Professional services automation (PSA) provider Kantata on June 16 introduced Expertise Agent, an AI system designed to help professional services firms answer cross-functional business questions and orchestrate actions across project management, resource planning, financial systems, and external tools.
Kantata announced the launch as part of new agentic capabilities in the Kantata Expertise Engine. Kantata said the release includes embedded generative business intelligence, self-executing workflows, and a professional-services knowledge graph designed to support more autonomous resource planning and project management.
The launch comes as professional services organizations face tighter capacity constraints, rising client expectations, and growing pressure to scale without simply adding headcount. Kantata’s January 2026 State of the Professional Services Industry research found 87% of professional services organizations plan to use AI agents as part of their workforce, while 89% of leaders said future revenue growth will depend more on how effectively they scale AI than how they scale headcount.
Gap Between Insight and Execution
Kantata is positioning Expertise Agent as more than a resourcing agent or analytics assistant.
The company said the agent can interpret complex questions, create other agents as needed, and coordinate work across PSA workflows and external systems. Example use cases include identifying at-risk projects before margins erode, matching resources to work based on skills and capacity, generating project plans from statements of work and briefings, and preserving institutional knowledge during handoffs.
“Services organizations don’t need more AI features layered on top of disconnected systems,” said Michael Speranza, CEO of Kantata. “They need a system that understands how their business works and can act on that understanding.”
The Expertise Engine’s services-native knowledge graph is central to that pitch. Kantata said the graph connects data from projects, people, systems, documents, communications, and meetings, then maps relationships between skills, delivery patterns, outcomes, and workflows.
That gives the agentic layer operational context. Instead of answering questions from static reports, Kantata wants the system to identify risks, recommend mitigations, build workflows, and apply lessons from successful delivery patterns across future engagements.
Analysis
What this means: Operational context separates useful agents from generic automation. Professional services workflows depend on the connection between projects, skills, utilization, margins, client commitments, and delivery risk. Vendors that embed agents into industry-specific operating data will have an advantage over standalone AI tools that cannot understand how work, capacity, and profitability interact.
Professional Services Firms Face AI Workforce Shift
The launch also reflects a broader change in professional services operations.
Kantata’s survey found more than 66% of professional services organizations have turned down work because of resourcing constraints, while 68% cited skill availability as a barrier, up from 45% the previous year. At the same time, 90% of leaders said their systems will soon need to attribute work, costs, and value across both humans and AI agents.
That finding moves AI agents from a productivity experiment into a workforce, costing, and governance issue. If agents contribute to project delivery, firms need systems that can measure their contribution, assign accountability, track value, and connect AI-driven work to financial outcomes.
The data foundation remains a challenge. Kantata’s research found only 12% of professional services leaders fully trust the data in their systems, down from 24% the previous year. While 88% said they trust AI outputs enough to use them in operational decisions, 89% said they spend significant time verifying those outputs.
That gap could become the constraint on agentic PSA adoption. Professional services firms may want autonomous workflows, but weak data quality, siloed systems, and limited transparency will make leaders cautious about allowing agents to act across project, resource, and financial processes.
Analysis
What this means: Agentic AI forces ERP leaders to manage mixed workforces. Kantata’s launch shows how professional services firms are beginning to treat agents as part of delivery capacity, not just productivity tools. ERP and PSA leaders will need models for assigning work, tracking contribution, costing effort, and measuring value across both people and AI systems.
Sponsor Industry‑Grade Research
Agentic PSA Tests New Operating Model
Kantata’s release points to a larger shift in how professional services firms may run delivery operations.
Traditional PSA systems help firms plan projects, assign resources, forecast utilization, manage budgets, and track delivery performance. Agentic PSA adds another layer of systems that not only show what is happening, but also recommend, coordinate, and execute responses based on operational context.
That changes the role of PSA inside the enterprise stack. It becomes less of a system of record and more of an execution layer for expertise, capacity, and profitability. For firms under margin pressure, the promise is faster risk detection, better resource matching, and more consistent delivery practices across teams.
The practical test will be governance. Firms will need to decide which actions agents can take autonomously, which require human approval, how AI work is costed, and how decisions are audited when agents operate across PSA, ERP, CRM, business intelligence, and customer-defined AI environments.
Analysis
What this means: Data trust will decide how far agentic ERP can scale. Kantata’s survey shows strong interest in AI agents, but also exposes low confidence in underlying business data and heavy verification work around AI outputs. Autonomous workflows require clean data, transparent logic, governed permissions, and auditable actions before agents can move from recommendations to execution.
