Metrics

Metrics

When you apply data to your process model, ProcessMind displays your chosen metric on the connections between activities. Click any connection to open a context menu and select the metric [Σ] you want to display. A connection shows the number of cases, the number of events, the average transition time between the two connected activities, cost, tCO₂e, or a custom expression — the transition time is measured per transition, not per case.

The following metrics are available for charts and KPI tiles. Throughput time metrics always describe the whole case — or the segment between the selected from/to elements — and are averaged per case:

Counts

• Count Cases
  Displays the total number of cases in the dataset or for a specific attribute. Useful for understanding the overall workload or case volume.

• Count Events
  Shows the total number of events recorded in the dataset. Ideal for tracking activity occurrences or event frequency.

• Count Activities
  Counts the total number of activity executions across all cases. Helps analyze workload distribution and activity intensity.

Attribute Metrics

• Sum Per Event
  Calculates the total sum of a numeric attribute per event (for example, total duration, total cost). Useful for cumulative insights.

• Avg Per Event
  Computes the average value of a numeric attribute per event (for example, average activity duration). Highlights typical performance levels.

• Count Distinct
  Counts the number of unique values for a selected attribute (for example, unique users, systems, or resources). Useful for diversity analysis.

• Min (Minimum)
  Shows the smallest value of a numeric attribute (for example, shortest activity time). Benchmarks best-case scenarios.

• Max (Maximum)
  Displays the largest value of a numeric attribute (for example, longest activity time). Highlights bottlenecks or extreme cases.

• Median
  Returns the middle value of a numeric attribute to reduce the impact of outliers.

• Modus
  Identifies the most frequently occurring value in a dataset. Useful for spotting dominant behaviors or attributes.

Throughput Time (Case Duration)

Measures the total elapsed time from case start to case end, per case. Cases without both boundaries (for example, still open cases) are left out.

• Sum Throughput Time (shown as Total cycle time in the metric list): Total cycle time of all cases, so it grows with the number of cases, not with their events.
• Average Throughput Time: Mean case duration.
• Min Throughput Time: Fastest case completion time.
• Max Throughput Time: Longest case completion time.
• Median Throughput Time: Middle value of case durations.
• 90th Percentile Throughput Time: Duration under which 90% of cases finish, useful for SLA benchmarking.

See Time Metrics for what is counted per case versus per step, and the wording used for each measure.

Processing Time (Active Work Duration)

Measures the time spent actively processing cases or activities, excluding idle times.

• Sum Processing Time: Total processing time across all cases.
• Average Processing Time: Mean active work duration.
• Min Processing Time: Shortest processing time recorded.
• Max Processing Time: Longest processing time recorded.
• Median Processing Time: Middle processing time value.
• 90th Percentile Processing Time: Active time within which 90% of cases or activities are completed.

Waiting Time (Idle Duration)

Measures the time cases or activities spend waiting, excluding active work.

• Sum Waiting Time: Total waiting time across all cases.
• Average Waiting Time: Mean waiting duration.
• Min Waiting Time: Shortest waiting time recorded.
• Max Waiting Time: Longest waiting time recorded.
• Median Waiting Time: Middle waiting time value.
• 90th Percentile Waiting Time: Waiting time within which 90% of cases or activities are completed.

Cost

Measures the cost associated with cases or activities, based on a mapped cost attribute.

• Sum Cost: Total cost across all cases.
• Average Cost: Mean cost per case or activity.
• Min Cost: Lowest cost recorded.
• Max Cost: Highest cost recorded.
• Median Cost: Middle cost value.
• 90th Percentile Cost: Cost below which 90% of cases or activities fall.

Ratios

• Ratio Cases
  Proportion of cases meeting a specific condition relative to the total. Useful for case-level comparisons.

• Ratio Events
  Proportion of events meeting a condition relative to the total. Useful for analyzing event distributions.

By selecting the right metric, you can tailor your process graph to highlight the insights most relevant to your analysis.

Process KPIs

A process KPI is a metric you commit to improving, with a named owner and a target. The metrics above are the raw measures; choosing which of them become KPIs is a decision about what this process is for.

A workable first set for a single process, and the reason each one earns its place:

KPI	Built from	What it tells the process owner
Cycle time (P50 and P90)	Throughput time	Typical experience, and worst-case experience
Waiting vs active ratio	Waiting time / processing time	Whether the process is slow or the queues are
Rework rate	Cases that loop back	Where quality fails, and at what cost
First-time-right share	Cases with no rework	The same story, easier to communicate
Conformance rate	Deviation from the intended model	Whether the approved standard is followed
Throughput	Cases completed per period	Capacity, not speed

Two cautions that decide whether these get used. Report P90 next to the average, because averages hide exactly the cases that generate complaints. And attach every KPI to a process rather than to a department, because a KPI owned by a function drifts away from the process it is supposed to describe.

Define the KPIs on the process model so they follow the model as it changes, instead of rebuilding a dashboard each time. Process health covers the built-in aggregates, and custom dashboards covers building your own.

Related topics

• Process Charts: chart types and customization.
• Expressions in Dashboards: build custom metrics with expressions.