Economic releases: separate the number from the surprise

Identify exactly what was measured

Start with the agency, series, units and reference period. The publication date is usually later than the period being measured. A monthly change, year-on-year change and annualised quarterly rate answer different questions and should not share an unlabeled column.

For inflation, a smaller positive rate means prices are rising more slowly, not that the overall price level has fallen. Check whether the figure is seasonally adjusted and whether it covers all items or a subset before comparing it with a chart or another headline.

Actual, expected and previous are separate inputs

A market surprise is a comparison with an expectation recorded before publication. A consensus figure needs its own provider, timestamp and definition. An official statistical agency generally publishes the observation, not a definitive forecast of what the market expected.

Suppose a fictional release is 0.3% against a recorded 0.2% expectation. That is a +0.1 percentage-point difference. It does not establish the direction of the next market move. Other components, positioning and changes in the expected policy path may matter at the same time.

Keep revisions visible

A previous number on a calendar may already have been revised. Keep the value known before the new release and the revised value in separate fields. Additional responses, updated seasonal factors and benchmarking can affect different series on different schedules.

Read the agency’s release notes rather than assuming every revision has the same cause. The revision policy for an unadjusted price index can differ from that for an employment estimate or a seasonally adjusted series. Preserve the data vintage when reviewing an old reaction.

Write a release note before a trade narrative

A useful short note records what changed, what was expected, what was revised and what remains uncertain. Add the observed market move with its time window as a separate fact. A move that follows a release is not proof that one headline caused it.

• Use the official calendar and the release’s own time zone.
• Keep source links, definitions and timestamps.
• Record contradictory details rather than choosing only the component that fits the story.