RFP responses and proposals with win/loss outcomes
Archives of requests for proposal (RFPs) and the proposals a company submitted, with drafts, review comments, pricing structures and whether each bid was won. Labs use them to train agents that draft long, compliant business documents against a customer’s requirements.
What a record contains
One bid: the buyer’s RFP and attachments, the company’s answers and supporting documents, internal drafts and review comments, and the result.
| Field | What it holds |
|---|---|
bid_id | Pseudonymous ID |
buyer_sector | Industry of the issuing organization |
questions[] | Requirement, answer and reviewer notes |
drafts[] | Versions before submission |
pricing_structure | Banded or removed |
result | Won, lost or withdrawn |
{
"bid_id": "B-1408",
"buyer_sector": "regional healthcare network",
"questions": [
{"id": "4.3", "requirement": "Describe your data retention controls.",
"answer": "Customer data is retained for the contract term plus 30 days...",
"reviewer_note": "Cite the retention policy section, not the summary."}
],
"result": "won"
}How AI labs use it
- Long-document drafting
- Draft against detailed requirements, section by section.
- Answer-library retrieval
- Reuse approved answers correctly.
- Evaluation
- Compare a model’s response with the winning submission.
Typical preparation requirements
Agreed with the supplier before any work begins. Typical requirements include:
- Buyer names and confidential RFP terms reviewed; some RFPs restrict disclosure
- Pricing banded or removed
- Third-party documents such as certificates and references excluded
Every dataset has a documented owner and confirmed licensing rights. See data governance on sourcex.si.
What makes a strong package
- Win or loss recorded for each bid
- Answer-library history with versions
- Several years across sectors
Who typically holds it
- B2B software companies
- Consulting firms
- Government contractors
- Construction and engineering firms
Know a company like this?
Introduce the company to SourceX. If its data deal closes, you can earn up to $100,000 in referral fees, paid after the buyer accepts the data and SourceX receives payment.
Refer a companyQuestions
Can public-sector bids be included?
Often the RFP itself is public, but the company’s response can contain confidential details, so it gets the same review as any other bid.
Need this data for a model?
Describe what you need: domain, volume, history, format and licensing terms. SourceX looks for companies that hold it and can license it.