HVAC and refrigeration service tickets with technician findings, 23 firms

Summary

Completed service tickets from 23 US HVAC and commercial refrigeration contractors, each with the dispatcher's intake note, the technician's findings, the work done and any follow-up proposal. Techs write in shorthand with pressures, temperatures and part sizes, which makes the notes hard and useful material for extraction and diagnosis models. Customer names, site addresses, phone numbers and on-site contacts are replaced with placeholders.

Collected to order. We collect it from US companies for your order. Scope, volume and timing are confirmed in writing before you sign.

Pricing on request

Request scope and pricing

At a glance

Planned scope. Volume, sources and timing are confirmed in writing before you sign.

Records: 4.6Mservice jobs, emails and service agreements
Size: 12.2 GB
Tokens, estimated: 1.3B
Years: 2018 to 2026
Firms: 23

Owner profile: 23 US HVAC and refrigeration service contractors, 8 to 95 staff each, mostly commercial work, Southeast and Southwest
Industry: Field services
Data type: Field operations
Headcount: 8 to 95 staff per firm
Region: United States (Southeast, Southwest)
Languages: English, Spanish
Formats: JSONL, Parquet
Updated: 2026-10-03

Sources

System	Kind	Records
ServiceTitan jobs and technician notes	Field operations	1.62M jobs
Google Workspace mail	Email	2.9M messages
ServiceTitan service agreements	Contracts	41K agreements

Good for

training dispatch agents to set priority and required skill from the caller's description
fine-tuning diagnosis models on symptom, reading and repair sequences
extracting equipment, parts and refrigerant data from technician shorthand
evaluating summaries of technician notes written for the customer

Fields

Field	Type	Description
job_id	string	Unique job id within the pool
firm_code	string	Placeholder for the contractor
job_type	enum	service_call, pm_visit, install, warranty or callback
priority	enum	emergency, same_day, standard or scheduled
created_at	timestamp	When the call was booked
completed_at	timestamp	When the technician closed the job
site_type	enum	restaurant, grocery, office, retail, warehouse, school, residence and others
intake_note	string	Dispatcher or CSR note taken at booking
tech_findings	string	Technician diagnosis with readings as written
work_performed	string	What was done on site
parts	array<string>	Parts used, generic descriptions with sizes
follow_up	string	Recommendations and proposals left with the customer

Example record

{
  "job_id": "HV23-JOB-2210547",
  "firm_code": "[COMPANY_14]",
  "job_type": "service_call",
  "priority": "emergency",
  "created_at": "2024-07-19T06:12:00-05:00",
  "completed_at": "2024-07-19T10:48:00-05:00",
  "site_type": "restaurant",
  "intake_note": "Walk-in cooler at 52F, staff moving product to reach-ins. Mgr [PERSON_88] on site till 2.",
  "tech_findings": "Cond fan motor seized on walk-in CU, comp tripping on high head. R-404A, suction 38 psig, head 410 psig before trip. Cond coil packed with grease and lint.",
  "work_performed": "Replaced cond fan motor and blade. Chem cleaned cond coil. Box pulled down to 38F in 70 min. Left running.",
  "parts": [
    "Condenser fan motor 1/3 hp 208-230V 1075 rpm",
    "Fan blade 18 in 27 deg CW"
  ],
  "follow_up": "Recommend PM agreement with quarterly coil cleaning. Bottom door gasket torn, gasket proposal sent to [EMAIL_88]."
}

One record, de-identified. Placeholders such as [PERSON_12] and [COMPANY_4] stay the same across the whole dataset, so threads still read correctly.

Licenses available

Exclusive
One buyer. Nobody else can license it during the exclusive term, usually 2 years.
Non-exclusive
Licensed to more than one buyer.
First look
One buyer gets 2 to 4 weeks to make an offer before anyone else sees the dataset.

What ships with it

Datasheet: what it is, what it is good for, verified size, sources, years, format and limits.
Provenance file: the owner shown by code, staff notice and consent status, exclusions, mapped to the EU AI Act Article 53 training-data summary and California AB 2013.
Manifest with a hash for every file.
QC report and de-identification report.
Your own copy, with unique canary records.

How it is de-identified

Default exclusions removed, files flattened to text, identifiers found in three layers and replaced with consistent placeholders, and the result measured against a 99% bar for each identifier type in each source.

The full method