Overview
The Daily Activity Count calculator tracks how many times each activity occurs on each calendar day, providing a detailed breakdown of activity frequency over time. Unlike the Daily Event Count calculator which shows total event counts per day, this calculator separates counts by activity type, allowing you to identify patterns and anomalies for specific activities in your process.
This calculator operates at the event level and groups data by both activity name and date, making it particularly valuable for understanding workload distribution, identifying data quality issues for specific activities, and monitoring operational patterns for individual process steps.
Common Uses
- Identify which activities experience volume spikes or drops on specific days
- Detect data quality issues affecting specific activities (missing activity data on certain dates)
- Analyze workload patterns for individual process steps across time
- Monitor staffing needs by tracking activity-specific workload by day
- Validate data extraction completeness for critical activities
- Compare activity frequency trends to identify process changes or bottlenecks
- Track seasonal or cyclical patterns for specific process activities
Settings
Time Period: While the setting includes a time period option, the calculator currently groups all data by calendar day (the date component only, ignoring the time). This means all activity counts are aggregated to the daily level regardless of when during the day they occurred.
Standard Fields:
- Title: Optional custom title for the calculator output
- Description: Optional description for documentation purposes
Examples
Example 1: Detecting Activity-Specific Data Issues
Scenario: Your invoice processing team reports that "Approve Invoice" activities seem to be missing from certain days in your event log. You want to verify whether this is a data extraction issue or if approvals genuinely didn't occur on those days.
Settings:
- Title: "Approval Activity Daily Tracking"
- Description: "Verify completeness of invoice approval data"
Output:
The calculator displays a table with three columns:
- Date: Each calendar day in your event log
- Activity Name: The name of each activity that occurred
- Count: How many times that activity occurred on that date
Example output:
Date Activity Name Count
2024-03-10 Submit Invoice 234
2024-03-10 Approve Invoice 198
2024-03-10 Pay Invoice 156
2024-03-11 Submit Invoice 248
2024-03-11 Approve Invoice 0
2024-03-11 Pay Invoice 12
2024-03-12 Submit Invoice 241
2024-03-12 Approve Invoice 215
2024-03-12 Pay Invoice 187
Insights: On March 11th, "Approve Invoice" shows zero occurrences while submissions and payments continued. This pattern indicates a data extraction problem rather than a genuine business pause, since you cannot have payments (156 the previous day, 12 on March 11th) without prior approvals. The dramatic drop from 198 approvals to 0, followed by recovery to 215 the next day, confirms a one-day data gap that needs investigation.
Example 2: Analyzing Activity Workload Patterns
Scenario: You're managing a customer service process and want to understand the daily volume patterns for different types of customer interactions to optimize staffing levels.
Settings:
- Title: "Customer Service Activity Volume Analysis"
- Description: "Track daily volumes for staffing optimization"
Output:
The output shows daily counts for activities like "Create Case," "Update Case," "Escalate Case," and "Close Case" across multiple weeks. When visualized as a line chart with one line per activity, you can identify:
- "Create Case" peaks on Mondays (backlog from weekend)
- "Close Case" peaks on Fridays (weekly cleanup efforts)
- "Escalate Case" shows consistent low volume except for month-end spikes
- "Update Case" remains relatively stable across all weekdays
Insights: The Monday spike in new case creation (often 40-50% higher than other weekdays) suggests you need additional staff on Mondays for case intake. The Friday peak in closures indicates staff prioritize completing cases before the weekend. Month-end escalation spikes align with customer urgency around billing cycles, suggesting you should schedule senior staff availability during these periods.
Example 3: Identifying Process Changes Over Time
Scenario: Your organization implemented a new procurement approval workflow on June 1st. You want to verify whether the new "Pre-Approval Review" activity is being used consistently and whether it replaced or supplemented the existing "Standard Approval" activity.
Settings:
- Title: "Procurement Workflow Transition Analysis"
- Description: "Track adoption of new pre-approval step"
Output:
The calculator shows daily counts for both approval activities across May and June:
Before June 1st:
Date Activity Name Count
2024-05-28 Standard Approval 145
2024-05-28 Pre-Approval Review 0
2024-05-29 Standard Approval 152
2024-05-29 Pre-Approval Review 0
After June 1st:
Date Activity Name Count
2024-06-01 Standard Approval 89
2024-06-01 Pre-Approval Review 58
2024-06-02 Standard Approval 72
2024-06-02 Pre-Approval Review 78
2024-06-03 Standard Approval 45
2024-06-03 Pre-Approval Review 103
Insights: The new Pre-Approval Review activity appeared immediately on June 1st and its volume increased daily (58, 78, 103) as staff adopted the new process. Meanwhile, Standard Approval counts decreased (145 to 89 to 72 to 45), showing the workflow change is working as intended. The combined total (147 on June 3rd) is similar to pre-change volumes (145-152), indicating the new step is supplementing rather than creating additional work. By mid-June, you'd expect Pre-Approval Review to handle most cases with Standard Approval limited to exceptions.
Output
The calculator produces a data table with the following columns:
Date (DateTime): The calendar date for each group of activity occurrences. The time component is always set to 00:00:00 (midnight) as the calculator groups by date only.
Activity Name (Text): The exact name of the activity as it appears in your event log. Activity names are case-sensitive and must match exactly to be grouped together.
Count (Number): The number of times this specific activity occurred on this specific date. This is always a positive whole number representing individual event occurrences.
The output can be visualized as:
- Line charts: Show trends for multiple activities over time (one line per activity)
- Stacked bar charts: Compare the composition of daily activity volumes
- Heat maps: Display activity intensity across dates and activity types in a calendar grid
- Pivot tables: Group by activity to see date-range totals, or by date to see activity distribution
- Filtered views: Focus on specific activities or date ranges of interest
Note: If an activity did not occur on a specific date, there will be no row for that activity-date combination in the output (the count will not appear as zero). Only actual occurrences are recorded. Events with missing timestamps or null activity names may be excluded or grouped together, depending on data quality.
This documentation is part of the mindzie Studio process mining platform.