Buy datasets & bulk dumps
Yes — you can buy our datasets. You can pull them live through an API key, or take a one-shot dump (a dated file package) for your warehouse, agents, or offline jobs.
This page explains, in plain English, what we sell, what is in a dump, what you may and may not do, and how to buy. List prices live on /pricing/. Card plans for finance start on /pay/.
What “a dataset” means here
Each idlidu track (finance, saas, pkgs, robots, and so on) is one dataset: a capped universe of public sources we archive on a schedule, plus the typed facts we extract from those snapshots.
When you buy a dataset, you are buying our product layer, for example:
- Dated extracts (plans, prices, owners, schemas — whatever that track publishes)
- Change events (what changed, when, with a content hash)
- Citeable indices and series (methodology version + sample or full history by plan)
- On higher SKUs: pointers or copies of raw archive objects (HTML, filings, and so on)
- Always: a methodology version and checksums so you can prove what you got
You are not buying the open web, a credit rating, legal advice, or a right to republish our corpus as your own product.
Two ways to buy
- Live API (monthly plan) — keys for HTTP and MCP. Call
/v1as often as your plan allows. Archive and above include rate-limited API export. Scale adds quarterly snapshots; Institution adds a rolling sync window. See /pricing/. - Bulk dump (one-shot or annual) — a versioned package (Parquet, JSONL, or ZIP as published per track) with a time-boxed signed download. Useful when you want the whole universe offline, not one API call at a time.
Bulk SKUs (per dataset)
USD list prices from the shared ladder. Soft-launch: delivery is quoted and scheduled — not a self-serve download button yet.
| What you buy | USD | What you get |
|---|---|---|
| Universe snapshot | 2,500 one-shot | Current v1 extracts + methodology version + checksums · ~30-day signed download |
| Raw + extract deep dump | 7,500 one-shot | Snapshot plus raw object pointers/hashes and extracts · 30 days of Archive-tier API for refresh pulls |
| Annual bulk refresh | 4,500 / year | Four quarterly universe snapshots |
| Scale dump addon | +500 / mo on Scale | Monthly snapshot instead of quarterly |
| Institution sync | Included in Institution ($3,999/mo) | Rolling export window (for example 7-day signed prefixes) · invoice / Net-30 |
| Rule / recipe pack (Shape C) | 5,000 one-shot | Versioned rules or recipes + citations index — not unlimited recompute |
What is usually inside a dump
- Manifest — track id, schema versions, as-of time, file list, checksums
- Extracts — the typed JSON we publish for each entity in the seed
- Events — change history for the covered window (plan-dependent)
- Indices / series — published points with methodology version
- Deep dump only — raw archive references (and sometimes bytes) so you can re-check our extract against the snapshot we stored
Formats follow the track (Parquet / JSONL / ZIP). We do not promise every track has the same column layout — each track has its own schema and methodology page.
License — what you may and may not do
Bulk dumps are licensed for your internal use (including your agents and internal tools). That is the same spirit as our Acceptable Use Policy for API access.
You may:
- Load the dump into your warehouse, lake, or eval harness
- Train or evaluate models on the licensed pack, if your contract says so
- Cite methodology and as-of times in your own systems
- Combine with your private data under your own policies
You may not:
- Resell, republish, or host our dump as a public corpus or competing API
- Strip methodology / claim the numbers as an official rating or endorsement
- Treat a dump as permission to ignore upstream site Terms for your own crawl
- Ask us for unlimited “scrape anything” as part of a dump SKU
Honest soft-launch notes
- Density varies by track. Some universes are dense; some are still seeding. Ask what as-of coverage you will get before you pay for a deep dump.
- Sources are gated. Each track has a SOURCES / ToS posture. A dump only includes what that track is allowed to archive.
- Raw HTML is hotter than extracts. Deep dumps with raw objects need a clearer rights conversation than extract-only snapshots.
- Delivery is quote-backed today. Card checkout on /pay/ covers listed finance plans and custom deposits first. Bulk one-shots ≥ ~$2,500 are usually invoice / Net-30 after email scope.
How to buy
- Pick the track or pack on /products/ (or a specialty pack on /pricing/).
- Choose live vs dump — monthly API, one-shot snapshot, deep dump, or Institution sync.
- Mail a short scope to [email protected] (template below). For listed finance plans you can also start on /pay/.
- We confirm coverage and price — USD quote; invoice / Net-30 when the SKU is ≥ ~$2,500. Soft-launch replies are multi-day, founder-run — not a ticket desk.
- Receive the package — signed download + checksums, and/or API keys for the refresh window named in the SKU.
Email intake (copy / paste)
Subject: idlidu dataset / bulk dump Org / contact: Dataset(s) you want (track name or pack): SKU of interest (snapshot | deep dump | annual | Institution sync | other): Use case (agents / internal analytics / model training — 1–3 sentences): Preferred format (Parquet | JSONL | ZIP): Need raw HTML/PDF blobs, or extracts + events only? Delivery deadline: Budget range (USD) or cap:
If the dataset does not exist yet
That is a custom collection & processing job — scoped ingest and pipelines into a private dump or feed — not a consulting memo. See /custom/. When the work fits our factory, pieces may later become a list-price track; your private grant stays as quoted.
What we do not sell
- Advisory research decks or “insights” retainers
- Opaque risk, credit, or malware scores
- Unlimited crawl rights or paywall bypass
- A support desk or Slack onboarding SLA
Next steps
Ask about a dump Full pricing Pay listed plans How idlidu works