Troubleshooting Join Path Issues in Atlassian Analytics Visual Mode
Title
Troubleshoot unexpected results from the join path in Atlassian Analytics visual mode
Platform Notice: Cloud Only - This article only applies to Atlassian apps on the cloud platform.
Summary
If your chart in Atlassian Analytics visual mode returns unexpected counts, duplicate rows, null values, or results that don't match Jira, one of the most common causes is the automatically selected join path. This article explains why results may differ from what you expect and how to fix them.
See also: This article expands on the join path guidance in Troubleshooting missing data or unexpected results | Atlassian Analytics | Atlassian Support. If your results are wrong for reasons unrelated to joins, that article covers additional causes including row limits, filters, and data freshness.
Why your results might be wrong
When you build a query in visual mode, Atlassian Analytics automatically joins tables using available foreign key relationships. If more than one valid relationship exists between tables, Analytics may select a join path that is technically valid but not correct for your reporting goal.
This is a common cause of:
duplicate records
unexpected null values
incorrect counts
results that don't match Jira or another expected source
How automatic join selection causes unexpected results
Some tables can connect to each other in multiple ways.
For example, the Jira Issue table can join to the Account table through different fields such as:
Reporter account IDAssignee account IDCreator account ID
If your chart is intended to show assignees but the generated join path uses the reporter relationship instead, the chart will return different results than expected.
How to fix it
Step 1: Review the join path
Open the chart editor.
In the query panel, select Join path.
Review the relationship diagram to see how the tables are connected.
Check whether the join columns match the question you are trying to answer.
If more than one join is available, the join path panel will show a More options label.
Step 2: Change the join path
Select Join path.
Select the alternative line or column name for the join you want to use.
The selected line will turn blue.
Select Run query to rerun the chart with the updated join path.
If the join path you need isn't available in visual mode, switch to SQL mode and write the join manually.
Common signs that the join path is wrong
Review the join path early if you notice any of the following:
your counts are slightly or significantly higher than expected
the same issue appears multiple times unexpectedly
some rows are missing even though the source data exists
the account-related column shows the wrong user role (e.g., reporter instead of assignee)
your chart looks reasonable at first glance, but doesn't match Jira when validated
Common examples
Issue and Account tables
If you want to report on who an issue is assigned to, confirm that the query joins through Assignee account ID rather than another account relationship.
Issue and sprint-related tables
Sprint reporting can be more complex because issue sprint data may flow through intermediate tables such as sprint history. Depending on the fields you add, this can produce multiple rows for a single issue if the issue has been in more than one sprint.
If you see duplicated issues in sprint reporting, review:
the join path
whether you are using sprint history data
whether you need additional filters for sprint name, ID, or status
Issue and SLA-related tables
If you join issue data to SLA data, each issue may have multiple SLA records. This can inflate counts unless the query is filtered to the specific SLA you want to analyze.
When to use SQL mode instead
Visual mode works well for many chart-building use cases, but you may need SQL mode when:
the required join path is not available as a selectable alternative
you need more control over the join conditions
you need to filter records as part of the join logic
you need more complex joins across several tables
Tip: A good validation method is to start with a simple table chart, add one new table at a time, and review the join path after each addition. This makes it easier to see when the results change unexpectedly.
Other causes of unexpected results
If changing the join path doesn't resolve the issue, also review:
Row limit reached – query truncating results
Filters applied to the wrong table
Data freshness lag – recently changed Jira data not yet available
Aggregation logic – counting rows instead of distinct issues
Timezone handling – records shifting across date boundaries
For a comprehensive checklist, see: Troubleshooting missing data or unexpected results | Atlassian Analytics | Atlassian Support
When to contact support
If you've reviewed the join path, filters, and row limits but still can't reconcile the results, contact support with:
the Jira filter or report you are comparing against
the Atlassian Analytics chart SQL or visual query
sample issue keys that appear to be affected
screenshots of both results