Data Mismatch for Visits and Unique Visitors in Adobe Analytics

Issue

The visits and unique_visitors (or similar visitor) metric values in your destination do not match the totals shown in Adobe Analytics Workspace for the same date range. Other metrics such as orders, revenue, and page_views may match as expected.

Environment

Connector: Adobe Analytics

Cause

The Fivetran Adobe Analytics connector fetches data from the Adobe Analytics Reporting API in time buckets determined by the Aggregation setting (API field: date_granularity; default: HOUR). For each sync, data is requested and stored one bucket at a time — for example, one row per hour or one row per day per dimension combination.

Visits and Unique Visitors are non-additive metrics in Adobe Analytics:

Visits: A visit session can span a time boundary. For example, a visit that starts at 11:50 PM and ends at 12:10 AM crosses a midnight boundary. Adobe counts it in both the 11 PM bucket and the 12 AM bucket. When the connector stores hourly or daily buckets and you sum them, boundary-spanning visits are counted more than once, producing a higher total than a single monthly query would return. See Adobe's Visits documentation for details.
Unique Visitors: Adobe de-duplicates unique visitors within the scope of a single reporting period. The same visitor appearing on five different days counts as 5 in daily buckets but as 1 in a monthly report. Summing daily unique visitor counts from the destination therefore overstates the true monthly unique visitor count. See Understanding Basic Visitor Metrics for details.

The destination data accurately reflects what the Adobe API returned for each individual time bucket. The discrepancy is not a connector defect, it is a fundamental property of how Adobe calculates these metrics.

Resolution

You can't reliably calculate an Adobe Workspace monthly total by summing hourly or daily visits or unique_visitors values, because the same visit or visitor can appear in multiple buckets. Choose the approach that best matches your use case:

Compare data at the same granularity

When validating data against Adobe Workspace, use the same date range in both. Instead of summing destination values over a month and comparing to a single monthly Workspace number, filter Workspace to a single day or hour that corresponds to one bucket in your destination.

Use a coarser date granularity

If monthly-level totals are more important than hourly or daily detail, you can create a new connection with the Aggregation setting (date_granularity) set to DAY or MONTH. This reduces the number of cross-boundary visits and, for MONTH, produces values that match Adobe's monthly de-duplication of unique visitors directly.

Because the Aggregation setting affects how data is aggregated and structured, you cannot change it on an existing connection. You must create a new connection with the desired value.

Use an additive metric for cross-period comparisons

For counts that must add up correctly across time periods, use additive metrics such as page_views, orders, or custom event counts instead of visits or unique_visitors. These metrics do not have the same boundary-spanning or de-duplication behavior.