A useful initial screening profile for businesses
- Operating maturity. SourceX looks for companies in the U.S., Canada, the EU, the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia (outside the U.S., Canada and the EU, at least 70% of the data must be in English) with 50 or more full-time employees at peak and several years in operation. Some programs set higher thresholds.
- Original, proprietary documentation. SOPs, internal knowledge bases, project histories, QA records, support workflows and similar material the company created itself.
- Rights to license. The company owns the material, or can license it without breaching client confidentiality, employee privacy or third-party contracts.
- The CEO's go-ahead. The company's CEO decides whether to open a discussion about licensing its proprietary data. Once a company is acquired, the acquirer's CEO decides; in a formal insolvency process the CEO alone may not be enough.
- Relevance to SourceX's current data needs. The proprietary material has to match a real program with AI developers worldwide. SourceX checks this during review.
Usually not a fit
- Personal consumer data, scraped data or data the company does not control.
- Material that belongs to the company's clients rather than the company.
- Businesses that only want to sell a one-off file with no documentation or context.
- Introductions made without the decision maker's knowledge.
Program-specific requirements
SourceX's current data needs differ in company size, operating history, geography, language and data category. SourceX applies the criteria of the specific program, not one universal minimum. A company that does not fit one program may fit another later.
Try the company fit checker for a preliminary, non-binding screening.
Why doesn't VC backing make a company eligible?
Because none of the checks SourceX screens on mentions funding. A company is screened on its size at peak (50 or more full-time employees, contractors excluded), years of records it created itself, its rights to license them, a CEO who authorizes the discussion and its fit with a current program. A term sheet proves none of these. A celebrated venture-backed company can fail every check, and a company that never raised outside capital can pass them all. Screen each company first, then promote only to the ones that pass.
For a platform team this changes the order of work. The useful venture channel is often wider than the funded portfolio: portfolio customers, LP-owned businesses, former portfolio companies now inside an acquirer. Screening comes before any portfolio-wide announcement, because a newsletter line that reads "our companies qualify" implies an acceptance nobody has given.
| Statement type | Statement | Basis |
|---|---|---|
| Program fact | Introduced companies need 50 or more full-time employees at peak; partners sign the SourceX Channel Partner Agreement before submitting introductions | Program terms, Version 2026-10-22.2, since replaced by version 2026-11-04 |
| Program fact | Every program applies its own criteria, and a company that misses one program's bar may still suit a different program later | The screening profile above, checked 2026-10-26 |
| Conditional | Whether a screened company is accepted, what it licenses and whether any reward follows | SourceX review and the company's own signed agreements |
| Our thesis | For an early-stage fund, more of the companies that pass the size and history checks sit outside the portfolio than inside it | A working hypothesis, not measured; the matrix below is how you test it |
| Term | How this section uses it |
|---|---|
| Operating entity | The legal entity whose employees created the records and that holds them today |
| Controller | Whoever directs that entity now: its board and CEO, an acquirer, or a trustee, receiver or assignee in a formal process |
| Mandate | The criteria of the specific SourceX program a company is screened against: size, history, geography, language and data category |
| Promotion | Any message that presents the program to one company or to a group of them |
Who controls the company today?
Name the operating entity and its current controller, and write down the date you checked. The fund's shareholding is not the answer. Delaware's General Corporation Law, for example, says the business and affairs of every corporation it governs are managed by or under the direction of a board of directors, unless the statute or the certificate of incorporation provides otherwise. Shareholders, a lead investor included, do not run the company by owning shares. Under the program, opening a licensing discussion and signing for the company fall to the CEO. A board seat lets a fund speak in the boardroom; it does not let the fund start that discussion.
What a fund knows about a company's systems usually arrived through board packs or information rights, and that knowledge comes with a duty. NVCA's model management rights letter, for instance, makes confidential information an investor receives or learns under it subject to the confidentiality provisions of the Investors' Rights Agreement. The fund's knowledge is a reason to ask the CEO, never material to pass along.
Three ownership facts deserve a date stamp before any outreach:
- The entity that holds the records. Not the fund, a holding vehicle or a brand name. After an acquisition, the records may sit inside the buyer's group, and for the reward cap that whole group, parent and subsidiaries alike, is a single company.
- The current CEO, and since when. Once a company is acquired, the acquirer's CEO decides for it, not the former board.
- Any formal process. In an assignment for the benefit of creditors, a receivership or a bankruptcy case, the CEO alone may not be enough, because someone else may control the assets. Who holds authority in each process is compared separately.
The mandate-by-ownership eligibility matrix
Ownership changes who says yes and where a check usually breaks. It never changes the checks themselves.
| Ownership today | Who authorizes | Size and history check | Rights check | Usual failure | First route |
|---|---|---|---|---|---|
| Founder-led, venture-backed, minority investors | The founder-CEO | Full-time headcount at peak; the year the company's own records start | Product data often sits in customer tenants; internal tickets, QA and project records are the company's own | Too young, or records belong to customers | The board member or founder relationship asks the CEO |
| Fund-controlled, board majority held by investors | The CEO | Unchanged by control | Unchanged by control | Fund enthusiasm stands in for the checks | The CEO, with the deal partner informed |
| Private equity platform with add-ons | The platform CEO, for the whole group | Group headcount at peak; each acquired business's history | Depends on how each business was bought | Records stranded in an add-on's old systems | The operating partner asks the platform CEO |
| Former portfolio company, since acquired | The acquirer's CEO | The group's; pre-deal archives may not have moved | Depends on the deal structure | Nobody at the fund still has a path | The former board member, only if the path is live |
| Owner-run, never raised outside capital | The owner, usually the CEO | Often long, steady histories | Client confidentiality clauses | Nobody has asked | Whoever holds the relationship: an LP, a supplier, a portfolio founder |
| Wound down or in a formal process | The CEO may not be enough; the estate's controller decides | Headcount at peak, not today's | The estate's control, plus personal data limits | Archives deleted during the shutdown | The controller of the estate, with counsel |
Funding stage appears in no column. A seed round, a Series D and no outside capital at all lead to the same five checks. Operating, acquired and wound-down companies can all qualify, and not every company does. How a sale was structured decides who holds the records afterward; asset and stock sales differ here.
Method note. The matrix applies the screening profile on this page and the program terms, both checked 2026-10-26, together with the program's settled rules: the CEO authorizes, one cap covers a parent together with its subsidiaries, and a staffing firm counts only its own full-time employees. The matrix leaves out demand, record quality and payout; SourceX assesses relevance to current data needs in review. A company can sit in two rows at once, such as a fund-controlled company in a receivership; use the stricter route. To use it as a worksheet, give each company one line: the date you checked ownership, each check marked pass, unknown or fail, and the first route.
Which mandate, language and history questions come before promotion?
Five questions, answered from a conversation with the relationship holder, never from files, samples or logins:
- What was full-time headcount at its peak, and when? Contractors are excluded. For a staffing firm, only the firm's own full-time staff count, never the workers it has placed on assignment with clients; report the two figures side by side, and make the introduction anyway if the records are strong.
- When do the company's own records start, and in which systems? Several years in operation is the screening baseline, and some programs set higher thresholds.
- Who created the records, and does any of them belong to clients? Client-owned material, scraped data and data the company does not control are usually not a fit.
- Where is the company, and what language are its records in? Companies in the U.S., Canada and the EU qualify, as do companies in the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia when at least 70% of the data is in English. The English-language rule is explained separately.
- Does the CEO know about the idea and want the conversation? An introduction that reaches SourceX before it reaches the decision maker is usually not a fit.
Mark each answer pass, unknown or fail, with a date. Unknown is not a fail; it is a question for the relationship holder.
Three screening decisions, one of them rejected (Illustrative)
Illustrative. All three cases are generic companies with invented facts, and none implies an outcome.
| Check | Case 1: Series C logistics software, 310 people | Case 2: Series A AI startup the fund promotes widely | Case 3: family-owned distributor, no outside capital |
|---|---|---|---|
| Ownership today | Founder-CEO, minority investors | Founder-CEO, minority investors | Owner-CEO, 26 years old |
| Full-time headcount at peak | 310: pass | 34 full-time plus 70 contractors: fail | 160: pass |
| Own records since | 2014, support and implementation systems: pass | 2023: fail | 2001, order and returns systems: pass |
| Rights | Own support and implementation records pass; customer shipment data in customer tenants does not | Scraped web text and customer uploads: fail | Its own records; supplier contracts to be read in review: unknown |
| CEO wants the conversation | Yes, after a board-member call | Yes | Not yet asked; a portfolio founder who supplies it will ask |
| Decision | Route: the company applies, describing its internal records only | Reject: not introduced | Route once the CEO agrees |
| What would change it | Rights review sets the scope | 50 full-time employees and records the company created itself | The CEO declining |
Case 2 is the one a fund is most tempted to promote, and its round size, press coverage and growth appear nowhere in the decision. Case 3 is the company the fund did not back and may not have thought of.
How do you route uncertain or excluded cases without implying acceptance?
Route each row by its weakest known check:
- Every known check passes and the CEO wants the conversation. Register, share your partner link, and let the company apply itself. Creating a link or sending an email does not by itself make an introduction eligible for a reward.
- Unknowns remain. Ask the missing metadata questions through the relationship holder, or suggest the CEO try the company fit checker, a preliminary, non-binding screen.
- A check fails. Record the failed check, the date and the trigger that would change it, and do not introduce. Failure is dated, not permanent.
- A formal process or a deadlocked ownership. Find the controller first; in a 50/50 ownership or under a trustee or court, the CEO alone may not be enough.
When the platform team mentions the program to the whole portfolio, describe the criteria and invite CEOs to screen themselves. Never list companies as eligible, and say "may be a fit; SourceX decides after review". Whether to run that as an opt-in program is weighed in opt-in versus mandatory portfolio programs, and the portfolio side of the referral process is covered for venture platform teams.
Where can this matrix be wrong?
- Programs change. Criteria follow each program and can move. A screen is valid on the date it was made; screen again when the profile above changes.
- Growth-stage portfolios. If most portfolio companies already employ hundreds of people and hold a decade of records, the thesis fails and the portfolio is the right place to start.
- Rights are decided in review. The matrix flags where rights usually break; it cannot clear a customer contract.
- Confidentiality. A fund partner who describes a company's systems from board materials may breach a duty the matrix cannot see.
- Relationship cost. Asking a portfolio founder to raise this with a supplier spends goodwill, and no data on how founders react has been published.
A measurable next step. Screen every portfolio company and 20 network companies in one sitting and count passes, unknowns and fails by ownership row. Thirty days later, count the CEOs who agreed to a conversation. If portfolio rows produce as many full passes as network rows, the thesis is wrong for your fund.
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 a company passes every check you can confirm and its CEO wants the conversation, register for a partner link, run it through the company fit checker and read the program terms before anyone contacts the company.
Whether a company qualifies for an introduction is a different question from whether to invest in SourceX; investor materials on the latter 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.