Table filter to show only highest selected week dynamically

Looker Q&A Looker Studio
,
1

Hi Guys,

I’m running into a challenge with a table in Looker Studio and would appreciate guidance.

Scenario:

  • I have a table with a week_number dimension.

  • Users can select multiple weeks using a filter, e.g., 10, 11, 12, 16.

  • I want the table to show only the rows corresponding to the highest selected week (16 in this example).

Important requirements:

  • No aggregation or consolidation across weeks — Top‑N or MAX aggregations are not acceptable.

  • This should work dynamically for any combination of past weeks, not just the latest week in the dataset.

  • Essentially, the table should always display only the highest week among the user-selected weeks, at row-level detail.

I’ve explored:

  • Top‑N filters → only work for charts, not tables, and aggregate data.

  • Calculated fields (MAX(week_number), Is Latest Week flags) → fail because tables cannot compare a row-level dimension to a metric.

  • Blended data → does not allow joining a row-level dimension to a dynamic MAX metric.

  • Parameters → would require manual selection of the week, not fully dynamic.

Given these constraints, is there any way within Looker Studio to achieve this behavior for tables, or is this fundamentally a platform limitation?

Thanks in advance for any ideas or suggestions!

2

Hi Sudith,

Try using a self blend join. joining same table to itself but with a aggregation change for second table.
In detail:

  1. Table 1: take the fields along with week number field.
  2. Table 2: take the week number field (as metric) but set the aggregation to MAX() – Do not add any dimensions here. This creates a single-value result representing the highest selected week.
  3. Not do an inner join of both tables on the basis of field week number.

When the user selects weeks 10, 11, 12, and 16 in the filter:

  1. Table 2 will calc the max(week number) across the filtered set which
  2. Inner join looks at table 1 and only keeps where week number == 16