NGO registry

Atlas view

NGO registry

api_v1.ngo_index

One row per voluntary organisation Atlas tracks, with its identity, branding and focus.

Try at /api ↗View in lineage ↗Report a data issue

First 5 rows · api_v1.ngo_index

snapshot 2026-05-12

orgnr	slug	name	brand_name	website_url	tier	chapter_data_shape	has_chapters	primary_focus	icnpo_code_1	icnpo_code_2	icnpo_code_3	chapter_count	has_supply
943838240	fire-h	4H Norge	—	https://www.4h.no	A	cms_bins	true	youth	—	—	—	0	false
970169113	diabetesforbundet	Diabetesforbundet	—	https://www.diabetes.no	A	cms_bins	true	patient_support	—	—	—	0	false
938498318	frelsesarmeen	Frelsesarmeen	—	https://www.frelsesarmeen.no	A	cms_bins	true	social	—	—	—	0	false
940190738	lhl	Landsforeningen for hjerte- og lungesyke	LHL	https://www.lhl.no	A	cms_bins	true	patient_support	—	—	—	0	false
971322926	mental-helse	Mental Helse	—	https://mentalhelse.no	A	cms_bins	true	patient_support	—	—	—	0	false

Built from

This Atlas view joins data from the following upstream sources. Click through for provenance, methodology, and citation.

Atlas internal model ↗

dim_chapter

Conformed chapter dimension across all ingested NGOs.

dim_chapter

Showing direct dbt parents. See full lineage in /lineage/ →

Columns

Column	Description
orgnr	9-digit Brreg organisasjonsnummer (text — leading zeros are preserved). Natural key for this view.
slug	kebab-case URL-friendly identifier (e.g. 'redcross', 'kirkens-bymisjon').
name	Legal name from Brreg.
brand_name	Common brand name when distinct from legal name (e.g. 'Sanitetskvinnene' vs N.K.S.). NULL when the brand equals the legal name.
website_url	NGO's primary website. Used as the outbound link from the index page.
tier	Structural-fit tier (A / B / C) from docs/research/ngo-landscape.md. v1 includes Tier A only.
chapter_data_shape	How this NGO's per-chapter activity data is shaped — drives which ingest pattern it uses. Values: 'api_canonical', 'cms_bins', 'programme_only', 'no_structure'.
has_chapters	Whether this NGO DECLARES that it has chapters — an editorial fact from dim_ngo, recorded when the organisation is curated. Not the same question as has_supply, which is chapter_count > 0 and therefore an ingest OUTCOME: does Atlas hold chapter rows for this organisation yet. They agree today only because all curated NGOs declare has_chapters true and all have been ingested. The first NGO curated before its chapters land makes them disagree, and a consumer filtering on the wrong one loses it without an error.
primary_focus	One-line description of the NGO's main mission area.
icnpo_code_1	ICNPO classification code (primary). See ref_icnpo. NULL when unclassified.
icnpo_code_2	ICNPO classification code (secondary). NULL when not used.
icnpo_code_3	ICNPO classification code (tertiary). NULL when not used.
chapter_count	Count of active chapters for this NGO (filtered by dim_chapter.is_active = true). Zero if the NGO has no chapters or all chapters are dormant. 🔴 AND A THIRD CASE, WHICH IS THE ONE TRUE TODAY: zero also means ATLAS HOLDS NO CHAPTER DATA. All eleven organisations currently report `has_chapters = true` with `chapter_count = 0`, and a consumer reading the pair gets a contradiction (urb-agents #1388). 🔵 The two columns answer different questions and both answers are correct. `has_chapters` is a property of the ORGANISATION — it is structured in local chapters — and comes from the roster. `chapter_count` is a property of ATLAS — how many it holds — and comes from `dim_chapter`, which is empty until the Red Cross data arrives. ⚠️ So `has_chapters = true, chapter_count = 0` reads "this organisation has chapters and Atlas has none of them", not "its chapters are all dormant". See `activity_catalog`, `distrikt_summary` and `kommune_local_chapters`, which are empty for the same reason and say so. 🔵 Same shape as the FHI suppression note: the data is right and the reading is the defect.
has_supply	Convenience boolean: chapter_count > 0. True when the NGO has at least one active chapter — i.e. it appears on the supply side of Atlas.

Sample query

Live PostgREST query that returns the first 5 rows. URL adapts to the current host — local dev shows api-atlas.localhost; the deployed site shows the configured API base.

https://api-atlas.urbalurba.com/ngo_index?limit=5

About this view

One row per voluntary organisation Atlas tracks, with its identity, branding and focus. 🔴 WHAT THIS RELATION DELIBERATELY DOES NOT CARRY. The demo consumer matches an organisation's local units in Brønnøysundregistrene by a name SUBSTRING it maintains itself, and it asked Atlas explicitly NOT to publish that (urb-agents #1387): "That is a heuristic we invented and are answerable for; ngo_index has no such column and should not." 🔵 The split it drew, and it is the right one: the ROSTER is ours — orgnr, canonical name, primary_focus, tier — and the MATCHING RULE is theirs. A published pattern column would make Atlas answerable for a guess it did not make, and would make one consumer's heuristic look like a property of the organisation. ⚠️ THIS NOTE EXISTS BECAUSE THE OBVIOUS RESPONSE IS THE WRONG ONE. That consumer has now deleted three hardcoded copies of Atlas data in two days, and each time the right fix was to publish what it was copying — served_as for the unserved list, window_years for the windowed-years map. The reflex "it keeps a copy, so publish it" is correct twice and wrong here. It keeps this copy on purpose. One row per NGO in dim_ngo with chapter_count and has_supply decorations. Backs the NGO landing page (atlas-frontend /ngo) and is equivalent to the listNgos() inline query. chapter_count counts only active chapters (dim_chapter.is_active), so a defunct NGO sits at 0 even if its historical chapters are retained. has_supply is the convenience boolean (chapter_count > 0). The full table is small (~11 rows in v1, growing as new NGOs land), so PostgREST consumers typically read all of it.