Private beta · September 2026

Every number Indonesia publishes. One API.

Ganita normalizes Indonesia's official statistics — BPS, Bank Indonesia, OJK — into clean, versioned time series with stable slugs. National macro, all 38 provinces, all 514 cities & regencies. REST + MCP, English-first docs, point-in-time history.

Free tier at launch · Waitlist gets the first API keys · One email at beta, no spam.

GET /v1/series/cpi.yoy
$ curl https://api.ganita.example/v1/series/cpi.yoy \
    -H "Authorization: Bearer ganita_····"

{
  "slug": "cpi.yoy",
  "name": "Inflation, consumer prices (YoY)",
  "unit": "%",
  "frequency": "monthly",
  "source": "BPS",
  "observations": [
    { "period": "2026-04", "value": 1.92 },
    { "period": "2026-05", "value": 1.87 },
    { "period": "2026-06", "value": 1.84 }
  ],
  "vintage": "latest"  // point-in-time on request
}
Illustrative response — endpoints go live in private beta.
Inflation · YoY
1.84%
BI Rate
4.75%
GDP Growth · YoY
4.87%
USD / IDR
16,240

Illustrative values as of the June 2026 release cycle. Live series ship with the beta.

549
BPS statistical domains, one gateway
38+514
provinces + cities & regencies covered
200+
hand-curated series with stable slugs
2
interfaces: REST for apps, MCP for AI agents

Coverage targets at general availability. Private beta starts with the 50 highest-demand national series.

Why Ganita

The data is public. Using it is the hard part.

Indonesia publishes world-class statistics — scattered across arcane APIs, monthly Excel drops, and 549 regional portals. Ganita does the janitorial work once, for everyone.

⚙

Stop parsing spreadsheets

Bank Indonesia ships monthly Excel zips. BPS's API speaks in opaque variable IDs and coded periods. We map it all so you just call cpi.yoy — ISO dates, real units, consistent JSON.

◫

Depth nobody else serves

Global vendors stop at national headlines — behind $10k+/yr enterprise seats. Ganita goes to province and city level: GRDP, poverty, HDI, wages and population for every one of Indonesia's 552 regions.

✳

Built for AI agents

A first-class MCP server, not an afterthought. One tool call returns a country snapshot or a full city profile — with source and release date on every number, so your agent can cite what it claims.

Coverage

National depth. Regional granularity.

Three tiers, one consistent series model. If BPS, BI, or OJK publishes it, it gets a slug.

TierWhat you getExample slugsCadence
National ×1 CPI & inflation, GDP, BI rate, FX & reserves, money supply, trade balance, unemployment, poverty, banking sector aggregates cpi.yoy gdp.real.qoq rate.bi trade.balance Daily–quarterly
Province ×38 GRDP by sector, provincial inflation (150 CPI cities), poverty, unemployment, minimum wage (UMP), HDI, population grdp.p32 poverty.rate.p32 wage.min.p31 Monthly–annual
City / Regency ×514 GRDP, poverty, HDI, population, labor stats, minimum wage (UMK) — plus the regional yearbook long tail grdp.r3273 hdi.r3273 pop.r3173 Annual

Developers & agents

One call, any region in Indonesia.

REST for your apps and dashboards. MCP for Claude, agents, and LLM workflows. Same registry, same guarantees.

MCP · ganita_region_profile
> ganita_region_profile({ "region": "Kota Bandung" })

{
  "region": "Kota Bandung",
  "codes": { "bps": "3273", "kemendagri": "32.73" },
  "population": 2530000,
  "grdp_idr_trillion": 348.2,
  "poverty_rate_pct": 3.9,
  "hdi": 82.5,
  "unemployment_pct": 8.1,
  "min_wage_idr": 4482914,
  "sources": ["BPS Kota Bandung", "BPS Jabar"],
  "as_of": "2025 release"
}
Illustrative response. Works for any of Indonesia's 552 provinces, cities & regencies.

Stable slugs

Human-readable series IDs that never change — not variable-ID soup. Rename-proof across source reshuffles.

Point-in-time vintages

Ask what a number said before it was revised. Backtest-safe history, the feature EM quants actually pay for.

Release webhooks

release.published and series.revised events the moment BPS or BI drops a release. No polling.

Region ID spine

Canonical BPS ↔ Kemendagri code mapping for every region — the join table all of Indonesian data joins on. Free tier.

Provenance on every number

Source publication, release date, and revision status per observation. Your agents cite; your auditors relax.

Public freshness SLOs

Per-series freshness on a public status page. If a pipeline is late, you see it before you have to ask.

Pricing preview

Self-serve. Between free-but-raw and $10k enterprise seats.

Priced in USD, billable in IDR. Locked at general availability.

Free
$0
  • Headline indicators (1-release delay)
  • Region reference & ID spine
  • 1,000 calls / month
Developer
$29/mo
  • Full national catalog, current data
  • MCP server access
  • 50,000 calls / month
Enterprise
$500+/mo
  • Point-in-time vintages
  • SLA & bulk export
  • IDR or USD invoicing

Preview pricing — subject to refinement at GA. Attribution to statistical sources passes through to your use under our terms.

For global teams

USD · English docs · agent-ready
  • Indonesia depth no global vendor carries — down to 514 cities
  • Point-in-time history for backtests and models
  • MCP tools your research agents can call and cite
  • One self-serve subscription instead of an enterprise sales cycle

For Indonesian teams

IDR billing · Bahasa support · local rails
  • Stop re-parsing BPS JSON and BI spreadsheets every release
  • Regional data for expansion, credit, and market sizing
  • Webhooks the moment a release drops — no polling scripts
  • Free tier generous enough for dashboards and media graphics

FAQ

Fair questions.

Where does the data come from — and is this legal?

Every series is sourced from official public releases: BPS (Statistics Indonesia), Bank Indonesia, and OJK publications. BPS's terms of use explicitly permit worldwide commercial use and redistribution with attribution, which we provide and pass through. Ganita is an independent product — not affiliated with, or endorsed by, BPS, Bank Indonesia, or OJK.

When can I get access?

Private beta opens September 2026 with the 50 highest-demand national series, REST and MCP. The waitlist gets the first API keys, in order. Regional (province + city) coverage follows at public launch.

Is the data real-time?

No — and nobody's is. Official statistics follow official release calendars (monthly CPI, quarterly GDP, annual regional data). What we guarantee is freshness against those calendars, published as per-series SLOs, plus webhooks the moment a release lands and vintage tracking when numbers get revised.

How does this relate to Tarutha?

Ganita is built by the same team, on the same platform. Tarutha covers Indonesian capital markets — IDX filings, KSEI, prices, news. Ganita covers the macro and regional statistics layer. They're designed to be used together, and a unified key across both is on the roadmap.

The numbers are already public.
Now they're usable.

Join the waitlist — first keys ship September 2026.

Join the waitlist