-yourself (DIY) engineering builds. Established automation leaders like AVEVA, Siemens, and Rockwell Automation maintain deeply entrenched hardware footprints, proprietary supervisory control and data acquisition (SCADA) systems, and on-premises historians across global manufacturing sites. However, these legacy architectures frequently trap operational telemetry in proprietary protocols and vendor-locked silos, impeding cross-facility contextual analytics. Concurrently, cloud-native challengers including Litmus and Sight Machine target rapid shop-floor telemetry ingestion, yet often lack multi-site enterprise knowledge graph modeling capabilities. Meanwhile, generic enterprise hyperscalers and horizontal data platforms lack native industrial protocol extractors and contextual ISA-95 asset hierarchy models. Consequently, industrial buyers must evaluate whether to assemble fragmented open-source data lakes or deploy purpose-built commercial platforms.[CP001, CP002, CP003, CP007, CP008]

3.2 Architectural Differentiation: Industrial Knowledge Graphs versus Legacy Silos

The core technical differentiator for Cognite Data Fusion (CDF) centers on its Industrial Knowledge Graph and automated data contextualization engine. Unlike conventional process historians such as OSIsoft PI, which primarily function as optimized flat time-series tag archives, Cognite maps unstructured documents, piping and instrumentation diagrams (P&IDs), 3D computer-aided design (CAD) models, and real-time sensor streams into a unified semantic graph. Independent procurement benchmarks evaluate Cognite exceptionally high on OT/IT/ET data integration (4.8/5.0) and industrial asset modeling (4.9/5.0), bolstered by over 90 pre-built connectors. Furthermore, the operationalization of Atlas AI and Cognite Flows introduces agentic AI capabilities that allow frontline operators to construct automated workflows and execute root-cause diagnostics directly against contextualized plant data. Conversely, Cognite demonstrates material limitations in native plant-floor edge execution and sub-second control loops, scoring only 2.8/5.0 in edge processing benchmarks. Because Cognite relies on containerized on-premises extractors that buffer telemetry for upstream processing rather than functioning as a distributed deterministic edge runtime, mission-critical safety systems and closed-loop control must remain governed by local programmable logic controllers (PLCs) and distributed control systems (DCS).[CP004, CP005, CP006, CP009, CP014]

3.3 Commercial Packaging, Deal Architecture, and Total Cost of Ownership

Cognite's go-to-market motion relies on high-value, multi-year enterprise SaaS subscriptions structured through customized master services agreements (MSAs) rather than transparent self-serve or public rate-card pricing. To protect high-margin software economics and satisfy ASC 606 revenue recognition standards, Cognite maintains a dedicated deal design governance function within its revenue operations group. This team actively unbundles low-margin, service-heavy implementation tasks into separate Professional Services, Development Accelerators, and Success Track advisory order forms. In contrast to legacy automation incumbents that penalize sensor instrumentation expansion via restrictive per-tag licensing fees, Cognite's platform subscription encourages broad enterprise connectivity. However, total cost of ownership (TCO) analysis reveals material procurement risks: year-one expenditure is heavily weighted toward billable consulting hours, custom extractor configurations, and contextual schema design. Moreover, because underlying consumption metrics (such as data-volume throughput and 3D ingestion rates) remain confidential and unindexed in public marketplace listings, enterprise buyers face potential budget expansion as multi-site telemetry scales beyond initial pilot projections.[CP010, CP011, CP012, CP013]

3.4 Moat Durability, Switching Costs, and Competitive Risks

Cognite's long-term competitive moat rests on high switching costs created by deep operational integration rather than contractual lock-in. By embedding open REST APIs, GraphQL endpoints, and Python SDKs into enterprise IT environments, Cognite positions CDF as a collaborative data liberation fabric that coexists with existing historians rather than demanding immediate rip-and-replace displacement. Once an industrial operator configures complex cross-asset relationships, custom machine learning pipelines, and frontline Flows workspaces across multiple facilities, extracting that institutional operational context and migrating to an alternative platform entails severe engineering disruption. Nevertheless, Cognite faces meaningful strategic vulnerabilities. Enterprise cloud software platforms such as Palantir and Databricks are expanding verticalized industrial manufacturing accelerators, while industrial conglomerates (including Schneider Electric, which maintains a pending acquisition agreement) seek to unify automation hardware with software platforms. Additionally, customer concentration in capital-intensive energy and process sectors exposes software expansion rates to macro commodity and capital expenditure cycles. Maintaining durable differentiation will require Cognite to defend its data modeling superiority while bridging the latency gap between cloud intelligence and plant-floor edge execution.[CP001, CP005, CP006, CP009, CP012, CP014]

3.5 Exhibits

4.1 Revenue Quality and Growth Signals

Cognite reported annual revenue above $170 million for fiscal 2025, while its separate ARR-bookings metric grew 36% year over year [CI001, CI002]. The two measures are not interchangeable: annual revenue is an accounting-period outcome, whereas bookings growth is a forward commercial signal whose opening balance and conversion to recognized revenue were not disclosed. Expansion evidence is strongest within a subset: Cognite reported 150% NRR for newer customers, not for the full installed base [CI003]. Product adoption may support expansion, with Atlas AI included in more than 70% of 2025 bookings and operationalized by 59% of customers measured by ARR [CI004, CI016]. Revenue concentration has moderated but remains meaningful because Cognite said roughly 40% of 2025 revenue came from outside oil and gas [CI005].

4.2 Pricing, Contracting, and GTM Economics

Public evidence points to a sales-assisted enterprise motion rather than a self-serve funnel. Cognite's own site directs prospects to request a live demonstration, while RFP.wiki reports that Cognite Data Fusion is purchased through custom subscription order forms and does not publish standard unit prices [CI008, CI006]. The same comparison says professional services, Success Track, and Development Accelerators are ordered separately from the core subscription [CI007]. That packaging can create multiple revenue streams, but public sources do not disclose their mix, discounts, contract duration, or realized price. Customer infrastructure and service costs can also include on-premises extractors, identity integration, and OT connectivity [CI009]. Consequently, list-to-realized pricing, CAC, sales-cycle length, and payback cannot be underwritten from the available material.

4.3 Unit Economics and Cost-Base Visibility

The public record provides growth proxies but not a complete unit-economic model. Newer-customer NRR of 150% suggests strong expansion in that cohort, yet it omits legacy cohorts, logo churn, contraction, and the denominator needed to generalize across the customer base [CI003]. Headcount grew 21% in 2025 to more than 800 employees, a scale indicator and potential operating-cost signal rather than a direct measure of burn [CI012]. RFP.wiki states that Cognite has no public EBITDA disclosure and that profitability and burn could not be verified in its review [CI010]. Public evidence also lacks consolidated gross margin, software versus services margin, cloud hosting cost, implementation utilization, CAC, and payback. The result is a credible top-line growth narrative without enough information to test operating leverage.

4.4 Capital Adequacy and Financial Verdict

Capital history does not establish current liquidity. Tracxn reports $225 million of total funding, while CNBC's 2026 profile lists $300 million; these incompatible totals require reconciliation against the capitalization table [CI013, CI014]. A documented $150 million 2021 investment valued Cognite at $1.6 billion, but that historical financing cannot be treated as cash still available in 2026 [CI015]. No assigned source discloses current cash, operating cash burn, runway, debt or credit facilities, covenant exposure, or the trigger for another financing. The financial verdict is therefore mixed: disclosed revenue and bookings indicators support commercial momentum, but revenue quality, margin durability, and capital self-sufficiency remain unverified. Underwriting should remain conditional on audited statements, cohort economics, contract data, and a current liquidity bridge.

4.5 Exhibits

5.1 Product surface and industrial workflows

Cognite delivers a platform rather than a single point application. Its documentation divides the surface into Flows and field operations, knowledge-graph and data-integration capabilities, production deployment, and developer tools. AWS describes Cognite Data Fusion as the layer that unifies IT, OT, and engineering data and exposes contextualized information through applications including Canvas, AI Copilot, InField, InRobot, Maintain, and Charts. Cognite's May 2026 launch release positions Flows above that foundation as an action layer where frontline users and developers assemble operational workflows. That release also names B. Braun and Idemitsu Kosan as users, but its adoption and speed claims remain company-issued and disclose neither a customer-count denominator nor an independent benchmark. The product thesis is therefore coherent: consolidate fragmented industrial data, contextualize it in a knowledge graph, and shorten the path from data to operator action. The diligence question is how consistently that thesis translates into production outcomes across the installed base.[CE001, CE007, CE009, CE013, CE014]

5.2 Architecture and deployment model

The documented architecture separates source, enterprise, and solution data models. Source layers preserve system data, the enterprise layer integrates and validates shared context, and solution views serve application-specific needs without routinely duplicating data. Cognite describes a property-graph model built from spaces, containers, views, nodes, edges, and shared identifiers. Direct relations are the default, while edges are reserved for relationship metadata or very high-cardinality many-to-many links; the documentation notes a current limit of 2,000 outgoing direct relations in a list. AWS adds a deployment view: purpose-built extractors and native AWS connectors ingest industrial sources, quality checks and entity matching construct the knowledge graph, and applications or AWS services consume the result. Cognite recommends Git-based configuration-as-code deployment through Toolkit across development, staging, and production. This is a specific, composable architecture, although the assigned evidence does not provide independent throughput, latency, or availability measurements.[CE002, CE003, CE005, CE006, CE007, CE008]

5.3 Maturity, controls, and technical risk

The Q3 2026 update shows uneven maturity across the portfolio rather than a uniformly production-ready stack. It labels some search, 3D, InField, Canvas, and custom-app capabilities as production, while Industrial MCP and the Agent Builder CLI are previews and Atlas AI Sandboxes remain private preview. That cadence is evidence of active delivery, but it also creates versioning, support, and change-management exposure for customers building on newer AI surfaces. The data-model documentation specifies separate schema and instance permissions, facility-level instance spaces, indexing, and controlled deployment. AWS describes a reference design using IAM, KMS, Secrets Manager, Shield, CloudWatch, and CloudTrail, but those controls characterize an AWS solution pattern rather than a Cognite-wide certification or audited control conclusion. A company-authored engineering job repost further indicates investment in multi-tenant, multi-cloud orchestration, observability, SLOs, and incident response; it is a hiring signal, not proof that stated reliability targets are met. Underwriting should therefore require service-level history, incident records, penetration-test summaries, certification scope, and workload-level cost and performance data.[CE004, CE010, CE011, CE012, CE015, CE016]

6.1 Customer Base Segmentation and Target Personas

Cognite targets enterprise accounts within asset-intensive industrial verticals, primarily oil and gas, energy, power and utilities, chemicals, pharmaceuticals, and discrete manufacturing. Within these organizations, commercial purchasing authority typically sits with executive digital transformation leaders such as Chief Digital Officers, Vice Presidents of Operational Technology, and Chief Information Officers seeking unified data contextualization. Frontline users include reliability engineers, plant operations managers, maintenance technicians, supply chain schedulers, and industrial data scientists. The commercial model addresses enterprise-scale facilities where cross-data-silo latency between enterprise resource planning systems and operational historian systems creates multi-million dollar downtime or yield penalties. Cognite's GTM motion relies on field sales reps paired with technical solution architects, enterprise integration partners such as Deloitte, Radix, L&T Technology Services, and RoviSys, and supply chain network partners like FourKites to land within major industrial complexes before driving departmental expansion.[CU001, CU002, CU003, CU004, CU005]

6.2 Adoption Trajectory, Consumption Scaling, and AI Workflows

Cognite's customer adoption trajectory exhibits heavy expansion from foundational time-series contextualization toward active agentic AI and workflow execution. In 2025, Cognite achieved 36% year-over-year ARR bookings growth, while monthly active users across customer deployments expanded by 26%. Operational consumption deepened dramatically: the contextualized knowledge graph spanning Cognite's customer base exceeded 80 trillion time-series data points, reflecting a 48% annual increase, while customer data modeling instances doubled and industrial AI token consumption expanded by nearly 500%. Adoption of Atlas AI expanded nearly 700% year-over-year, reaching 59% of overall customer accounts weighted by ARR. Furthermore, by mid-2026, more than 30% of Cognite's customer base and strategic integration partners had been enabled on Cognite Flows to execute root cause analysis and non-conformance workflows, demonstrating rapid uptake of new operational experience modules.[CU006, CU007, CU008, CU009, CU010, CU011]

6.3 Named Customer Proof, Deployment Scale, and Value Delivery

Publicly referenced enterprise accounts validate that Cognite Data Fusion and Cognite Flows operate in live mission-critical production environments rather than stalled proof-of-concept sandboxes. German medical technology multinational B. Braun utilizes Cognite Flows to establish real-time asset transparency and operational intelligence across manufacturing operations. In Japan, refining and petrochemical enterprise Idemitsu Kosan deployed Cognite Flows to systematically resolve asset non-conformance incidents and accelerate engineering root cause analysis. Norwegian exploration and production operator Aker BP has integrated Cognite at the core of its operational model, scaling hundreds of AI agents across offshore assets. Joint deployments with engineering integration partners such as Radix, L&T Technology Services, and RoviSys demonstrate third-party delivery repeatability. However, outside selected marquee references, individual account ARR contributions and contract durations remain confidential.[CU012, CU013, CU014, CU015, CU016]

6.4 Retention Dynamics, Account Durability, and Concentration Gaps

Cognite reports 150% Net Revenue Retention (NRR) for newer customer cohorts, signaling substantial net expansion once industrial accounts integrate operational data models and activate high-value analytics. Enterprise expansion is structured around a Moonshot commitment to deliver $100 billion in cumulative customer value by 2035, supported by formal Value Realisation Workshops, dedicated steering committees, and quota-carrying commercial sales teams. Nevertheless, severe disclosure limitations persist. Cognite reports cohort-specific NRR for newer customers rather than a blended net retention rate across its entire historical customer base, masking potential attrition among legacy or un-expanded accounts. Furthermore, total paying enterprise client counts, logo churn rates, gross revenue retention (GRR), and revenue concentration among top anchor clients—such as founding affiliate Aker BP—are completely omitted from audited disclosures, leaving account vulnerability unquantified.[CU006, CU014, CU015, CU016]

6.5 Exhibits

7.1 Regulatory and Legal Compliance Framework

Cognite operates as a data processor for industrial enterprises across North America, Europe, and Asia, necessitating formal compliance with regional data protection standards including the European Union General Data Protection Regulation (GDPR). The company's standard Data Processing Agreement establishes strict operational boundaries, requiring notification of personal data breaches to data controllers without undue delay and no later than 72 hours following discovery. Furthermore, Cognite contractually warrants that data exports and automated migration tools remain accessible during a 30-day transitional period following any potential insolvency event before permanent erasure. However, because Cognite processes operational telemetry and personal data across multinational cloud and hybrid on-premises environments, cross-border data transfer restrictions and foreign authority subpoena access represent persistent compliance friction points that require active governance.[CR001, CR002, CR003]

7.2 Operational Complexity and Integration Friction

Industrial AI deployment across heavy-asset manufacturing and energy facilities exposes Cognite to severe integration and operational risks. Frontline adoption of Cognite Data Fusion and Cognite Flows requires contextualizing disparate enterprise resource planning systems, warehouse management systems, and operational technology (OT) historians. Third-party benchmarks indicate that while Cognite excels at contextual data modeling and enterprise multi-site integration, it lacks native protocol gateway depth and relies heavily on customer-side or partner-led integration tooling. Commercial evaluations emphasize that Cognite Data Fusion is sold through custom enterprise subscription orders rather than transparent self-serve tiers, with professional services, Success Track, and Development Accelerators billed separately. Incomplete visibility into consumption thresholds and integration labor creates total cost of ownership volatility for industrial buyers.[CR004, CR005, CR006, CR007, CR008]

7.3 Partner Dependence and Customer Concentration

Cognite's commercial trajectory demonstrates significant counterparty and customer concentration risk. Historical company records show anchor deployment relationships with major oil and gas operators, notably Aker BP, which served as Cognite's foundational customer prior to Saudi Aramco Overseas acquiring an equity stake in 2022. To scale delivery across complex manufacturing workflows, Cognite relies heavily on system integrators and solution partners such as Deloitte, Radix, L&T Technology Services, and RoviSys. If strategic systems integrators deprioritize Cognite architectures or key anchor energy accounts decelerate expansion, net revenue retention and customer acquisition velocity would suffer material compression. Monitoring partner enablement bandwidth and large-account renewal schedules remains an essential diligence requirement.[CR009, CR010, CR011, CR012]

7.4 Exhibits

8.1 Investment Thesis and M&A Transaction Context

Cognite's investment thesis has transitioned from an independent SaaS unicorn expansion story to an acquired industrial intelligence powerhouse. On June 30, 2026, Schneider Electric announced an agreement to acquire 100% of Cognite Holding B.V. in an all-cash transaction valued at $3.1 billion. The acquisition price reflects an 18.2x multiple on Cognite's 2025 revenue, which exceeded $170 million. Cognite's core strength lies in its unified industrial data model within Cognite Data Fusion (CDF) and its agentic AI layer Atlas AI, which together bridge complex operational technology (OT) and enterprise IT data across heavy-asset industries. Under Schneider Electric's ownership, Cognite is planned to be integrated with AVEVA and consolidated within Schneider Electric's Industrial Automation business, directly combining industrial data contextualization with AVEVA's CONNECT platform across global industrial sites.[CV001, CV002, CV003, CV004]

8.2 Financing History and Valuation Multiple Discipline

Prior to the $3.1 billion acquisition agreement, Cognite raised approximately $225 million across three primary private financing rounds. In October 2020, Accel led a $75 million Series A round that valued the company at $550 million. In May 2021, growth equity firm TCV led a $150 million Series B investment that propelled Cognite to unicorn status at a post-money valuation of $1.6 billion. A subsequent secondary transaction in February 2022 brought in Saudi Aramco at a $1.5 billion valuation mark. While secondary and pre-IPO marketplaces such as Forge Global tracked limited marketplace trading activity without active pricing quotes, third-party market tracker Latka and aggregator Multiples both evaluate the $3.1 billion acquisition price as an aggressive 18.2x multiple on $170 million of historical ARR/revenue. This premium reflects strategic scarcity value in industrial GenAI and agentic workflow orchestration rather than a standardized software trading multiple.[CV005, CV006, CV007, CV008, CV009]

8.3 Scenario Analysis, Execution Risks, and Diligence Asks

Evaluating Cognite requires examining outcomes under closed transaction integration, extended regulatory delays, or a broken deal scenario. In the base case, the all-cash acquisition closes in the coming quarters following customary regulatory reviews, delivering liquidity at $3.1 billion and positioning Cognite within AVEVA to cross-sell into Schneider Electric's global customer base. The bull case assumes accelerated enterprise penetration where agentic tools like Cognite Flows and Atlas AI drive ARR expansion well beyond 36% bookings growth, unlocking deeper industrial automation synergies. Conversely, the bear scenario examines antitrust or cross-border regulatory scrutiny that could delay closing or force concessions, leaving Cognite exposed to high cash burn and standalone valuation reset toward standalone private software multiples of 8x to 10x ARR.[CV010, CV011, CV012, CV013, CV014]

8.4 Exhibits