The AI landscape doesn't move in one direction — it lurches. Some techniques leap from experiment to table stakes in a single quarter; others stall against regulatory walls, technical ceilings, or organisational inertia that no amount of hype can dislodge. Knowing which is which is the hard part. The State of Play cuts through the noise with a rigorously maintained index of AI techniques across every major business domain — classified by maturity, evidenced by real-world adoption, and updated daily so you always know where you stand relative to the field. Stop guessing. Start knowing.
A daily newsletter distilling the past two weeks of movement in a domain or two — delivered to your inbox while the index updates in the background.
Each dot marks the weighted maturity of practices within a domain — hover for a brief summary, click for more detail
AI that analyses system logs to discover actual business processes, identify bottlenecks, and recommend optimisations. Includes conformance checking and process variant analysis; distinct from workflow orchestration which automates processes rather than discovering how they actually work. Scope covers AI/ML-enhanced process discovery and optimisation; basic log analysis, manual process mapping, and traditional process-mining tools without ML are out of scope.
Process mining has matured into a proven discovery discipline with established vendor ecosystem and quantified ROI across sectors — yet a structural execution gap now bounds adoption momentum. The practice extracts actual process flows from system event logs, surfaces bottlenecks, and recommends optimisations, bridging the persistent gap between how work is documented and how it actually executes. Platforms from Celonis, UiPath, SAP Signavio, and Microsoft Power Automate (OCPM GA May 2026) handle billion-record datasets with real-time monitoring and conformance checking. Recent deployments document concrete results: Molex achieved 90% PO confirmation and 87% touchless invoices on supply chain; Celonis reports $7.5B in tangible business value identified across customers; Rabobank reduced Six Sigma cycles from 9-12 to 4-6 weeks; June 2026 evidence shows AstraZeneca and Arm validating process intelligence as foundational for AI success. However, three structural barriers now dominate: (1) discovery without execution — Celonis identifies problems but lacks built-in execution capabilities, requiring separate tools and organisational change integration; (2) governance-constrained scaling — 81% of enterprises report data governance as blocker to AI initiatives; 80% of implementation effort consumed by data engineering; (3) infrastructure cost forecasting failures — PoC-to-production infrastructure costs 3-5x prototypes, with 30-42% project abandonment post-PoC despite high pilot success. The defining tension for good-practice tier is that technical capability and vendor ecosystem maturity have proven ROI, but organisational infrastructure (governance, data engineering, cost discipline, change capability) and the execution gap create persistent deployment barriers independent of platform feature advancement.
Vendor ecosystem consolidates with AI-process mining convergence positioning discovery as foundational layer for enterprise automation. June 2026 platform evolution: Celonis launches Context Model providing real-time digital twin for Enterprise AI with Ikigai Labs (MIT-linked decision intelligence) acquisition, enabling simulation and forecasting alongside discovery; Microsoft Power Automate OCPM (May 2026) auto-enabled for all 500M+ users; SAP Signavio Journey-to-Process Analytics integrates customer-experience signals with operational data; deepset partners with Celonis for sovereign AI (military, cybersecurity, national infrastructure sectors). Market validation accelerates: Gartner category rebranding from "Process Mining" to "Process Intelligence" reflects ecosystem maturity; analyst recognition (PEAK Matrix 2025, Gartner MQ 2026) consolidates Celonis, UiPath, SAP Signavio, Pegasystems as leaders. Geographic expansion: India represents 260 Global Capability Centres (Mercedes-Benz, Hitachi Energy, AstraZeneca deploying; targeting 1,800+ centres), signaling emerging-market adoption.
Deployment evidence validates production ROI across sectors. June 2026 cases: AstraZeneca/Alexion (pharma) using Celonis Process Intelligence for clinical drug development, reducing timelines by 12 months; Arm using process intelligence as AI orchestration foundation; building materials supplier achieved 49% P2P cost reduction with 6,658 variants discovered; manufacturing (SalientProcess across 600+ deployments) documents 20% downtime reduction and 50% faster supply exception response. Cumulative evidence: 30,000 fewer late payments monthly, 60% throughput increases, €1M IT savings, 400+ compliance rules automated, 90% HR cycle reductions. Celonis reports $7.5B in tangible value with 120 Value Champions, validating enterprise-scale adoption.
Yet structural barriers now dominate adoption momentum. Emerging tension: discovery capability has matured but execution integration lags. Competitive analysis reveals Celonis (market leader) lacks built-in execution—organizations must use separate orchestration, automation, and approval tools, creating integration and change-management complexity. June 2026 research confirms governance as binding constraint: 81% of enterprises report data governance blocking AI initiatives; 93% encounter permission/access issues in pre-production; only 15% have foundational AI governance capabilities. Infrastructure barriers intensify: PoC-to-production infrastructure costs 3-5x prototypes; 30% of gen-AI projects abandoned post-PoC despite working pilots; 42% of enterprises abandoned most initiatives in 2025 (tripling from 17% in 2024). Data engineering consumes 80%+ of PM implementation effort; OCPM adoption faces usability and object-centric data modeling hurdles limiting mainstream scaling. Practitioner assessments document the non-digital blindness: process mining captures application-centric flows only, missing email, phone, Excel, informal channels—requiring precedent digital process mapping (40% of project time) as prerequisite.
— Independent tech journalism covering named customer implementations (AstraZeneca Alexion, Arm) alongside deepset-Celonis sovereign AI platform launch, positioning process intelligence as foundational context for AI success and enterprise orchestration.
— IBM Gold Partner SalientProcess documented outcomes across 600+ manufacturing deployments: 20% unplanned downtime reduction, 50% faster supply exception response, AI-driven quality defect detection, predictive maintenance, NPI documentation automation.
— Independent research synthesis on agentic AI adoption identifying governance maturity (not intelligence) as deployment ceiling, with 80% of implementation effort on data/governance/integration, establishing process discovery and architecture as foundational infrastructure limits.
— Rothbaum Consulting case at Krones (multinational beverage): process mining identified automation opportunities in operational purchasing that were not feasible with standard SAP, demonstrating discovery enabling optimization beyond ERP boundaries.
— Competitive analysis identifying critical execution gap: Celonis (discovery and intelligence tool) identifies problems but lacks execution capabilities, revealing structural limitation in process mining as standalone practice—organizational barrier to operational impact.
— Analysis documenting PoC-to-production gap affecting AI initiatives including process mining: infrastructure costs 3-5x prototype, 30% of gen-AI projects abandoned post-PoC, 42% of enterprises abandoned most initiatives in 2025, highlighting organizational barriers beyond tool maturity.
— ECIS 2026 peer-reviewed research from TU Munich and SAP Signavio identifying seven opportunities and nine challenges in OCPM adoption, documenting that industry is beginning advanced process mining adoption with unclear barriers and enablers.
— Peer-reviewed BPM 2026 conference paper advancing process mining methodology by introducing conformance-aware loss functions for deep learning, enabling neural networks to internalize process model structure alongside prediction accuracy.
2018: Process mining emerges from academic research into early enterprise adoption. Real-world deployments in financial services (PostFinance, VTB Bank) and manufacturing (named Fortune 500 customers) validate business value. Celonis leads vendor market with 5-70x performance improvements and first cloud-native platform. Academic research confirms real-world applicability but reveals methodological gaps in precision measurement.
2019: Process mining moves into broader enterprise adoption. Banking sector deploys process mining for internal control auditing and evidence generation; Fortune 500 adoption confirmed (Coca-Cola Europacific Partners on SAP Signavio). Forrester validates economics with 123% ROI documented. Academic research remains sparse (only 32 ML+process mining articles published 2014-2018) but advances methodology (uncertainty quantification, ERP integration). Practitioner skepticism emerges: lean consultants and vendors question ROI relative to direct observation, highlighting adoption barriers around organizational change and bottleneck risk.
2020: Process mining extends into ERP transformation (SAP S/4HANA migrations) and SME operations. Named deployments document concrete ROI: Uber identifies $20M in improvements; Siemens saves tens of millions on supplier payment and order-to-pay optimization over six years. However, awareness-adoption gap persists: 70% of decision-makers see process insight as essential for RPA, but only 31% deploy tools. Peer-reviewed research identifies four core adoption challenges (planning, selection, implementation, cultural embedding); practitioner assessments highlight data preparation and change management friction. IEEE Task Force launches vendor-agnostic global survey to benchmark adoption patterns. Academic research advances predictive monitoring (€2.5M ERC grant on Apromore platform), signaling evolution from tactical discovery to operational intelligence. Market forecast projects $1.4B process analytics market by 2023.
2021: Process mining consolidates into mainstream enterprise adoption with documented ROI. Accenture achieves 75% cycle-time reduction across 50-country procurement transformation on Celonis and SAP platforms. Deloitte and IEEE Task Force conduct industry-wide adoption surveys, revealing that organizational challenges (planning, change management, data prep) dominate technical barriers. Peer-reviewed Delphi study with 40 experts identifies 30 opportunities and 32 adoption challenges. Celonis ecosystem deepens with partner mechanisms for deployment and abstraction (academic case study). Delivery Hero and other enterprises adopt process mining for operational optimization. Market trajectory remains positive despite practitioner skepticism on ROI vs. simpler methods.
2022-H1: Process mining adoption expands with strategic scaling methodologies emerging. Forrester survey shows 61% of 818 decision-makers rank process mining as top improvement technology with 12-month deployment intent, but execution gaps widen: only 16% have complete process visibility, and 72% still rely on manual methods. Celonis-Fraunhofer joint study documents successful multi-process deployments while identifying cross-functional scaling challenges. Academic research focuses on adoption phase guidance (ECIS 2022 method on use case selection). However, critical countervailing signals emerge: healthcare shows no systematic uptake beyond research pilots; peer-reviewed research questions gap between insights and implementation; Springer book chapter documents ethical and fairness risks (FACT criteria). Adoption barriers remain organizational rather than technical—ROI justification, stakeholder buy-in, data preparation, and organizational resistance dominate reported failures.
2022-H2: Process mining reaches inflection point with documented scale and critical barriers. Production deployments show compelling ROI: Max Mara achieves 90% resolution time reduction, Tostem Thailand identifies 44 improvement themes, Deloitte-supported engagements quantify manual effort at scale (54% spreadsheet time in O2C processes). Market growth accelerates: $527M in 2021 projects to $26B by 2031 at 47.9% CAGR. However, practitioner research reveals structural adoption barriers: 27-expert study identifies gaps in stakeholder involvement and continuous deployment; 79% of companies cite skill gaps; critical assessments highlight data quality dependencies and risk of process-mining centricity without organizational discipline. Vendor platforms mature (UiPath GA, Celonis CoE frameworks), but success depends on organizational rather than technical factors.
2023-H1: Process mining consolidates ecosystem maturity with 50+ documented case studies across global industries, confirmed via IEEE Task Force aggregation. Vendor platforms advance: UiPath 2023.4.1 adds Kubernetes support and app templates, signaling cloud-native investment; analyst recognition (Everest Group) confirms market leadership with named customer deployments. Real-world case evidence emerges from City of Vienna auditors using PM for complete purchase-to-pay audit coverage (2,550 cases, eliminating sampling need). However, peer-reviewed research identifies fundamental ML-process mining integration challenges: ML models rely on ad-hoc assumptions misaligned with process data distributions, and learning procedures ignore concurrency constraints. Academic frameworks mature (use-case taxonomies, maturity models) but adoption barriers remain organizational—skill gaps, data preparation complexity, and stakeholder involvement gaps dominate reported implementation difficulties.
2023-H2: Process mining reaches inflection point in enterprise deployment scope and scale. Market analysis projects growth from $1.8B (2023) to $12.1B by 2028 at 45.6% CAGR. Vendor platforms mature: UiPath 2023.10 adds inline transformations editor and KPI metrics tools; community projects (multiprocessmining.org) demonstrate technical advancement with 5x-10x performance improvements in object-centric variants, enabling analysis of 100M+ event datasets. Adoption metrics show sector-wide penetration: HFS Research survey (260 enterprise leaders) reports 56% adoption in customer service, 53% in IT, and 55% in supply chain (in production or scaled up). Real-world deployments expand: automotive manufacturing company deploys process mining on SAP S/4HANA for supply chain optimization, automating exception handling and reducing manual effort; market shows strategic shift from cost reduction (accounts payable) to revenue-impacting initiatives (ERP migrations, customer-facing processes). However, research on object-centric process mining—the next-generation approach—notes that 'adoption in real-world analyses remains limited,' signaling continued technical barriers to broader variant adoption. Organizational adoption barriers persist: skill gaps, data quality dependencies, and implementation complexity remain primary obstacles to scaling beyond early adopters.
2024-Q1: Process mining consolidates mainstream adoption with peer-reviewed validation of organizational deployment mechanisms. Academic research (ICIS 2024, 30 expert interviews) confirms organizations actively leverage process mining for process change, documenting adoption mechanisms and value translation. Object-centric process mining reaches named enterprise deployments: Freudenberg case study demonstrates 10% working capital improvement, validating OCPM viability in complex manufacturing supply chains. Market fundamentals strengthen: process mining software market forecast at $3.4B in 2024, growing at 20.3% CAGR to $10.3B by 2030, with manufacturing leading at 30% share. However, critical adoption barriers persist: practitioner evidence documents 80%+ of implementation effort consumed by data engineering and event log construction; auditing sector shows limited North American uptake despite potential, with auditor resistance and data quality challenges blocking systematic deployment. Organizational barriers (skill gaps, data dependencies) remain more significant than technical capability.
2024-Q2: Process mining continues technical advancement and market consolidation. ECIS 2024 research demonstrates object-centric approaches deliver measurable improvements (7% prediction lift) on production Order-to-Cash processes, validating next-generation methodology efficacy. Market leadership consolidates: Gartner and Everest Group recognize Celonis and UiPath as category leaders, confirming vendor platform maturity. However, organizational adoption barriers intensify: Utrecht University study of 17 organizations identifies seven distinct challenges in translating insights to improvements, with data preparation and change management complexity dominating barriers. Practitioner assessments highlight technical implementation friction points: process conformance checking faces format compatibility, naming mismatches, and manual process capture barriers requiring custom workarounds. Despite continued analyst validation, organizational rather than technical capability remains the primary constraint to broader adoption.
2024-Q3: Process mining reaches wider sectoral adoption with quantified deployment success in retail and manufacturing. Zespri (NZ$4.2B revenue, kiwifruit marketer) achieves 27% Vendor Invoice Management cycle time reduction and purchase order conformance improvement from 65% to 88% using Celonis over 18 months; Globus luxury retailer reduces order cancellations by 20% and reaches 99.9% on-time delivery. Vendor platforms mature: UiPath releases July 2024 GA features (dashboard filters, conformance checking enhancements, UK data sovereignty). Academic research (PM² methodology) continues documenting deployment value and integration obstacles. However, research-backed practitioner assessments highlight persistent failure rates: over 2 in 3 solutions achieve less than 10% ROI; data preparation consumes 80%+ of implementation effort; project terminations driven by momentum loss, internal politics, and stakeholder disengagement. Critical adoption barriers (change management, organizational alignment, data engineering complexity) dominate implementation challenges more than technical product capabilities.
2024-Q4: Process mining continues demonstrating measurable enterprise ROI with diversified deployment across procurement, financial services, and utilities. Accenture globally deploys Celonis across 50 countries on SAP platforms, reducing requisition approval cycle from 60 to 15 hours (75% reduction) and discovering 14,000 process variants; Florida Crystals (sugarcane grower) identifies millions in missed discounts and duplicate payments within first month; Eversource Energy utility uses process mining for operations optimization via EY partnership. Vendor platforms mature further: UiPath 2024.10 GA adds enhanced conformance checking and data sovereignty features; Apromore gains multi-analyst recognition as Leader. However, independent research from Utrecht University identifies five "terminators" blocking adoption sustained success (data preparation burden, interest loss, expertise deficits, incentive misalignment, organizational denial), confirming organizational barriers remain the primary constraint despite quantified ROI evidence.
2025-Q1: Process mining continues ecosystem expansion with vendor platform innovation and reaffirmed market growth projections. SAP Signavio releases Journey-to-Process Analytics GA capability (March 2025), integrating experience and operational data for new deployment patterns. Market projections accelerate: process mining market valued at $3.1B in 2024, projected to reach $23.3B by 2030 at 40.3% CAGR, up from prior forecasts. Sector-specific deployments demonstrate ROI: telecom and HR firms deploy process mining for workflow optimization, achieving 40% customer satisfaction increase and 35% time-to-hire reduction respectively. However, critical practitioner assessments reinforce persistent adoption challenges: common pitfalls (data quality, misalignment, collaboration gaps, change management neglect) continue blocking success; independent expert Delphi study identifies 30 opportunities and 32 adoption challenges, confirming organizational barriers remain the primary constraint to broader scaling despite improving vendor platforms and documented deployment ROI.
2025-Q2: Process mining consolidates AI integration pathway with organizational scaling research advancing the field. Academic research documents organizational challenges in scaling process mining beyond pilots, proposing Value Management Capability Framework for sustained value realization. Vendor platforms evolve: UiPath February 2025 release adds data transformation connectors enabling end-to-end process visibility across integration services and automation execution. Industry adoption metrics show AI-process mining convergence accelerating: Deloitte survey (120+ respondents) documents 25% of organizations already integrating AI with process mining, with 74% planning AI integration for automated decision support and behavior prediction. Real-world deployments expand to construction sector: Portuguese construction company deployed Celonis for Accounts Payable optimization and process discovery. However, organizational adoption barriers persist: practitioner research identifies common pitfalls (data availability challenges, silo thinking, lack of defined objectives, data privacy considerations) blocking project success, and highlights necessity of Business Process Management discipline alongside process mining for value realization—confirming organizational capability remains the primary constraint despite maturing vendor platforms and expanding AI integration.
2025-Q3: Process mining reaches inflection point in deployment ROI documentation and methodological maturity. Forrester Consulting study of Celonis deployments documents 383% ROI and six-month payback, with quantified benefits across multiple operational domains ($44.1M composite benefit, 33% to 86% automated delivery, $24.5M inventory savings, labor cost reductions through full automation). Consulting and practitioner literature advance process mining methodology: Rothbaum Consulting documents ERP transformation risk mitigation through three-phase process mining deployment; Roland Woldt releases new book with six-step framework for successful projects, addressing common execution pitfalls. However, critical practitioner assessments reveal persistent scope limitations: process mining provides only partial process view (application-centric, unaware of manual activities and cross-system policies), requiring supplementation with BPMN translation and enterprise architecture frameworks for full organizational value realization. Evidence balance shows strong positive ROI signals tempered by documented limitations requiring organizational discipline for effective deployment—confirming practice remains at good-practice tier with execution barriers (methodology, scope limitations, organizational alignment) dominating technical capability as constraints to broader adoption.
2025-Q4: Process mining consolidates vendor platform innovation with continued market growth acceleration and balanced evidence on technical barriers. Vendor releases: UiPath December 2025 GA introduces Process Insights dashboard for automated anomaly detection; SAP Signavio's Journey-to-Process Analytics integrates customer experience and operational data. Market analyst validation (Everest Group State of Market 2025) confirms mature provider landscape (Celonis, UiPath, SAP Signavio, ServiceNow) with sustained growth forecasts accelerating to $23.3B by 2030 at 40.3% CAGR. Case study aggregation (51 documented deployments) validates ROI: 43% bottleneck reduction, 4% elimination of unnecessary steps, 35% automation increase, 52% rework time reduction. AI-process mining integration accelerates: Deloitte survey shows 25% of organizations already combining AI with PM, 74% planning integration for decision support. However, critical evidence emerges on technical barriers: expert opinion documents SAP environment limitations (incomplete event logs in VBFA, CDHDR tables), constraining full operational visibility; consulting assessments document traditional PM projects consuming 12-18 months and £200,000-£300,000 before ROI, with high failure rates; case study review reveals pre-mapped solutions (Process Pilot approach) deliver results in days vs. months, signaling complexity barriers in standard implementations. Organizational barriers remain primary constraint: data preparation (80%+ effort), change management complexity, stakeholder alignment, expertise gaps dominate reported implementation challenges more than technical platform capability. Evidence balance (positive ROI deployments + critical technical/organizational barriers) confirms good-practice tier stability with execution rather than capability as primary advancement constraint.
2026-Jan: Process mining continues deployment momentum with ecosystem integration deepening and new use cases emerging. Vendor platform integration advances: UiPath Maestro integration enables automatic Process Optimization app generation with conformance checking and anomaly dashboards. Real-world deployments expand: Eissmann Automotive (German automotive manufacturer) deploys Celonis enterprise-wide, achieving 30% throughput time reduction on Purchase-to-Pay and production; Celonis applies process mining to supply chain emissions tracking (Scope 3) for sustainability applications. However, implementation barriers remain dominant constraint: vendor assessment documents persistent data quality, access, and skills gaps; expert reassessment (van der Aalst) warns of organizational understanding gaps and AI risks (bloating, semantic loss); required skills shift toward object-centric data, process-aware AI, and change management, indicating technical capability outpaces organizational absorptive capacity. Good-practice tier stability confirmed: deployment ROI validated but execution barriers (data engineering, change management, organizational alignment) constrain advancement pace.
2026-Feb: Process mining platforms advance with continuous feature delivery and domain expansion. Vendor platform maturity deepens: UiPath February 2026 GA adds data residency in Switzerland and root cause analysis features, expanding compliance and analytical capability; Celonis launches Robotic Systems Intelligence Manager (with LeafLabs) extending process intelligence to robotics and autonomous supply chain systems, achieving 50% acceleration in development cycles at Pickle Robot Company. Market validation sustains: Celonis maintains 120+ Value Champions generating $8.1B in measurable value, validating enterprise scale adoption. However, emerging evidence surfaces tool maturity limitations: peer-reviewed research reveals substantial suitability gaps for process mining tools in IoT and sensor-driven contexts, highlighting that business-process-oriented platform architecture constrains applications in industrial and emerging domains. Organizational barriers remain persistent: critical assessments document 80% data preparation effort, integration complexity, and misapplication risks (PM overuse for automation where simpler methods suffice). Good-practice tier stability reaffirmed: technical capability continues advancing but adoption remains constrained by implementation complexity, data engineering burden, and organizational absorptive capacity rather than platform feature maturity.
2026-Mar: Process mining consolidates production deployment evidence across functions and sectors with vendor platform advancement. Platforms mature: Microsoft Power Automate adds object-centric process mining GA (May 2026) auto-enabled for all customers; SAP Signavio and Celonis continue ecosystem expansion. Case study aggregation from independent sources documents production outcomes: five unrelated organizations achieved 30k fewer late payments/month, 60% throughput increase, €1M IT savings, 400+ compliance rules automated, 90% HR cycle reduction; five named enterprises (Deutsche Telekom, Fujitsu, Mercedes-Benz, Uniper, Vinmar) report quantified production ROI across manufacturing, energy, distribution. Celonis Value Champions metric sustains: 120 organizations at $10M+ documented value each, totaling $8.1B. However, peer-reviewed research (Eindhoven University, van der Aalst) documents 27 distinct event log quality issues undermining 100+ organization implementations; independent advisory (Adapt Digital) confirms true process mining captures only application-centric flows, missing manual work and informal processes. Emerging AI augmentation (Fraunhofer research on LLMs in process mining) signals next-generation methodology maturation, but organizational barriers remain dominant: 80% data preparation effort, shifting skill requirements toward object-centric data and process-aware analytics, and artifact completeness constraints. Good-practice tier stability maintained: deployment economics validated but execution barriers (organizational absorptive capacity, data engineering complexity, scope limitations) prevent advancement despite strong platform maturity signals.
2026-Apr: Process mining extends sectoral adoption and AI integration momentum with reinforced market growth signals and enterprise deployment breadth. Market analysis confirms acceleration: process mining software valued at $1.68B (2025) with 40.2% CAGR through 2035; German Mittelstand market at €0.85B growing 18% annually with SME deployments achieving 33% Purchase-to-Pay cycle reductions. Enterprise adoption drivers quantified via large-scale survey (1,649 leaders, 5 regions, 13 industries): 85% aim for autonomous operations within 3 years; 76% identify process bottlenecks as primary blocker to AI ROI; 82% recognize process intelligence as foundational to AI success. Sector-specific deployments expand: Molex achieved 90% PO confirmation and 87% touchless invoices via Celonis; Rabobank reduces Six Sigma DMAIC cycle from 9-12 to 4-6 weeks; multiple Order-to-Cash implementations (Siemens 11% rework reduction, 3M 92% order automation, Sysmex $15M cycle improvement). Vendor ecosystem matures: Microsoft Power Automate object-centric process mining (GA April 2026) auto-enabled for all customers; UiPath documents production use cases covering existing process optimisation, new workflow orchestration, and end-to-end improvement. Practitioner evidence documents that initiative failures occur both before tool deployment and after go-live, signalling implementation rather than technical barriers. Critical barriers persist: 80%+ of implementation effort consumed by data engineering; skill requirements shifting toward object-centric data and process-aware analytics. Good-practice tier stability confirmed with organisational absorptive capacity as binding constraint.
2026-May: Process mining consolidates platform maturity and mainstream availability with continued ecosystem deepening and a category-level rebranding. Gartner names ARIS, Celonis, Pegasystems, and SAP Signavio as 2026 Process Intelligence Leaders, marking an industry shift from "Process Mining" to "Process Intelligence" as the category designation. Celonis acquires MIT-linked Ikigai Labs, coupling process intelligence with decision intelligence (planning, simulation, forecasting) and signaling platform evolution toward AI-augmented optimization. Vendor platforms advance toward standardization: ServiceNow Process Mining auto-installed free on all instances; UiPath adds root cause analysis dashboard (February 2026 GA) and conformance checking documentation; Microsoft Power Automate 2026 Wave 1 ships Process Intelligence Studio and Object-Centric Process Mining support, extending mainstream availability. Deloitte Global Process Mining Survey documents sustained ROI: 80% of users report value delivery, with cost-savings expectations rising from 46% to 59% over two years. Market fundamentals strengthen: process mining market valued at $2.3B (2025) growing to $36B by 2034 at 34.58% CAGR. Analyst recognition sustained: Everest Group names Celonis Leader (sixth consecutive year) and Star Performer (fourth consecutive year). Critical adoption barriers persist: data preparation consumes ~80% of PM project time; practitioners document structural constraints (event log dependency, inability to capture unstructured work and cross-team handoffs) as organizational barriers. Good-practice tier stability confirmed: ecosystem maturity and analyst category evolution signals balanced against persistent implementation complexity.
2026-Jun: Process mining reaches inflection point in healthcare sector adoption and strategic AI-enterprise context integration. Healthcare deployment validation: Novo Nordisk (major pharma) operationalizes 100+ AI agents powered by Celonis Process Intelligence to reduce clinical drug development time by 12 months, validating deployment in highly regulated, complex processes. Celonis PI Day coverage (AstraZeneca Alexion, Arm) and deepset-Celonis sovereign AI platform launch position process intelligence as foundational AI-success context across enterprise and national-infrastructure sectors. SalientProcess documents outcomes across 600+ manufacturing deployments (20% unplanned downtime reduction, 50% faster supply exception response), adding cumulative sector breadth. Discovery-execution gap crystallises: competitive analysis confirms Celonis identifies problems but lacks built-in execution capabilities, requiring separate orchestration tools and creating systemic integration and change-management complexity. Governance emerges as the deployment ceiling: independent research synthesis documents governance maturity (not model intelligence) as primary agentic AI adoption ceiling, with 80% of implementation effort on data/governance/integration and 81% of enterprises reporting governance-related AI initiative delays. Infrastructure cost failures deepen: 30% of gen-AI projects abandoned post-PoC, 42% of enterprises abandoned most initiatives in 2025, with PoC-to-production infrastructure costs running 3-5x prototypes. ECIS 2026 peer-reviewed research documents domain-specific event log gaps, proposing LLM augmentation. Good-practice tier stability maintained: ecosystem maturity confirmed with strategic AI platform integration, yet organizational absorptive capacity and the discovery-without-execution structural gap prevent advancement past good-practice.