Collaboration
Collaboration Graph Data
Filename
collaboration-graph/organizationName_ANONYMIZED_COLLABORATION_GRAPH_YYYYMMDDZ00000000000N.extension
Variables:
default prefix: collaboration-graph/
organizationName: the slugified name of the organization
YYYYMMDD, week this file contains data from. Week is in ISO format, being Monday first day of the week.
due to size file might be split into N consecutive chunks using the suffix 0000000000N.
extension, depending on the format and compression, see below:
Available formats
NONE,GZIP
.csv[.gz]
NONE,GZIP
.json[.gz]
NONE,DEFLATE,SNAPPY
.avro
NONE,SNAPPY,GZIP,ZSTD
.parquet[.gz]
File contents
Provides a flattened version of the collaboration graph for the company on a given week. Each row has a source employee id, meaning is the person that initiated the collaboration work and a target employee id, the person with whom collaborated. Also the total amount of time of that collaboration during the week, in seconds and in minutes.
Sample row
NDJSON example (note examples are pretty printed for documentation purposes otherwise will be presented in one row each)
Collaboration graph schema
week
DATE
Week when collaboration happened
employeeIdSource
STRING
Employee that initiated the collaboration
employeeIdTarget
STRING
Employee affected by employeeSourceId events
weeklyCollaborationTimeSeconds
INTEGER
Time of collaboration in the week expressed in seconds. Collaboration is the sum of the estimated time employeeIdSource spent on all events that were related with employeeIdTarget.
weeklyCollaborationTimeMinutes
FLOAT
weeklyCollaborationTimeSeconds but expressed in minutes
Collaboration Graph Data Split by Tool
Filename
collaboration-graph-split/organizationName_ANONYMIZED_COLLABORATION_GRAPH_SPLIT_YYYYMMDDZ00000000000N.extension
Variables:
default prefix: collaboration-graph-split/
organizationName: the slugified name of the organization
YYYYMMDD, week this file contains data from. Week is in ISO format, being Monday first day of the week.
due to size file might be split into N consecutive chunks using the suffix 0000000000N.
extension, depending on the format and compression, see below:
Available formats
NONE,GZIP
.csv[.gz]
NONE,GZIP
.json[.gz]
NONE,DEFLATE,SNAPPY
.avro
NONE,SNAPPY,GZIP,ZSTD
.parquet[.gz]
File contents
Provides a flattened version of the collaboration graph for between individuals in the company on a given week. Each row has a source employee id, meaning is the person that initiated the collaboration work and a target employee id, the person with whom collaborated, the tool they used to collaborate and the total amount of time of that collaboration during the week, in seconds and in minutes (rounded) For a pair of employee source and employee target there will be as many rows as different tools they used to collaborate on a given week.
Collaboration graph schema
week
DATE
Week when collaboration happened
employeeIdSource
STRING
Employee that initiated the collaboration
employeeIdTarget
STRING
Employee affected by employeeSourceId events
sourceType
STRING
The type of events (gcal, gdrive, outlook-cal, zoom, slack, jira...)
weeklyCollaborationTimeSeconds
INTEGER
Time of collaboration in the week expressed in seconds. Collaboration is the sum of the estimated time employeeIdSource spent on all events of the particular sourceType that were related with employeeIdTarget.
weeklyCollaborationTimeMinutes
FLOAT
weeklyCollaborationTimeSeconds but expressed in minutes
Sample rows
NDJSON example (note examples are pretty printed for documentation purposes otherwise will be presented in one row each) of two rows for the same pair of collaborators, but different tools
Last updated