What is process intelligence?
Process intelligence is the capability to observe, analyse, and act on how processes actually execute across an organisation. It reconciles the process content you already hold into one queryable model, so ownership, risk, resilience, and compliance can be questioned directly, with provenance for every answer.
Most organisations have plenty of process documentation. What they lack is the ability to interrogate it. Diagrams describe intent. Runbooks describe procedure. Neither tells you, on demand, which production-critical process has no owner, or which control no longer maps to the risk it was meant to cover. Process intelligence closes that gap. It treats the operating model as something you query, not something you read.
Process intelligence is not process mining, and it is not business intelligence.
The three are routinely confused. They answer different questions, and only one of them treats the operating model as a connected structure.
Process mining reconstructs the path a case actually took by replaying event logs from your transactional systems. It is powerful for a single, high-volume, system-bound process. It is silent on ownership, on control, and on the relationships between one process and the next. Mining is a technique that lives inside process intelligence, not a synonym for it.
Business intelligence aggregates numbers into dashboards. It tells you a metric moved. It does not tell you which process produced that metric, who is accountable for it, or which policy governs it. Business intelligence reports on outcomes. Process intelligence explains the structure that produces them.
Process intelligence reconciles your whole operating model into one canonical structure you can question directly. Ownership, risk, resilience, and compliance read from a single source. The unit of analysis is not a chart or a single case. It is the relationship between processes, the people who own them, the systems they depend on, and the controls that govern them.
How process intelligence works.
Process intelligence is built in a sequence. Each step makes the next question answerable, which is why value compounds rather than arriving all at once.
Reconcile the sources
Process content rarely lives in one place. It sits in iGrafx repositories, Visio diagrams, SOPs, BPMN and XML exports, and RACI matrices. Process intelligence reconciles these into one vendor-neutral canonical model with namespaced identifiers, so the same process described in two systems resolves to one entity.
Enrich with governance context
A reconciled model is still just a map until it carries meaning. Enrichment attaches the context that makes a process a governance asset: ownership, controls, risk classification, policy linkage, and dependency paths. This is where a diagram becomes evidence.
Interrogate with provenance
With the model built and enriched, you can ask ownership, impact, risk, and compliance questions in plain language. Every answer returns the source object, the relationship path it travelled, and a confidence score. Nothing is a black box. The answer is audit-ready by construction, not by a quarter-end scramble.
What process intelligence makes possible.
The capability is only as useful as the questions it answers. These are the ones organisations cannot answer reliably without it.
Ownership and accountability. Find the production-critical processes with no documented owner before an incident finds them for you. Trace any process to the person accountable for it and the control that governs it.
Risk and control effectiveness. Identify processes that carry risk with no mapped control, and controls that no longer connect to the risk they were written for. Control effectiveness becomes a query, not an annual exercise.
Operational resilience. Surface single points of failure and dependency chains. See what breaks downstream when a system goes offline, and which important business services sit on top of it.
Three-lines-of-defence visibility. The three-lines-of-defence model assumes each line can see the same process reality. Most organisations cannot provide that. Process intelligence gives the first, second, and third lines one canonical view, in real time, with the control context already attached.
AI readiness. Before you delegate a process to an AI agent, you have to be able to see it. Process intelligence is the precondition for any credible automation programme, which is the subject of the next section.
Why process intelligence comes before AI, not after.
An agentic enterprise operating model is one where AI agents execute work alongside people. It only holds together if authority, evidence, and oversight are designed in before anything reaches production.
The common failure mode is to buy AI features, run pilots, and retrofit governance later. The agents then act outside boundaries no one can prove, and the programme stalls at the first audit. The sequence that works runs the other way: platform first, governance embedded, automation last.
Process intelligence is the foundation that sequence stands on. You cannot safely delegate a process to an agent if you cannot see how it runs, who owns it, and where a human must remain in control. Intent-first workflows define the outcome to reach rather than a rigid script. Human gates mark the points where a person must decide before execution continues. An immutable audit trail records every stage transition and gate decision, which is the evidence regulators and boards ask for.
This is the difference between experimenting with AI and building a capability with it. The model you can interrogate today is the model you can govern an agent inside tomorrow.
Model-first or document-first, process intelligence works with both.
Organisations manage process knowledge in one of two paradigms. Process intelligence reads from either, and rewards the more structured one.
Document-first organisations treat narrative procedures, policies, and manuals as the primary artefact. The knowledge is real, but it is locked in prose. Process intelligence ingests that content and structures it, which is often the first time the organisation can see its operating model as a connected whole.
Model-first organisations treat structured process models, such as BPMN diagrams or an iGrafx repository, as the authoritative source. Process intelligence reaches its full value here, because the relationships it needs are already explicit. It reads the repository, reconciles it, and adds the intelligence layer without disturbing the source.
Neither starting point is wrong. The point of a first diagnostic is to establish which one you are, how much of your content is already structured enough to query, and what to model next.
From content you already hold to questions you can defend.
Process intelligence is a capability. IGX360 Insights is how IGX delivers it, built from the process content you already maintain.
You do not start by building a new repository. You start with what you have. A first diagnostic connects IGX360 Insights to a representative slice of your existing content and shows you what it can already answer: gaps, orphans, ownership ambiguity, and unmitigated risk. You keep the read-out either way.
If you have no BPM foundation yet, the IGX360 Platform gives you one. If you already have content but cannot query it, IGX360 Insights is the intelligence layer that makes it answerable. The diagnostic tells you which is the right next step.
Process intelligence: frequently asked questions
What is process intelligence?
Process intelligence is the capability to observe, analyse, and act on how processes actually execute across an organisation. It turns scattered process content into one queryable model, so ownership, risk, resilience, and compliance questions can be answered directly with provenance.
How is process intelligence different from process mining?
Process mining reconstructs the path a case took from event logs, one process at a time. Process intelligence makes the structure of the whole operating model queryable: ownership, controls, dependencies, and policies, not just the sequence of events. Mining is a technique. Process intelligence is a capability.
Is process intelligence the same as business intelligence?
No. Business intelligence aggregates metrics into dashboards. It tells you a number moved. Process intelligence tells you which process produced the number, who owns it, and what governs it.
Do I need a BPM tool to use process intelligence?
No. Process intelligence reads from the process content you already hold, iGrafx first, plus Visio, SOPs, and BPMN/XML. It adds an intelligence and governance layer on top rather than replacing your modelling platform.
Why is process intelligence a prerequisite for AI?
AI agents execute work inside your processes. If you cannot see how a process actually runs, who owns it, and where a human must decide, you cannot safely delegate it to an agent. Process intelligence supplies that visibility and the audit trail that governance requires.
How long does it take to get value from process intelligence?
Weeks, not years. A first diagnostic reads your existing content and produces a decision-ready view of repository maturity, relationship gaps, and the questions you can already answer, before any wider commitment.