SDI Intelligence

The SDI (Sustainable Development Investments) layer applies the SDI ruleset on top of the revenue data: It augments some company assessments with bespoke per industry formulae taking outcomes into consideration to better understand and classify the revenue of the entities. It then aggregates these up into a classification system for all entities.

Both tables are versioned — filter is_current = TRUE.

`sdi_formula_results`

The output of the SDI revenue-allocation formula: for all cases where

Column	Type	Description
`sdi_formula_result_id`	VARCHAR (UUID)	Primary key
`entity_id`	NUMBER	Who. Joins to `entities`
`year`	NUMBER	Fiscal year
`segment_id`	NUMBER	The business segment the revenue comes from; joins to `segments`
`segment_name`	VARCHAR	Segment name, denormalised for convenience
`taxon_id`	NUMBER	The taxonomy activity the revenue is allocated to; joins to `taxonomy`
`taxon_name`	VARCHAR	Taxon name, denormalised for convenience
`revenue_amount`	FLOAT	Revenue of the segment, USD
`revenue_pct`	FLOAT	Segment revenue as a share of the entity’s total revenue
`allocated_revenue_amount`	FLOAT	Portion of that revenue allocated to the taxon by the SDI formula, USD
`allocated_revenue_pct`	FLOAT	Allocated revenue as a share of total revenue
`sdg_alignment_type`	VARCHAR	Alignment of the allocation: `positive`, `negative`, `potential`, `potentially_positive`, `potentially_negative` or `not_covered`. Included for cases where the formula has a determination but no matching taxon.
`confidence`	NUMBER	Confidence score of the mapping

`sdi_classifications`

The headline SDI result: one classification per entity and year. A company can have a higher confidence of being decisive versus majority if the decisive threshold (10%) is met with more confidence than the majority one (50%).

Column	Type	Description
`entity_id`	NUMBER	Who. Joins to `entities`
`year`	NUMBER	Fiscal year
`end`	DATE	Period end date
`sdi_classification`	VARCHAR	The entity’s SDI classification for the year
`sdi_classification_confidence`	NUMBER	Confidence score for the classification
`is_sdi_decisive`	BOOLEAN	Whether the classification is decisive under the SDI ruleset
`is_sdi_decisive_confidence`	NUMBER	Confidence score for the decisiveness flag

The set of entities in scope for SDI is flagged on the entity itself: entities.is_sdi_covered.

Example: positively aligned revenue per company

SELECT sfr.entity_id,
       e.name,
       sfr.year,
       SUM(sfr.allocated_revenue_amount) AS positive_revenue_usd,
       SUM(sfr.allocated_revenue_pct)    AS positive_revenue_pct
FROM NET_PURPOSE_REVENUE.REVENUE.SDI_FORMULA_RESULTS sfr
JOIN NET_PURPOSE.ENTITIES.ENTITIES e
  ON e.entity_id = sfr.entity_id AND e.is_current = TRUE
WHERE sfr.is_current = TRUE
  AND sfr.sdg_alignment_type = 'positive'
GROUP BY sfr.entity_id, e.name, sfr.year
ORDER BY positive_revenue_pct DESC;

Example: screen a universe by SDI classification

SELECT e.entity_id,
       e.name,
       e.primary_isin,
       sc.year,
       sc.sdi_classification,
       sc.sdi_classification_confidence
FROM NET_PURPOSE_REVENUE.REVENUE.SDI_CLASSIFICATIONS sc
JOIN NET_PURPOSE.ENTITIES.ENTITIES e
  ON e.entity_id = sc.entity_id AND e.is_current = TRUE
WHERE sc.is_current = TRUE
  AND e.primary_isin IN ('US0378331005', 'GB0007980591')
QUALIFY ROW_NUMBER() OVER (
    PARTITION BY sc.entity_id ORDER BY sc.year DESC
) = 1;