Rovo credit details (of a prompt)

Question

Hello,

I have a question regarding the Rovo credits usage.

Starting from 3/12, Atlassian will begin charging for overuse of credits. However, at the moment it is completely unclear how many credits an Rovo chat actually consumes.

As a result, it is impossible to determine why someone has used a large number of credits or how many credits specific prompts consume.

Today you can only see totals, but no details.

For me is it strange that Atlassian has decided to make Rovo credits chargeable when there is no way to determine in detail wherefor the credits were used for. Some transparency is required when you charge overuse.

So, when will Atlassian provide more detailed insights into Rovo credits usage?

My position, if Atlassian does not provide detailed visibility into how credits are consumed,they cannot expect customers to pay for additional credit usage.

Kind regards,

Dieter

Subhrajit Giri · Answer

Hi @Dieter Lauwers ,

You raise a very valid concern regarding Rovo credit consumption and the need for greater transparency, especially as credit overage charges are introduced. Without detailed usage insights, it becomes challenging for administrators to identify which prompts, workflows, or AI agents are consuming the most credits and optimize usage accordingly.

Greater visibility into credit consumption would help organizations plan budgets, establish usage policies, and confidently scale their AI adoption.

At Empyra, we help enterprises understand and optimize their Atlassian Rovo adoption. You may find these resources useful:

* Atlassian Rovo Services – Explore how enterprises can implement and optimize Rovo for their business workflows.
* Rovo Credits Calculator – Estimate Rovo credit requirements and plan your organization's AI usage.

Hopefully, Atlassian will introduce more granular credit usage reporting to help customers make informed decisions about their Rovo adoption.

Sami Shaik · Answer

@Dieter Lauwers I cannot speak to intent, but the IDs are resolvable today with two more exports, and the gaps you are seeing are already on Atlassian’s public tracker.

Resolving the user IDs. The actor column is the Atlassian account ID. Export the directory from Atlassian Administration > Directory > Users > Export users: the CSV has a User ID column (“The unique identifier for a user”), User name and Email (Export users from a site). A lookup on that column against the credits export gives you names and emails per row, and a pivot on name by month gives you the per-person figure you are after.

Resolving the agent IDs. That one is a known gap: ROVO-1267 (Suggestion, created 30 Sep 2026, Gathering Interest) asks for an “Agent name” column, and its listed workaround is to export the agents list and cross-reference. Worth a vote if you want the column.

Blank Feature and Agents cells. If some rows come back empty in those two columns, that is ROVO-1265 (Bug, created 28 Sep 2026, Gathering Impact). Atlassian is asking for examples there, so a comment with your export date and a count of blank rows is the fastest way to move it.

Two limits to plan around, from the Platform usage page: the export “covers up to 3 months of usage or the first 1 million metered events, whichever is reached first”, and the download link expires 24 hours after the email. If you want a running record, schedule the export monthly and keep the files.

Prompt-level detail is not in any of this. That is a fair ask, and the two tickets above are the right place to put it on record.

Dieter Lauwers · Answer

Hi Arkadiusz,

Export of data is a starting point, but it contains ID's of actor, site, …. not the effective agentname, actorname, … not so clear for me.

Did Atlassian intentionally choose to use IDs in the reporting to make it less transparent?

Who is also not happy with the way how Atlassian it makes all unclear/offy?

Dieter