URGENT BUG: Dimension filters break Time Series charts by splitting boundary weeks across months (creates false drops)

Data Studio Data Studio Q&A
1

Hello Looker Studio Team,

There is a critical data-handling bug in Looker Studio that completely breaks the integrity of Time Series charts when using custom dimension filters for time granularity (e.g., switching from Year/Month to Year/Week).

This bug distorts financial data, shows false performance drops, and makes client-facing dashboards completely unreliable.

The Issue:

When a chart is set to a Year/Month dimension, and a user applies adimension filter to view data by Year/Week, Looker Studio fails to re-aggregate the chart timeline correctly. Instead of adjusting the chart axis to weekly buckets, it retains the monthly grouping structure underneath.

As a result, weeks that cross over the boundary of two months (e.g., a week starting in late January and ending in early February) are split into two separate data fragments.

The Impact:

  • A full week of revenue data is chopped into fragments of 2 or 3 days.

  • The chart displays a severe, artificial drop towards zero on those boundary weeks (see attached screenshot).

  • This is not a data source issue (it happens across Google Sheets, BigQuery, and GA4 connectors alike). It is a core Looker Studio aggregation rendering bug.

Why current workarounds are unacceptable:

  • Using Drill-down: While drill-down works, it forces us to train hundreds of non-technical stakeholders to find and click hidden chart-header arrows instead of using standard, intuitive dashboard filters.

  • Parameters with CASE fields: Looker Studio Time Series charts do not support native date parameters inside dynamic CASE WHEN statements without breaking the timeline configuration.

Steps to Reproduce:

  1. Create a Time Series chart with any metric and Year/Month as the dimension.

  2. Create a Dropdown/Advanced Filter using a Year/Week (ISO) dimension.

  3. Select a specific week that spans across two months.

  4. Observe the massive, incorrect dip in the line chart where the week is chopped in half.

Year / Month

image
image588×181 15.9 KB

Year / Week

image
image597×169 30 KB

This bug severely undermines data accuracy and professional reporting. We need the product team to look into how Looker Studio handles cross-dimension time filtering on Time Series elements immediately.

Please acknowledge this issue and let us know when a hotfix can be expected.

1 Like