Process intelligence is what process mining has grown into. Process mining draws the map of how work really flows. Process intelligence adds why the work slows down, whether the rules were followed and whether a fix worked. Gartner's 2026 Magic Quadrant is titled Process Intelligence Platforms.
Most tools in the category follow one process at a time. Namuda follows every record across all the work it touches. This post explains why that matters.
The problems sit between processes
The process map on the wall comes from a workshop. The real process has loops, waits and shortcuts nobody drew, and most problems sit where one process hands over to another.
A claim belongs to a policy, a customer and an adjuster. A payment belongs to an invoice, a purchase order, a project, a grant and a vendor. Follow only the claim, or only the payment, and you miss the slowest hand-offs. You also miss most of the integrity risk, because that hides in the links.
Three tools, each seeing one part
Slow work, broken rules and integrity risk usually sit on the same records: the claim that was reopened, the offer cancelled after two weeks of silence, the payment released before the grant agreement was signed.
| Tool | What it sees | What it misses |
|---|---|---|
| Process mining tool | The delay | The rule that was broken |
| Compliance tool | The rule | The delay |
| Audit analytics tool | The transaction | How it connects to the rest |
| BI dashboard | The total | The record, and why |
Namuda sees all of it on the same record, and follows the fix until the data says it held.
Four lenses on one data model
Every view in Namuda reads the same event log. The record you see on the map is the same record a rule flags, a driver explains and a goal counts.
| Lens | What it does |
|---|---|
| Process mining | Builds the map from the event log: each object type's path, where they meet, wait, stop and loop back |
| Process intelligence | Drivers, change detection, reports for your industry against peers, data quality, and an AI guide that walks the map |
| Compliance | Rules on all records at every data load, a risk score per unit, ranked and explained signals |
| Improvement | Goals with targets, and improvements with their expected effect, judged by the data at a later load |
What it answers
Buyers rarely ask for a category. They ask one of five questions.
| Outcome | The question |
|---|---|
| Performance | Where does time go? |
| Cost | Where is money lost? |
| Risk | Where is risk building up? |
| Compliance | Are our rules followed, on every record? |
| Proof | Did the fix work? |
One story, the same claims
On a synthetic log of a fictional Norwegian insurer (October 2025 to May 2026), the answers land on the same records.
- Performance. Half of all claims are paid within 10.9 days of first notice. Outside one storm, the biggest driver of the wait is asking the customer for documents: 20.8 days with a document request, against 12.0 without.
- Cost. 1.5 million kroner was paid on lapsed cover, all of it in one office.
- Compliance. 4.5% of settlements above 50,000 kroner were approved by their own assessor.
- Proof. The recommended action becomes an improvement with its expected effect, and the data judges it at a later load.
The delay, the cost and the deviation were the same claims.
How it reads the data
Namuda reads data as objects and the events between them, using OCEL 2.0, the open standard for object-centric process mining. That is what lets one rule catch a payment made before its contract was signed, and lets one score compare every country office on the same terms.
It connects to the systems your work runs in, such as SAP, Dynamics 365, Salesforce, ServiceNow, HubSpot and Infor, or starts from the extracts you already have: CSV, Excel or Parquet.
It no longer takes a specialist team
Turning raw data into an event log used to be a consultancy project, and reading a process map needed a specialist. Now the AI does the groundwork and people decide. It proposes how to build the log, asks what your data means, suggests what to look at and explains what it finds.
Process intelligence is no longer a project for a centre of excellence.
How to start
A fixed-price pilot on one process, on data you already have. Your team can run it, we can run it with you, or a partner can.
Talk to us about the process to start with.