You can access a metrics page that provides insights into how people are using Overleaf at your organization.
Date ranges
The date range for the metrics defaults to the last 30 days, displayed by day. You can change this to any date range, displayed by day, week, or month, by clicking on the date range and selecting a new one. This option is on the top right-hand side of the Metrics page.
Understanding the metrics
Each visualization on the metrics page includes a tooltip that provides more information about the displayed graph. When looking at each plot, mouse over the (?) to see tooltips.
You can also mouse over the plots for additional information and control which categories you see by clicking on the legend.
Downloading metrics data
Types of metrics available
Active Projects
This shows the number of unique Overleaf projects edited at least once in the selected time period (day, week, or month).
Affiliations
This shows the number of unique affiliated users per time period (day, week, or month).
The initial drop in users when we enable the subscription is due to the reconfirmation period, which is usually set to 18 months. Once the subscription is enabled, all users who have not logged in during the last 18 months are removed from the plot. If these users subsequently reconfirm their affiliation, they will be included in the plot again.
Total Session Duration (hours)
This shows the total length of editing sessions, in hours, per time period (day, week, or month). This is how long your users are spending on Overleaf.
New Affiliations
This shows the number of new unique affiliated users per time period (day, week, or month).
Active Users
This shows the number of unique users that started at least one Editing Session on a given day, week, or month. An Editing Session starts when a user opens a project and lasts until there are 30 minutes of inactivity.
Editing Sessions
This shows the number of editing sessions per time period (day, week, or month).
An Editing Session starts when a user opens a project and lasts until there is 30 minutes of inactivity. Activity includes moving the cursor, typing, and clicking.
Project Activity
This shows the number of projects that had at least one Editing Session within the time period (day, week, or month).
The different colors show the breakdown of the total length of Editing Sessions on each of those projects over the past year. Click the dots in the legend to toggle the category display on and off. Double-click on a dot to show only that category.
New Projects
This shows how many new projects your users created on Overleaf per time period (day, week, or month).
Collaborators
This shows the average number of collaborators per project. Collaborator count is the number of distinct users that had at least one Editing Session during the time period (day, week, or month).
To be included, at least one collaborator must belong to your institution, and projects must have had at least one Editing Session. For ‘Internal Collaborators’, only users who belong to your institution are counted.
Click the dots in the legend to turn the category display on and off.
Collaborations
This shows the number of projects that had more than one collaborator. A collaborator is a user who had at least one Editing Session during the time period. The colors represent the different numbers of collaborators.
To be included, at least one collaborator must belong to your institution.
Internal Collaborations
This shows the number of projects that had more than one collaborator belonging to your institution. A collaborator is a user who had at least one Editing Session during the time period. The colors represent the different numbers of collaborators.
The data provided on the metrics page can be downloaded using the Download CSV option on the Metrics page. The data available here is aggregate data. High-level user access data can be downloaded from the Admin Hub page ad described in .
As time goes on, inactive users (i.e. ones who have not logged in for more than 18 months) will slowly be removed from the metrics, ensuring a more accurate representation of the number of active users. .
