Outcomes: `all_data`

all_data is the consolidated, analysis-ready view of the Net Purpose data universe, and covers the majority of the question set. Everything the foundation layer produces — reported facts, calculated KPIs and estimated values — lands here, already standardised, and enriched with intensities and year-on-year changes.

Where two foundation datasets produce the same (entity_id, question_id, year) triple, both rows will be in the all_data table, with the default_sort_order asc providing the standard Net Purpose resolution.

If you only query one table in the share, query this one.

Versioned — filter is_current = TRUE.

`all_data`

Column	Type	Description
`all_data_id`	VARCHAR	Primary key
`entity_id`	NUMBER	Who. Joins to `entities`
`question_id`	NUMBER	What. Joins to `questions`
`canonical_question_id`	VARCHAR	The question’s readable slug (`metric_id` / `kpi_id`), for convenience
`year`	NUMBER	Fiscal year
`start_date` / `end_date`	DATE	Exact period the value covers
`value`	FLOAT	The value, in the question’s standard unit; NULL when not disclosed
`unit_id`	NUMBER	Standard unit. Joins to `units`
`data_type`	VARCHAR	How the value was produced: `Reported`, `Calculated`, `Estimated`, `Not Disclosed`, plus revenue and fund roll-up variants
`data_origin_class_id`	NUMBER	1 = Reported, 2 = Derived, 3 = Estimated (see Conventions)
`is_not_found`	BOOLEAN	`TRUE` when we researched the question and the company does not disclose it
`year_rank`	NUMBER	1 = most recent year for this entity and question
`default_sort_order`	NUMBER	Net Purpose default deuplication order
`evic_per_million`	FLOAT	Value per million USD of the entity’s EVIC (enterprise value including cash)
`revenue_per_million`	FLOAT	Value per million USD of the entity’s revenue
`yoy_delta`	FLOAT	Absolute change versus the previous year
`yoy_perc`	FLOAT	Percentage change versus the previous year

`all_data_origins`

Provenance for each all_data row: which foundation-layer records produced it. A row can have multiple origins (for example a calculated value with several inputs).

Column	Type	Description
`origin_id`	VARCHAR (UUID)	Primary key
`all_data_id`	VARCHAR	The `all_data` row
`data_type`	VARCHAR	Origin kind
`fact_id`	NUMBER	For reported values: the fact. Joins to `revenue_facts` / `outcomes_facts`
`kpi_result_id`	VARCHAR	For calculated values: the KPI result. Joins to `kpi_results`
`technology_id`	VARCHAR	For estimated values: the modelled technology

Example: a company’s latest values across all questions

SELECT q.name        AS question,
       ad.year,
       ad.value,
       u.name        AS unit,
       ad.data_type,
       ad.yoy_perc
FROM NET_PURPOSE.OUTCOMES.ALL_DATA ad
JOIN NET_PURPOSE.REFERENCE.QUESTIONS q ON q.question_id = ad.question_id
JOIN NET_PURPOSE.REFERENCE.UNITS u     ON u.unit_id = ad.unit_id
WHERE ad.is_current = TRUE
  AND ad.entity_id = 12345
  AND ad.year_rank = 1
  AND ad.is_not_found = FALSE
QUALIFY ROW_NUMBER() OVER (PARTITION BY entity_id, q.question_id ORDER BY default_sort_order asc) = 1
ORDER BY ad.default_sort_order;

Example: All estimated data

SELECT e.name,
  e.entity_id,
  q.name,
  ad.value,
  u.name,
  ad.year
FROM NET_PURPOSE.OUTCOMES.ALL_DATA ad
JOIN NET_PURPOSE.ENTITIES.ENTITIES e
   ON ad.entity_id = e.entity_id
JOIN NET_PURPOSE.REFERENCE.QUESTIONS q
   ON q.question_id = ad.question_id
JOIN NET_PURPOSE.REFERENCE.UNITS u
   ON u.unit_id = ad.unit_id
WHERE ad.is_current
  AND ad.data_origin_class_id = 3
ORDER BY entity_id asc, year_rank desc

Outcomes: all_data

all_data

all_data_origins

Example: a company’s latest values across all questions

Example: All estimated data

Outcomes: `all_data`

`all_data`

`all_data_origins`