Skip to main content
Open Performance in the agent’s left sidebar, or select Open performance on the agent’s home page, to review its credit usage, task outcomes, evaluation results, and task history. The page has three tabs: Insights, Tasks, and Evaluations.
Performance page on the Insights tab with Usage & Credit Trend, Task outcomes, the overview metrics Credits, Tasks, Avg. credits per task, and Members, an activity grid, Impact report, Evaluations, and Reflections
Performance reporting depends on your analytics permissions. Access to a saved conversation also follows Agent access, including your organization’s conversation visibility policy. For organization-wide reporting, see Insights.

​
Insights

Pick a date range, and optionally Compare to previous period to see the change. The chart’s interval is chosen for you to suit the range, and you can switch to any other interval that still makes a readable chart:
Intervals that would produce a single bar or thousands of them are not offered, so a short range cannot be charted by month and a year cannot be charted by the hour.
Usage & Credit Trend charts credits and tasks so you can distinguish a rise in spend from a rise in usage. Hover a point to see cost per task. The overview on the right gives you the headline numbers for the range: Further down, two breakdown tables:
  • Credits by: Models, Sources (where tasks came from, such as a trigger, Slack, or the app), Triggers (credits charged per trigger), Members, and Connectors.
  • Usage by: Connectors, Skills, Triggers, Artifacts, Subagents, and Knowledge Sources. The Triggers view lists Runs, Failed, and Last run for each trigger.
Reading it. High credits with few tasks usually means long tasks on an expensive model: check Credits by Models and consider a cheaper model or a lower summarization trigger. Lots of tasks with almost no connector usage often means people are asking the agent things it has no tools for.

​
Task outcomes

Under the trend chart, Task outcomes charts how tasks ended over time, on the same range and interval as the rest of the tab:
Read it next to Usage & Credit Trend. Cost falling while Failed rises is a cheaper model doing worse work, not a saving. A steady band of Approval Required usually points at a missing tool or an instruction gap for one specific kind of request.

​
Which kinds of tasks cost the most?

Cost vs completion plots each kind of task the agent handles, such as Answer a question or Produce content. Higher means a better completion rate, further right means more credits per task, and a bigger dot means more tasks. Hover a dot to see Tasks, Credits per task, Completed, and Tool calls per task. Gumloop labels tasks shortly after they finish. Why tasks didn’t complete lists the reasons tasks failed, with a count for each. If you can view the agent’s configuration, select a dot or a reason to open the matching tasks on the Tasks tab.
Cost vs completion chart with a tooltip for Answer a question showing Tasks 5, Credits per task 49.4, Completed 80%, and Tool calls per task 3.6, above the Why tasks didn't complete list

​
Is a connector failing?

Usage by > Connectors shows each connected app’s Share of tool calls, Failed share, Typical (median) response time, and Calls. Needs reconnect marks a connector whose account you need to refresh.
Credits by table on Models with Grok 4.7 and DeepSeek V4.1 Flash, and Usage by table on Connectors listing Exa and Firecrawl with Share, Failed, Typical, and Calls columns

​
What changed, and what it did to cost

Markers along the bottom of Usage & Credit Trend show when the agent was changed, so you can line a change up against the spend and volume around it. Each marker covers one interval of the chart. Select one to see the changes inside it, newest first, with who made each one:
  • Agent created
  • Model changed: the model it moved to, and the one it came from
  • Instructions edited
  • Connectors, skills, and knowledge sources added or removed
  • Reflections the agent applied. Select one to open the suggestion.
When enough data exists on both sides, the card also shows how cost per task moved after the change, so a model swap’s effect is visible without doing the arithmetic. It appears only once both the before and after windows have completed, and it stays quiet when the movement is within the agent’s normal noise.
A change made before the agent ran any tasks is marked No tasks yet, since there is nothing yet to compare against.

​
Impact report

Have Gumloop send this summary out on a schedule, by Email or to a Channel, to the people you add under To. Useful for keeping a team lead in the loop without giving them the builder.

​
Evaluations and Reflections

Both features live in the summary panel on the right of this page, and both are switched on from here: See the linked guides for criteria, tags, data points, schedules, reports, and API details.

​
Evaluations tab

Once evaluations are on, this tab shows the grades: how completed tasks scored against your criteria, which ones failed, and the trend over time. That is where you catch a regression after changing the instructions or the model.

​
Tasks tab

Review this agent’s task history to see who started each task, its source, and its credit cost. Open a saved task you have access to when you need to investigate an evaluation result. Filter the list by task label, Failure reason, Date range, and Credits (tasks over a credit amount). Task labels come in two groups, named Kind of work and Area unless your organization renamed them. With analytics access, incognito tasks also contribute to counts and can appear as Incognito Chat. Creator and usage metadata can remain visible, but there is no durable message transcript to open. See Incognito mode.

Evaluations

Define criteria and grade the agent’s answers.

Reflections

Let the agent propose its own improvements.

Credits

What the agent is charged for.

Organization Insights

The same picture across every agent and team.