Assessment Lifecycle
How are assessments added, changed, or retired?
A series may be initiated, terminated, or changed at any time, driven by the availability and quality of raw data from the sources in each country. Because most data comes from public official sources, the continuity of an assessment ultimately depends on those sources continuing to publish data that meets the assessment's requirements — and the commitment is to quality over continuity: an assessment that can no longer be produced reliably is changed or retired rather than published at a lower standard.
Every addition, change, and retirement is documented in the monthly, publicly available Intratec Release Notes. Methodology changes are additionally announced in advance, with a defined transition period, before they take effect; how such changes reach subscribers is detailed on the Model Selection and Evaluation page and in the Validation & Accuracy section.
What happens when an assessment's data falls short?
When data for an assessment is insufficient, inadequate, or unavailable for a prolonged period, one of four approaches is adopted, in the service of keeping the published series representative:
- Use a model — estimate the assessment with a model built on good previous and/or related data.
- Change the data source — replace the failing source with another that meets the requirements.
- Alter the assessment — adjust the assessment's definition so it can be produced reliably from the data available.
- Discontinue the assessment — if the lack of suitable data persists, retire the series rather than publish unreliable figures.
Short-term gaps — a source that is merely lagging or temporarily unavailable — are handled earlier in the pipeline through model-generated Preliminary estimates and quick source replacement, as described on the Source Timing and Coverage page.
How are assessments kept relevant over time?
Relevance is maintained through the same methodology review process that governs the report generally: assessments evolve as reviews act on raw-data changes, modeling enhancements, and subscriber feedback, so a series is kept accurate rather than left to drift. The review cadence, the three review triggers, and the shadow-testing step before a revised method takes effect are detailed in How often is the methodology reviewed?.