Many sources, integrated
into one queryable view.
What we integrate
Court decisionsputusan & jurisprudence
Indexed
Regulationsstatutes & instruments
Indexed
Legal scholarshiptheory & doctrine
Indexed
Media & social mediasignal beyond the record
Crawled

Each information source is classified, tokenized, and clustered — then connected, so legal insight emerges from the relationships between data, not just the documents themselves.

01 / Sub-products

Four sub-products under Analytic — indexation, analytic, crawler, legal insight.

Sub-product 01 · Indexation

Structured indexation

Raw legal information — court decisions, regulations, scholarship — classified, tokenized, and clustered into structured records. The substrate everything else is built on: messy source documents turned into queryable data.

Classify Tokenize Cluster
Sub-product 02 · Analytic

Correlation & analytic

Find correlation between each case and other cases and regulation. Trends in judge consideration, the recurring material in trials, and verdict patterns within case clusters — surfaced as analytics, not buried in documents.

Trends Patterns Case clusters
Sub-product 03 · Crawler

Data crawler

A crawling engine that gathers information from multiple sources at once. Each crawl extracts content and documents into one place — pulling not only legal information but also signal from media and social media.

Multi-source Extract Consolidate
Sub-product 04 · Legal Insight

Legal insight & research

Legal insight ammunition for professionals to develop the most informative legal opinion for a client. Connections between each piece of legal information, regulation links, and predictive information of a case using data gathering.

Insight Predictive Connection
Capability · Data Extraction

Data extraction

Extract data to make reports and insight from each legal document or other information source. The bridge from the corpus to a deliverable — at the volume that ministry archives actually hold.

Reports At scale Document-level
Capability · Visualization

Advanced visualization

Change the perspective of information with engaging visualization. Advanced visualization using GIS and D3 — maps and interactive charts that give more meaning to the data, displayed interactively for the user.

GIS D3 Interactive
02 / Featured · Graph

Where the real value lives — in the edges between regulations, not the texts themselves.

CITATION GRAPH · UU 12/2022 TPKS UU 12/2022 UUD 1945 UU 8/1981 UU 23/2002 UU 7/1984 PP 30/2025 PERPRES 98/2024 PERPRES 55/2024 PERMEN 8/2024 PERJA 6/2021 CITES · 4 CITED BY · 5 FOCAL NODE
Featured dataset · Graph backbone

Every regulation, traced up to its constitutional basis and down to every implementing instrument.

Indonesian regulation is a directed graph — but until we built it, no one had assembled it. Our citation graph resolves cross-references across the entire corpus: which Acts a regulation derives its authority from, which lower instruments implement it, which provisions have been amended by later law, which have been struck by the Constitutional Court.

For policy analysts and AI builders, this is the substrate that turns "regulatory text" into "regulatory knowledge."

4.2M+Resolved edges
11Regulation tiers
D3Visualisation-ready
DEFINITION DIFF · "USAHA KECIL" Usaha Kecil 18 DEFINITIONAL VARIANTS · 7 ACTIVE UU 9/1995 SUPERSEDED "...memiliki kekayaan bersih paling banyak Rp 200 juta tidak termasuk tanah dan bangunan." UU 20/2008 SUPERSEDED "...memiliki kekayaan bersih lebih dari Rp 50 juta sampai dengan paling banyak Rp 500 juta ..." PP 7/2021 IN FORCE "...memiliki modal usaha lebih dari Rp 1 miliar sampai dengan paling banyak Rp 5 miliar ..." ML DIFF · CHANGE IN MEASURE (KEKAYAAN → MODAL) ML DIFF · 25× INCREASE IN UPPER BOUND → 15M SMEs RECLASSIFIED UNDER PP 7/2021
Featured dataset · Definition library

Eighteen different definitions of "usaha kecil" — reconciled across forty years of legislation.

One of Indonesia's hardest regulatory problems is also one of its quietest: the same term, defined differently in different laws, applied inconsistently across ministries. We built an ML-powered diff engine that surfaces these definitional drifts term by term, version by version.

For Indexa's own drafters, this is how we avoid creating the nineteenth conflicting definition. For our clients, this is how a research team or a compliance system can finally see — at a glance — which definition is currently in force, which is superseded, and what changed.

200K+Definitions extracted
MLAuto-diff engine
APITerm lookup endpoint
CASE-LAW SCHEMA · CORRUPTION 12-variable structure PUTUSAN MA-RI 1234 K/PID.SUS/2024 01 · NOMOR PUTUSAN 1234 K/Pid.Sus/2024 02 · PASAL DITERAPKAN Pasal 2 (1) UU 31/1999 03 · KERUGIAN NEGARA Rp 4.7 miliar 04 · LAMA PIDANA 7 tahun 05 · DENDA Rp 300 juta 06 · UANG PENGGANTI Rp 4.7 miliar 07 · JABATAN TERDAKWA Pejabat eselon III 08 · WILAYAH Jawa Tengah 09 · SEKTOR Pengadaan barang & jasa 10 · TINGKAT PERKARA Kasasi 11 · TANGGAL PUTUSAN 14 Agustus 2024 12 · KAIDAH HUKUM "Unsur memperkaya diri..." → JOINS TO GIS · ANALYTICS · JURISPRUDENCE TREE
Featured · Case analytic

Court decisions, distilled into structured variables.

A court decision is rarely useful in PDF form. We have built an NLP pipeline that reads decisions and writes back structured data: applied articles, sentence length, financial losses, defendant attributes, geographic origin, and the legal principle itself.

For the Attorney General's Office, 50,000 narcotic cases became the baseline data set for the Narcotic Requisitor Calculator — analytics that put structure around requisitor decisions and monitoring. The same pipeline powers Indexa Analytic, and is available to apply to your own research question.

50KNarcotic cases analysed
NLPReads decisions to data
AGORequisitor baseline
Engineering · provenance

How the data actually gets from a government website into a queryable database.

A five-stage pipeline, running continuously. Every regulation in our corpus has a clear chain of custody: source URL, capture timestamp, parser version, classification confidence, and reconciliation status.

→ Stage 01
Crawl

Distributed crawlers monitor multiple government portals and ministry sub-sites, plus media and social-media sources — pulling new instruments and signal into one place.

→ Stage 02
Extract

OCR for scanned gazettes, structured-document parsing for native digital. Layout-aware extraction preserves bab, pasal, ayat hierarchy.

→ Stage 03
Classify

NLP classifier assigns regulation tier, subject domain, issuing authority. Confidence scores expose ambiguous cases for human review.

→ Stage 04
Resolve

Cross-references in "mengingat" and "menetapkan" clauses resolved to graph edges. Definitions extracted from Pasal 1 and diffed against prior versions.

→ Stage 05
Serve

Elasticsearch index, REST API, GraphQL endpoint, and bulk-export JSON/CSV. Versioned snapshots for reproducible research.

03 / Who uses it

Three kinds of people — and what they do with the data.

i.

Ministries & agencies

Internal regulatory inventory, policy harmonisation across line ministries, regulatory impact assessment, and AI/RAG pilots over the agency's own legal corpus. Custom slices and on-premise deployments available.

For: Bappenas · Kemenkumham · BPHN · sector ministries
ii.

Research institutions

Quantitative legal scholarship, comparative law studies, longitudinal sentence-disparity research, regulatory-network analysis. Reproducible snapshots with stable identifiers for citation.

For: Universities · think tanks · CSO research arms
iii.

Legal-tech builders

Foundation data for compliance platforms, contract review tools, case research products, and LLM-based legal copilots. API access with attribution and rate tiers calibrated to your product.

For: Indonesian legal-tech & AI startups
04 / Access

How the platform reaches the people who need it.

Mode 01 · Public

Legal Search Engine

The public-facing search interface — Pencari Informasi Putusan and Pencari Kesamaan Putusan. Search legal information, regulation, and court decisions with one swipe.

  • Web interface
  • Search & browse decisions
  • Find similar rulings
  • Connect regulation to case
Mode 03 · Engagement

Custom data work

Bespoke data extraction, classification pipelines, and thematic analysis built around an institution's research question — as in the AGO narcotic requisitor baseline. Suited to ministries and research partners.

  • Custom data extraction
  • Thematic analysis packs
  • Pipeline co-development
  • Joint research engagements
05 / Coverage

The hierarchy of Indonesian law, tier by tier — the structure we index against.

Tier
Instrument type
Issuing authority
Position
01
Undang-Undang Dasar 1945 & amendments
MPR · DPR
Constitution
02
Ketetapan MPR
MPR
Tier II
03
Undang-Undang & Perpu
DPR · Presiden
Tier III
04
Peraturan Pemerintah (PP)
Presiden
Tier IV
05
Peraturan Presiden (Perpres) & Keppres
Presiden
Tier V
06
Peraturan Menteri (Permen) & Kepmen
Kementerian (38)
Tier VI
07
Peraturan LPNK & Lembaga Negara
LPNK · BI · OJK · MA
Agency
08
Peraturan Daerah (Perda) Provinsi
38 Provinsi
Regional
09
Perda Kabupaten / Kota
514 Kab./Kota
Regional
10
Putusan Mahkamah Agung & lower courts
MA · PT · PN
Case law

The seven-tier hierarchy of Indonesian legislation (UU 12/2011), plus agency regulations, regional Perda, and court decisions — the structure Indexa indexes and integrates against.

Data ethics · provenance

All of this data is public-record material — but how we serve it matters.

Every record carries its source URL and capture timestamp. We do not redact, edit, or paraphrase regulatory or judicial text — what you query is what the gazette published.

For case law, personal identifiers in decisions follow the Supreme Court's own publication policy. Our pipelines respect anonimisasi conventions, and we provide tools for additional redaction where research ethics require it.

Need the data?
Or a custom slice built for your research?

[email protected]
For builders & researchers

Tell us what you're trying to query, and we'll tell you how the platform fits — and what additional analysis might be worth commissioning.

[email protected]

For ministries

Custom data extraction, classification pipelines, and thematic analysis built around your own legal corpus and research question.

[email protected]

Office

Gedung Setiabudi 2, Lantai 2, Suite 207 B-C
Jl. H. R. Rasuna Said Kav. 62
Kuningan, Jakarta

+62 877 2999 2727