South Korea-based Puzzle Data is a relatively new entrant in the Process Mining software market. The customer base includes companies in manufacturing, transportation, retail, mobile gaming and public institutions like hospitals and Korea’s Supreme Prosecutors’ Office. Puzzle Data’s ProDiscovery solution is made up of around 30 widgets (“puzzles”) that can be individually configured and added to any dashboard. While a Conformance Checking functionality is still in development, ProDiscovery offers sophisticated widgets for process discovery, statistical analyses, social networks and organizational graphs. Dashboards can be shared with users in the same organization. The architecture is designed for Big Data processing and is built on the open-source environments Apache Hadoop, Spark and the Elasticsearch engine.

Tool Name
Puzzle Data Co., Ltd. (Seoul and Pohang, South Korea)
Company Size
11-50 employees
Free Trial
Upon request
, On-Premises
Embedded In
Tested Version
2.0.42 (SaaS) - Tested in 04/2020
Data Management
Import File Types
Database Connections
JDBC drivers, e.g. Oracle
Adobe Analytics
Integrated ETL Functionality
Data Anonymization & Pseudonymization
Data Loading
  • Data Refresh: Incremental data loading, appending new data to an existing set of data
  • Scheduled Jobs = Automatic data loading in defined time intervals
Data Refresh
, Scheduled Jobs
Scheduled Jobs upon request
Character Encodings
UTF-8 compatibility tested with special characters and various languages: Korean, Japanese, Trad. & Simplif. Chinese, Hebrew, Arabic, Russian
UTF-8 (verified
Attribute Types
, Event-level
Specify Business Hours
Working week
, Multiple shifts/day
, Exclude days
, Holiday calendar
Holidays can be defined by the administrator
Define Event Order
Manual definition of event ordering in case of identical timestamps. This criterion does not consider automatic ordering by the tool.
By selected column, by custom sequence of activities
Start/End Timestamp
2 timestamps
Process Discovery
» Process Graph
As-Is Process Visualization
Directly-Follows Graph (horizontal, vertical)
Additional Heuristics Miner available with adjustable thresholds
Export As-Is Process Graph
For data exports (e.g. CSV) see “Export Reports” criterion
Performance Highlighting
Visual highlighting of process bottlenecks
Active time
, Idle time
Process Animation (Replay)
Adjust speed
, Adjust timeframe
, Switch time mode
, Zoom in case
Search and Filter in Graph
  • Search and find activity names (relevant for spaghetti-like graphs)
  • Filter activities/transitions directly from graph
, Filter
Graph Abstraction
Frequency Metrics
Activity frequency, Case frequency, Paths ratio
Paths ratio = # of arc / total # of arc
Time Metrics
The term “duration” is used when both active and waiting/idle times can be displayed
Avg/med duration
Additional Graph Metrics
Cost metrics
, Custom metrics
» Process Analysis & Analytics
Process Benchmarking
Visual comparison
, Metric comparison
Process Benchmarking (Different Logs)
Visual comparison
, Metric comparison
Root Cause Analysis
Variant Breakdown by
“Duration” refers to the case throughput time
Case count, event count, unique activity count, case-level performance (duration, active time, waiting time)
Aggregation for sorting by time: min/max/avg/med/std.dev
Case and Activity List
Activity List
, Case List
, Case List for Variants
View Case Details
Rework Analysis
Edge/Transition Details
From-to activities: List of ingoing and outgoing activities for any selected activity
List of all transitions
, From-to activities
Conformance Checking
Compare As-Is and Target Process
Target Model Creation
Import model
, Auto-create from as-is
, Create new
In-Graph Conformance Visualization
List of Compliance Violations
Four-Eyes Principle
Sequence Filtering
“(Not) Directly followed by” filtering
Conformance Root Cause Analysis
Operational Support
Alert Generation
Predictive Analytics
Recommendations (Prescriptive Analytics)
Advanced Enhancement Capabilities
Organizational Mining
Scenario Simulation
Decision Rule Mining
Views, Monitoring and Reporting
Export Reports
Events (MXML, TXT), Cases (CSV), Variants (XML, PNG, SVG)
Export Charts and Tables
Custom Dashboards
Custom charts
, Custom tables
Custom Metrics/KPIs
KPI Thresholds
Deviation from defined thresholds can be detected through color differentiation
Advanced Charts
Dashboard offers >5 different chart types
World Map
Latitude & longitude coordinates
, Location by attribute (e.g. country codes, city names)
Save Filter Settings
UI Languages
English, Korean
Share & Collaborate
Share selection
; Dashboard sharing; Memo widget for comments
Security & Compliance
Role-Based Access
User Authentication
Audit Logs

Distinctive Focus and Features

  • Process analytics: ProDiscovery’s “puzzle” library includes sophisticated widgets with various chart types that allow the user analyze the process from many dimensions.
  • Big Data processing: Puzzle Data’s solution is designed to process large datasets, being built on the open-source environments Apache Hadoop, Spark and Elasticsearch.

I’m a vendor. Submit a change request.

Disclaimer: The timeliness of provided information is based on the tested version and date as stated under “Tested Version”. No guarantee can be given about the correctness and accuracy of the information contained.

Click to enlarge

Tap to enlarge

Appetite for an additional visual impression of ProDiscovery?

Keen on exploring more

Process Mining tools?

Have a glance at the features and capabilities of other tools.