Skip to content

Join only the shared time scope of two research tables

Compute an explicit common interval before comparing outputs with different requested windows.

Guides

Join only the shared time scope of two research tables — what should the operator record?

Two tables may overlap in time without being directly comparable across their full extent. A shared-window view can be useful for operational reconciliation, but it should not erase the nonoverlapping evidence. Keep each table's original scope and create a clearly named intersection view with its own lineage.

Practical operations check
Compute an explicit common interval before comparing outputs with different requested windows.
Store left_window and right_window from the manifest, distinguishing requested dates from the actual observed timestamp range. Observed minimum and maximum timestamps describe returned records only; they must not replace the requested scope when determining the intended relationship between the two tables.
Name the common interval and excluded scopes explicitly, avoiding population or trend claims based on differences between returned subsets. The published comparison should reference both original windows and the intersection contract, allowing recipients to understand excluded evidence without assuming it vanished or became invalid.
01 / Guides

Record both scope contracts

Store left_window and right_window from the manifest, distinguishing requested dates from the actual observed timestamp range. Observed minimum and maximum timestamps describe returned records only; they must not replace the requested scope when determining the intended relationship between the two tables.

02 / Guides

Compute the intersection

Use the later start and earlier end under the same boundary convention; an empty intersection should produce no forced comparison. If the intersection ends before it begins, mark no-common-window and retain both original tables, rather than shifting a boundary until some overlap appears.

03 / Guides

Partition eligible rows

Create in-common-window, outside-common-window and unknown-time buckets for each table, preserving the original membership references. An unknown timestamp is neither inside nor outside by evidence; keep that third state explicit so the common-window view does not overstate its eligibility checks.

04 / Guides

Inspect join coverage

Compare only compatible timestamps and identifiers within the intersection, explaining why unjoined rows remain unresolved rather than calling them missing from the source. A matching identifier outside the common interval can remain a historical association but should not be counted as an eligible common-window row in this particular transformation.

05 / Guides

Publish the bounded comparison

Name the common interval and excluded scopes explicitly, avoiding population or trend claims based on differences between returned subsets. The published comparison should reference both original windows and the intersection contract, allowing recipients to understand excluded evidence without assuming it vanished or became invalid.

Keep the boundaries clear.

  • Compute an explicit common interval before comparing outputs with different requested windows.
  • Operational record: Record both scope contracts. Context badges do not certify access.

Start with a read-only check: ask your agent to report actual source coverage before collecting records or exporting an audience.

Primary references

Documentation reviewed 2026-11-15. Product capabilities, client availability and plans can change. Third-party marks identify their products, not an affiliation.

Know what social media shows about you.
Start with a question.

Get started
montage