Ask questions about content performance in natural language. The agent converts requests into analytics reports based on Viewtracker data. Instead of opening a report and setting filters, you describe the numbers you need in Rovo Chat, and the agent returns a tabled report built from Viewtracker data.
The provided answer is an initial summary. If you need help interpreting the results or deciding on next steps, ask, and Rovo will assist.
The agent is part of Viewtracker - Analytics for Confluence Cloud. It does not replace the built-in reports. Use the reports to explore data interactively. Use the agent when you have a specific question, want an answer without leaving the page you are on, or want a report delivered on a schedule.
What you can ask for
One skill handles every request. A skill produces one type of report. For the prompts, options, and output of each one, see What the Viewtracker Agent can do.
|
Report |
What you get |
|---|---|
|
A ranked list of the most viewed content in one or more spaces. Use it when leadership asks "what's our most-read content this quarter?" and you need the answer in the next five minutes. |
|
|
A performance snapshot for a single page, blog post, or attachment. Use it when you're about to update a page and want to know if anyone's actually reading it before you invest the time. |
|
|
A list of content filtered by Confluence content status and by view or edit activity. Use it during your monthly space cleanup to identify pages still in Draft that haven't been touched in 90 days. |
|
|
Aggregated metrics for one or more spaces. Use it to compare how two department spaces are performing, or to report space health to a space owner. |
|
|
The users who viewed a specific piece of content. Use it when you need to confirm that key stakeholders actually saw a project update before a decision meeting. |
|
|
The users who edited a specific page or blog post. Use it to recognize active contributors or spot pages that depend on a single author — a knowledge-concentration risk. |
|
|
A side-by-side comparison of several pages or blog posts. Use it when you've published a new version of an article and want to see if it's outperforming the old one. |
|
|
The members of a user group who did not view a space or a piece of content. Use it for quarterly compliance checks: "who in the Security team hasn't viewed the updated data-handling policy?" |
|
|
A support request in the Elevatic helpdesk that carries your feedback about the app and the agent. Use it when something in the agent didn't work as expected — one prompt files a support request without leaving Confluence. |
How the agent builds a report
-
The agent matches your request to the skill that fits your intent. If several skills could fit, it asks a short question or explains which skill it chose.
-
It turns your request into concrete parameters: Scope, date range, filters, sorting, and number of results.
-
For every parameter you leave out, it applies the documented default and names that default above the table.
-
It builds the table from Viewtracker tracking data and Confluence metadata.
In Rovo Chat, each answer has three parts: The parameters used, the tabled report, and a list of options for refining the report. Reports the agent delivers through an automation contain only the parameters and the table.
The agent uses only data that Viewtracker tracked and metadata that Confluence provides. It sorts, filters, groups, and reformats that data, and calculates values such as sums and percentages. It never estimates or extrapolates, and it never fills a gap with an assumed value. When the data you ask for does not exist, the agent states what is not possible instead of approximating.
Key concepts
-
Skill: One report type with its own parameters, defaults, and output. You do not need to select a skill by name. The agent picks it from your prompt.
-
Context page: The Confluence page you have open when you start the chat. The agent uses this page, or its space, as the default scope when your prompt does not name one.
-
Query ID: The identifier the agent returns when a report takes longer than 25 seconds. Use it to pick up the result later.
What you need
-
Viewtracker - Analytics for Confluence Cloud, version 8.0.0 or later.
-
Rovo is available in your Confluence Cloud site.
-
View access to the content or space you ask about, and access to its Viewtracker Space and Content Reports.
-
No admin rights. Setting up a scheduled report additionally requires access to Atlassian Automation.
The agent needs no separate installation and has no settings of its own. It arrives with the app.
What you see depends on your permissions
The agent applies your Confluence permissions and the Viewtracker report permissions to every request. Content and spaces you cannot access aren't listed in a report or accepted as a scope.
|
Situation |
What happens |
|---|---|
|
You name a space you cannot access |
The agent asks you to select a space you can access. |
|
You name a space whose Space and Content Reports are restricted to other groups |
The agent asks you to select a space you can access. |
|
You name several spaces and can access some of them |
The agent asks whether it should continue with the spaces you can access. |
|
You name content you cannot access |
The agent asks you to name content you can access. |
|
You name content whose Viewtracker report is restricted |
The agent asks you to select content you can access for reporting. |
What the privacy setting changes
The Viewtracker data privacy setting decides whether the agent can name individual users. A Confluence global admin sets it in the Global Report under Privacy.
|
Privacy setting |
Content viewers, content contributors, and non-visitors |
|---|---|
|
Standard |
Available to every user with report access. |
|
Hide viewers |
Available only to members of the permitted user group. |
|
Extended privacy |
Not available. The agent explains that the privacy settings prevent the report. |
In extended privacy mode and in hide viewers mode for users outside the permitted group, the content lifecycle report hides the Last viewer column. Content owner and last editor stay visible.
Limitations
-
Viewtracker tracks content from the moment it is installed. Content that was never viewed, edited, or commented on since then is unknown to Viewtracker and does not appear in a report.
-
A custom date range covers a maximum of 365 days.
-
The date ranges today, this week, this month, and this year are incomplete because the period has not ended. Comparisons for these ranges use the completed previous period.
-
The content lifecycle report covers one space per request and supports only native Confluence content statuses. Statuses from third-party apps are not available.
-
The space performance report covers up to 5 spaces per request. The top-performing content report covers up to 100 spaces.
-
Reports that take longer than 25 seconds return a query ID instead of a table.
-
In an automation, the agent cannot ask you for missing parameters. Name every parameter in the prompt.
-
The agent is available on Confluence Cloud. There is no Data Center equivalent.
Related pages
Continue with these guides to learn how to request, retrieve, and schedule reports with the Viewtracker Agent.