Dataset

Open Synthetic Schema Catalog for Operational Business Data

Sukhrobjon (Rob) Ismoilov
Sukhrobjon (Rob) Ismoilov
Founder & Principal, Main Street Wealth

What does a clean, de-identified operating record actually look like? Four synthetic 100-row samples — HVAC dispatch-to-invoice, plumbing camera inspection, freight load-to-carrier-invoice and legal matter-to-billing — with a field dictionary showing what each column replaces in real data.

Last Updated November 16, 2026
4 min read
1 cited sources
Share:
Quick facts
Schemas4
Rows per sample100
Real data usedNone — fully synthetic
LicenseCC BY 4.0

Section 1

The schemas

SchemaVerticalRecord chainFields
HVAC Dispatch-to-Invoice Chain (Synthetic)Residential HVACOne row per inbound service call, carried through booking, dispatch, diagnosis, the options presented, what was sold, the invoice and payment. The linked chain — not any single field — is what makes this shape valuable.17
Commercial Plumbing Camera Inspection & Estimate (Synthetic)PlumbingOne row per sewer-camera inspection, linking pipe material and age, the coded defect found and where, severity, the recommendation and whether the estimate was accepted.14
Freight Brokerage Load-to-Carrier-Invoice (Synthetic)Freight brokerageOne row per brokered load: lane, equipment, quoted shipper rate, carrier rate and margin, appointment performance, detention and accessorials, proof of delivery and the carrier’s invoice and payment.15
Legal Matter-to-Billing Narrative (Synthetic)Law firmOne row per time entry: matter type, timekeeper role, task category, hours, rate, a templated narrative with no names or facts, write-downs and collection. Real billing narratives are privileged or confidential; this shows the shape only.10
Every row is generated from a seeded random-number generator. No row is derived from any real company’s records, and the value distributions are invented for illustration — they are not market statistics.

Section 2

Why publish synthetic schemas

Owners preparing a file and buyers’ procurement teams both need a shared picture of the target: which fields survive de-identification, what replaces names and addresses, and which links between records have to be preserved. Buyers say the links are the value — “the connection, the diversity, the traceability of all of these data points.”[1] Each schema here keeps the chain intact while every direct identifier is a token or a generalisation.

Use them to map your own export, to brief a developer writing a de-identification script, or as a test fixture for the PII scanner.

Frequently asked questions

Is any of this real customer data?
No. Every row is generated from a seeded random-number generator; no row is derived from any real company’s records, and the distributions are invented for illustration.
Can I use these samples?
Yes, under CC BY 4.0 with attribution to Main Street Wealth. They are suitable as fixtures and examples, not as statistics.

References

Sourced from primary industry reports, government data, SEC filings, and peer-reviewed valuation research. Retrieved at time of publication.

  1. 1.Business Insider — Micro1 is making a $1 billion bet on AI’s latest data craze — Seller interviews, the range micro1 says it pays, and a Gartner estimate of AI data spending, 26 October 2026 (2026)

Want a straight read on what your records are worth?

Send us the systems, years and rough record counts. We tell you what is licensable, get competing written quotes, and you pay nothing if buyers pass.

Request the evaluation form
Calendar

This article is provided for general educational purposes and is not legal, tax, investment, or accounting advice. Multiples, fee ranges, and timelines are illustrative ranges compiled from third-party sources current as of November 16, 2026 and will vary by transaction and market conditions. Consult qualified counsel before acting on any information contained here.