Ask what was knowable at the time.

Define the information boundary before interpreting a historical result.

Separate the clocks

A research record can distinguish the period being measured, the time a value was published, the time it was collected and the time it was revised. These fields answer different questions and should not be silently substituted for one another.

Declare the rule before the test

Specify which publication timestamp makes an observation eligible. Where the timestamp is unavailable, document the uncertainty and avoid presenting the result as an exact historical reconstruction.

Keep revisions visible

Store a new version alongside the earlier value when practical. A revised series can be useful for describing an economy or company; a historical decision test needs an explicit policy for which version it uses.

A minimal working template

Suggested fields: series ID; observation period; published at; collected at; revision ID; value; unit; original source; eligibility rule; quality note. This page defines a documentation template and does not supply a market-data feed.