On requestReasoning· AI reasoning

Professional Decision Reasoning Traces Dataset

How experienced accountants, supply-chain planners, marketers and engineers think through real work decisions, written out step by step by the professionals themselves.

See pricing · from $790 Get free samples

On request: this dataset is produced to your specification after you order. Typically 3–4 weeks for 200 traces and 6–8 weeks for 1,000.

Example record

{
  "id": "prt-000087",
  "risks": [
    "A promotion or holiday would raise demand; check the marketing calendar before waiting."
  ],
  "domain": "supply-chain",
  "decision": "Do not order today; schedule the order in about 3 days or when stock reaches 900 kg.",
  "language": "en",
  "scenario": "A coffee chain sells about 420 kg of roasted beans per week across 6 stores. The roaster's lead time is 10 days. Current stock is 1,100 kg. Should the planner place an order today?",
  "confidence": "high",
  "assumptions": [
    "Lead time is reliable",
    "No promotion planned in the next two weeks"
  ],
  "author_role": "supply-chain planner, 8 years",
  "reasoning_steps": [
    "Daily demand is about 420 / 7 = 60 kg.",
    "Demand during the 10-day lead time is 60 × 10 = 600 kg.",
    "Sales vary by roughly ±20% week to week, so I keep about 5 days of safety stock: 300 kg.",
    "Reorder point = 600 + 300 = 900 kg.",
    "Stock is 1,100 kg, above the reorder point, which leaves about 3 days before we reach it."
  ]
}

Pricing

Fixed prices for standard packages, produced to your spec. Pick a package and a license to see the exact total, or ask for a custom quote.

Packages compared
Feature	Starter	Pro	Custom
examples	200	1,000	Any quantity
What's included	200 traces in JSONL Domain, role and experience labels Datasheet, schema, license and checksums	1,000 traces in JSONL Everything in Starter Domains and decision types chosen by you Expert quality score per trace	Your quantity, annotations and deadline
Datasheet, license, checksums	Included	Included	Included
Research license	$790$3.95 / example	$2,990$2.99 / example	Quote
Commercial license	$2,490$12.45 / example	$8,900$8.90 / example	Quote
Exclusive license	$7,900$39.50 / example	$26,900$26.90 / example	Quote

How ordering works

1Choose a package and license and place your order. No payment is taken online.
2Within 48 hours our team contacts you to confirm the spec and invoices a 50% deposit.
3We produce and deliver in parts so you can review early. Typically 3–4 weeks for 200 traces and 6–8 weeks for 1,000.
4Pay the balance and download the final dataset.

Custom quote

Need another quantity, an Enterprise license or annotations for this dataset? Available here:

Domain and skill labels
Expert quality score per trace
Paired strong/weak traces
Your own domains and scenarios

Request a custom quote

1. Choose a package

Starter · 200 examples200 traces, reviewed by a second expertPro · 1,000 examplesLarger set with a custom topic mix agreed with you

2. Choose a license

ResearchAcademic and personal, non-commercial use$790CommercialTrain and ship commercial models$2,490ExclusiveThe data produced for you is never licensed to anyone else$7,900

200 traces in JSONL
Domain, role and experience labels
Datasheet, schema, license and checksums
Train, fine-tune and evaluate models
Deploy trained models commercially
Keep your models if the license ends

Price per example: $12.45
Deposit, invoiced after you order (50%): $1,245
Balance on final delivery: $1,245
Delivery: Typically 3–4 weeks for 200 traces and 6–8 weeks for 1,000
Total: $2,490

Order now · $2,490

No payment now: place your order and within 48 hours our team confirms the spec and invoices the deposit. Fully refundable if we cannot deliver the agreed spec.

Overview

AI assistants are moving from answering questions to doing work: planning stock, reviewing a budget, choosing a supplier. What they lack is examples of how experienced people actually reason through these decisions, including the assumptions they make and the risks they weigh.

Each record is a realistic work scenario from operations, finance, marketing, HR or engineering, followed by the step-by-step reasoning of a professional with years of experience in that role, their decision, the assumptions behind it and what would make them change their mind.

Example records

Illustrative records in the delivered format. Request the sample pack for real records written by our experts.

Record 1 of 1 · illustrative

{
  "id": "prt-000087",
  "risks": [
    "A promotion or holiday would raise demand; check the marketing calendar before waiting."
  ],
  "domain": "supply-chain",
  "decision": "Do not order today; schedule the order in about 3 days or when stock reaches 900 kg.",
  "language": "en",
  "scenario": "A coffee chain sells about 420 kg of roasted beans per week across 6 stores. The roaster's lead time is 10 days. Current stock is 1,100 kg. Should the planner place an order today?",
  "confidence": "high",
  "assumptions": [
    "Lead time is reliable",
    "No promotion planned in the next two weeks"
  ],
  "author_role": "supply-chain planner, 8 years",
  "reasoning_steps": [
    "Daily demand is about 420 / 7 = 60 kg.",
    "Demand during the 10-day lead time is 60 × 10 = 600 kg.",
    "Sales vary by roughly ±20% week to week, so I keep about 5 days of safety stock: 300 kg.",
    "Reorder point = 600 + 300 = 900 kg.",
    "Stock is 1,100 kg, above the reorder point, which leaves about 3 days before we reach it."
  ]
}

Get a free sample pack

We email you real examples from this dataset.

Technical specifications

Authors: Professionals with at least 5 years in the role (role and years recorded per record)
Domains: Operations and supply chain, finance and accounting, marketing, HR, software and hardware engineering
Language: English or Vietnamese
Fields: Scenario, role, reasoning steps, decision, assumptions, risks, confidence
Review: A second professional in the same field reviews each trace
Format: JSONL (one record per line), UTF-8; CSV on request
AI use: Written by people; screened for LLM-generated text. Any AI assistance is disclosed in the datasheet
License: Research, Commercial, Enterprise or Exclusive

Use cases

Agent and copilot training
Teach work assistants to reason like a domain professional before acting, not just produce a plausible answer.
Reasoning evaluation
Compare your model's reasoning with an expert's on the same scenario, step by step.
Preference data
Order paired traces (strong vs weak reasoning, labelled by an expert) for preference optimisation.
Enterprise knowledge capture
Commission traces in your own domain to encode how your best people make decisions.

How this data is made

1Problems sourced or written by vetted domain experts
2Each solution written step by step by a person, never generated
3A second expert checks every step and the final answer
4Disagreements resolved by a lead reviewer
5Screened for copied text and LLM-generated writing
6Difficulty, topic and skill labels added
7Exported as JSONL with a schema you can rely on
8Datasheet and license packaged with the delivery

Read our full process

Provenance & legal

100% made by people: no scraping, no generative AI
Datasheet documenting how the data was made, checked and its limitations
Commercial license that lets you keep models trained on the data
Supports training-data documentation under the EU AI Act

Licensing options

Dataset-specific notes

Scenarios are realistic but invented or anonymised: no client names, figures or confidential data from real employers.
Authors sign a statement that the reasoning is their own and contains no confidential information.
Each trace records the author's role and years of experience, never their name.

Delivery format

One JSON record per line, with the fields shown in the example records above and a SCHEMA.md describing each one.

Each delivery contains:

professional-reasoning-traces-v1/
├── data/train.jsonl
├── data/sample.jsonl
├── SCHEMA.md
├── DATASHEET.md
├── LICENSE.pdf
└── checksums.sha256

Frequently asked questions

Who writes the traces?

Working professionals we recruit and vet for each domain, for example supply-chain planners or chartered accountants. Their role and years of experience are recorded with each record.

Can we choose the domains and decisions?

Yes. Most buyers send a list of decision types their product supports and we write scenarios around them.

Are the scenarios based on real companies?

They are realistic but invented or fully anonymised. We do not use any employer's confidential information.

How does ordering and payment work?

Create a free account and place your order on this page; nothing is charged online. Our team contacts you to confirm the spec and invoices a 50% deposit, with the balance due on final delivery. The deposit is refunded in full if we cannot deliver the agreed spec.

What is the difference between non-exclusive and Exclusive?

With Research or Commercial we may license the examples produced for your order to other buyers later. With Exclusive they are never licensed to anyone else or added to our catalog.

Can I order a different quantity or extra annotations?

Yes. Request a custom quote with the quantity, annotations and deadline you need. Prices per record fall as quantity grows.

Need a variation?

Different topics, languages, difficulty, quantity or labels? Describe it and we reply within 48 hours with matching samples and a quote.

Request a quote

Browse all datasets