When referring a business in the U.S., Canada, the EU, the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE or Saudi Arabia (outside the U.S., Canada and the EU, at least 70% of the data must be in English) to SourceX, you don't need to delve into their confidential data. Your role is to identify companies in the U.S., Canada, the EU, the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia that are likely to possess valuable, well-structured proprietary operational data suitable for AI developer programs. A key indicator for this potential is their "documentation maturity" – how well they create, maintain, and utilize internal records. This guide will help you spot the right characteristics.
What is Documentation Maturity?
Documentation maturity refers to the extent and quality of a company's internal records, standard operating procedures (SOPs), knowledge bases, and other materials that detail their operations, processes, and historical activities. For SourceX, mature documentation often correlates with the availability of rich, structured, and valuable data that can be licensed for AI training and research.
Why it Matters for Referrals
companies in the U.S., Canada, the EU, the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia with strong documentation practices are more likely to:
- Possess rich, structured data: Well-documented processes often generate structured data logs and records that are highly valuable.
- Understand their own data assets: They can more easily identify and articulate what data they have and its potential utility.
- Have clear data rights: Companies that create their own operational documentation often have clear ownership of the underlying data.
- Simplify data preparation: Organized internal records can streamline the process for SourceX to evaluate and prepare data for licensing, should the company choose to proceed.
Indicators to Look For (Without Accessing Data)
You can assess documentation maturity through conversations and observation, without requesting access to any proprietary information:
- Internal Knowledge Bases: Does the company maintain a wiki, intranet, or internal portal for sharing operational knowledge, best practices, or project histories?
- Standard Operating Procedures (SOPs): Do they have formal, written procedures for common tasks, workflows, or client interactions? Illustrative: A company with detailed SOPs for its customer support department often generates structured support ticket data, call logs, and resolution paths.
- Training Materials: Do they have comprehensive training modules for new employees, detailing how to perform specific tasks or use internal systems?
- Project Management Systems: Beyond just task lists, do their project management tools capture detailed project histories, decisions made, lessons learned, and outcomes?
- Quality Assurance (QA) Records: For product or service-oriented businesses, do they log QA findings, test results, and defect resolution processes?
- Change Management Processes: How do they document changes to their products, services, or internal systems? Is there a formal record of updates and their rationale?
- Process Improvement Initiatives: Companies that actively engage in process improvement often have excellent documentation of their current and past states, and the impact of changes.
What's Usually Not a Fit
Conversely, certain situations suggest a lack of documentation maturity or data that would not be suitable for licensing with SourceX:
- Reliance on Individual Knowledge: If critical operational knowledge resides solely in the heads of a few long-term employees, it's a red flag.
- Lack of Standardization: Highly informal operations with no consistent way of performing or recording tasks.
- Focus Solely on Client Data: Companies whose primary data asset is client-specific information they do not own or cannot license broadly.
- No Internal Systems: Businesses that rely entirely on external, off-the-shelf software with no unique internal data generation or documentation.
Asking the Right Questions
When speaking with a potential referral, you might ask questions like:
- "How do you onboard new employees and get them up to speed on your internal processes?"
- "Where do your teams go to find information about how to do their jobs or solve common problems?"
- "What systems do you use to track project histories and outcomes?"
These questions can give you insight into their documentation practices without needing to see the actual content. Remember, the company will decide what data scope it is willing to provide and agrees commercial terms, not you.
For a preliminary, non-binding screening of a company's general fit, you can direct them to our company fit checker. Understanding these indicators can significantly improve the quality of your referrals and their potential to qualify for SourceX's current data needs.
Explore the referral process steps or learn more about who qualifies.
Rewards are calculated as the greater of $50,000 plus 5% and 15% of the compensation SourceX actually pays the company you refer, with no cap. A lead, meeting or signed agreement alone does not trigger payment. See the program terms.
A quiet mature business may deserve the first introduction because the fit tests reward accumulated history, not momentum. SourceX screens for 50+ full-time employees at peak, years of operating records the company created itself, rights to license them and a CEO who wants the conversation. A 20-year operator whose headcount has been flat, or has shrunk from a higher peak, often passes more of those tests today than a celebrated three-year-old scale-up. Use growth to find companies; use documentation maturity to choose whom to ask first.
These sections turn the documentation indicators above into one decision for venture platform teams and other network owners: which company to ask first. Program facts come from SourceX's program terms and who qualifies page, checked 2026-10-26. The companies are Illustrative, and the screen and ranking are this page's proposed method, not SourceX data.
Why might a quiet mature business deserve the first introduction?
Because every test it must pass looks backward. Growth measures how much a company changed over a few years; the fit facts measure what it has already built and kept.
| Statement | Type | Basis |
|---|---|---|
| Introduced companies need 50 or more full-time employees at peak, contractors excluded | Program fact | Terms, section 1 |
| Good candidates have several years in operation, documentation they created, rights to license and an authorized sponsor; under the program the CEO authorizes | Program fact | Who qualifies page and program rules |
| Whether a company's records match a current data need | Conditional | Checked only in SourceX's review; criteria differ by program |
| In most networks, the most-discussed companies are not the ones with the deepest licensable history | Thesis | This page's hypothesis, tested by the screen below |
What a growth signal measures, and what it misses
A growth signal is a rate of change. The size test is a level, read at the company's peak. They answer different questions, and a network that ranks by the first will misjudge the second.
The standard definition makes the gap concrete. A Bureau of Labor Statistics Monthly Labor Review article describes the OECD's high-growth firm as one with 10 or more employees whose growth averages more than 20 percent a year over three years, measured by employment or turnover. A firm at that 10-employee floor growing exactly 20 percent a year ends the three years with about 17 employees: high-growth by definition, and far short of 50 full-time employees at peak.
Survival points the other way. The BLS reports that 34.7 percent of business establishments born in 2013 were still operating in 2023. A company that has kept going for a decade has passed a filter most never pass, and each of those years can leave records behind. The figure counts establishments from one cohort, not firms, and says nothing about record quality, so treat it as a base rate, not a score.
| Term | How this page uses it |
|---|---|
| Growth signal | A change measure: headcount growth, a new funding round, hiring posts or press coverage |
| Attention | How often a company comes up in your network; never a fit fact |
| Peak headcount | The highest number of full-time employees the company has had, contractors excluded |
| Continuous records | Operating records kept in systems the company runs, with no gap where history was dropped |
| Quiet mature business | A company with a long operating history and little attention, whatever its current growth |
| Capacity | Whether the CEO and the system owner can give an inventory a few hours next quarter |
The maturity-versus-hype screening comparison
Run both companies through the same seven questions, answered from conversation and public facts, never from files. Illustrative: two generic companies with invented facts.
The celebrated scale-up. A four-year-old vertical software company, much discussed after its growth round. It grew from 20 to 48 full-time employees in two years, re-platformed its product last year and keeps most operating data inside customer tenants. The CEO's next two quarters belong to the round's hiring plan.
The quiet operator. A 23-year-old commercial HVAC service contractor, still run by its founder as CEO. Full-time headcount peaked at 210 in 2019 and is 160 today. It has kept work orders, technician notes and quotes in one field-service system since 2011, with a written procedure for each job type. Nobody in your network has mentioned it in years; one of your LPs sits on its advisory board.
| Question | Celebrated scale-up | Quiet operator | How to answer without records |
|---|---|---|---|
| 1. 50 or more full-time employees at peak? | No: 48 at peak | Yes: 210 at peak | Ask peak headcount, contractors excluded |
| 2. Years of continuous records in a system it runs? | Unclear: four years, split by a re-platform | Yes: 15 years in one system | Ask which system, and since when |
| 3. Records the company created itself? | Mostly no: inside customer tenants | Yes: its own work orders and notes | Ask whose records they are |
| 4. Written procedures, QA and change records? | Thin: practices still forming | Strong: one procedure per job type | Ask how a new hire learns the work |
| 5. Any contract that limits reuse? | Unknown | Unknown | Note it for SourceX's rights review |
| 6. A relationship owner who can ask a CEO open to the idea? | Unclear: the CEO is absorbed by hiring | Yes: the LP can ask the founder-CEO | Ask the relationship owner first |
| 7. Capacity next quarter? | Low | Medium: winter is the busy season | Ask about the CEO's calendar, not yours |
On the six fit and access questions, the quiet operator passes five and the scale-up passes none, failing two outright; capacity, the seventh, only sets the timing. Rights stay unknown for both until SourceX reviews them. A celebrated name does not cure a failed size test, and a quiet one does not create rights.
Shrinking is not disqualifying. The size test reads the peak, so the contractor's fall from 210 to 160 does not matter for size. What a shrinking company should prompt is a different question: did record-keeping shrink too? Ask whether the system and its history survived the cuts.
A quiet company that still fails
A 31-year-old document-processing bureau with 180 full-time employees is as quiet and mature as companies get, and it still fails. Most of what it holds is its clients' documents, and material that belongs to clients rather than the company is usually not a fit. Its own job tickets would have been the asset, but it dropped nine years of history when it left its old system in 2016. Two questions settled it: whose records are these, and when does your history start? Age answered neither.
Rank the first conversations with ownership and capacity included
Sort on observable keys, in this order, and let attention break ties only:
- Peak size and continuous records both known and passing.
- Records the company created itself.
- A relationship owner who has agreed to ask the CEO.
- Capacity, or a dated event such as a documented system migration that makes now the right time.
| Rank | Company (Illustrative) | Keys met | Why it sits here |
|---|---|---|---|
| 1 | The 23-year-old HVAC contractor | 1, 2, 3 and 4 | The LP will ask the founder-CEO; procedures are written |
| 2 | A 70-person freight audit firm, flat for eight years, leaving its legacy system in March | 1, 2 and 4 | The migration date is real timing; nobody has asked the relationship owner yet |
| 3 | A 12-year-old staffing firm with 85 internal full-time staff and 900 placed workers | 1 and 2 | Only the 85 count toward size; describe both numbers and introduce it if its records are strong |
| 4 | The celebrated scale-up | None yet | Revisit when full-time headcount passes 50 and it controls its own records |
| 5 | The document-processing bureau | Fails key 2 | Client-owned material and lost history |
Method note. Questions 1 to 5 restate SourceX's published screening profile and terms (checked 2026-10-26); questions 6 and 7 are operating questions that decide whether an ask can happen at all. The growth definition comes from the BLS article cited. The screen does not measure demand for any company's records, record quality, price or payout, which only SourceX's review can assess, and the rank is a reading order for your outreach, not a valuation. For the records themselves, see mature operational workflows, recognizing several years of operating history and whether old data has value for AI.
Where this argument is wrong
- A company can be both. A 12-year-old firm with 300 employees that is also growing fast passes everything. The contrast here is attention against history, not a case against growth.
- Quiet can mean undocumented. Some long-lived companies run on paper, inboxes or one person's memory, the reliance on individual knowledge flagged above as a poor fit.
- Rights. Mature service firms often hold client material, and SourceX treats material that belongs to clients as usually not a fit.
- Demand. Relevance to current data needs is checked in review, and requirements differ in size, operating history, language and data category, so a strong row can be declined today and fit another program later.
- Synthetic data. Epoch AI's report on AI in 2030 says developers are unlikely to run out of data for large-scale training before 2030 and that multimodal and synthetic data could extend the supply; if that holds, demand for real records could narrow.
- Base rates. The BLS survival figure covers establishments from a single cohort and says nothing about records.
A measurable next step. Pick ten companies from your network: the five most discussed and the five longest-running you can name. Put all ten through the seven questions within two weeks, then count for each group the companies with four of four fit facts known and the CEOs who agreed to a conversation. If the most-discussed group matches the longest-running group on both counts, the thesis is wrong for your network, and attention is a fine place to start. The venture platform team overview covers the rest of the program from the portfolio side.
Next step
Refer a company. No cap on rewards. Default reward on a company's first qualifying transaction: the greater of $50,000 plus 5% and 15% of Actual Supplier Payout, with no cap; further qualifying transactions first agreed within 12 months of the first Closing pay 15%. Eligibility and payment conditions apply, including the USD 100,000 Minimum Transaction, and a signed Schedule A or addendum can differ. A lead, meeting or signature alone does not trigger payment.
When the screen puts a quiet company first, register for a partner link, confirm the basics on who qualifies, read the program terms and ask the relationship owner before anyone contacts the CEO; the company fit checker gives a quick, non-binding first pass, and more network playbooks sit in the guides.
Putting a quiet contractor ahead of a scale-up is a referral judgment, and judging SourceX as a company to invest in is a separate one; investor materials are available on request through the SourceX investor overview.
This article is for information only and is not an offer to sell or a solicitation of an offer to buy securities. Any offering by SourceX is made only to verified accredited investors through its offering documents.