A path that ends at a custodian is unfinished, however warm it feels. That is the one rule a partner can act on before any data exists, and the study below is built to test whether it is right.
Where each claim on this page stands
| Claim | Standing | Basis |
|---|---|---|
| The CEO authorizes a licensing discussion and signs for the company | Program rule | SourceX program decision applied across refer.sourcex.si |
| An introduction the decision maker does not know about belongs on the usually-not-a-fit list | Published screening guidance | Who-qualifies page, checked 2026-10-26 |
| Credit goes to the earliest valid introduction inside the 90-day attribution window and attaches when the company submits a verified application | Program terms | SourceX terms, Version 2026-10-22.2, since replaced by version 2026-11-04 |
| Whether a reached company is accepted, licenses anything or produces a reward | Conditional | SourceX review and a signed agreement |
| Reach depends on the introduction path | Hypothesis | Unmeasured; the protocol below tests it |
| Term | Meaning here |
|---|---|
| Authorized data owner | The CEO, whose approval opens a licensing discussion and whose signature binds the company |
| Sponsor | A CFO, COO, general manager or department head who owns records day to day and can carry the question to the CEO |
| Custodian | The IT manager or system administrator who keeps the records and cannot authorize their use |
| Reach | The CEO has heard the proposal from someone they trust and has answered, yes or no |
| Disposition | The final status of a case on the cutoff date |
| Unknown case | A case whose reach status cannot be established by the cutoff |
Reach is not agreement. A CEO who hears the proposal and declines has been reached: a success for reach and a no for the conversation. Keeping the two apart is what makes the study useful to someone deciding how to introduce.
The decision you can make before any data exists
Prefer paths that end at the CEO with the fewest hand-offs, and treat a path that ends at a custodian as unfinished. The failure points below are hypotheses the study would test, not findings.
| Path | First person the introducer reaches | Hand-offs to the CEO | Likely failure point |
|---|---|---|---|
| Direct | The owner-CEO | None | The CEO declines |
| Sponsor | CFO, COO or general manager | One | The sponsor never raises it |
| Custodian | IT manager or system administrator | Two or more | Enthusiasm without authority; it stalls below the sponsor |
| Adviser | Outside accountant, lawyer or banker | One | The adviser's own rules or reluctance |
| Board or investor | A director or investor | One | The CEO feels pushed, or the director defers to management |
The roles are set out in data owner vs custodian. Asking an intermediary for an introduction is covered in how to ask a client for an introduction, and choosing between a sponsor and the CEO in introduce through a sponsor or directly to the CEO.
The protocol: frame, consent and variables, published before any result
Fix the cohort, the consent rules and the variables in writing, date them, and only then start counting.
- Sampling frame. Every introduction attempt logged by participating partners during one calendar quarter, the entry window. Partners opt in before the window opens and log every attempt, including those that never got past the first person.
- Cutoff. Dispositions are read on one date, 90 days after the window closes. A case without a CEO answer by then is unknown, not failed.
- Consent. Partners consent to sharing their logs. Companies are never named: each is coded by sector and headcount band. No records, samples, documents or credentials are collected, and quotes appear only with the partner's approval.
- Proposed analysis. Reach by path, reported as a range; the agreement rate among reached CEOs; days from introduction to the CEO's answer by path, with open cases treated as censored at the cutoff. No model or causal claim that was not written down before the window opened.
| Variable | Values |
|---|---|
| 1. Path type | Direct, sponsor, custodian, adviser, board or investor |
| 2. Age of the introducer's relationship | Under 1 year, 1 to 5 years, over 5 years |
| 3. Peak full-time headcount band | Under 50, 50 to 199, 200 or more |
| 4. Control structure | Owner-run, executive-run, investor-controlled, or flagged (two equal owners, a bankruptcy trustee or court control) |
| 5. Hand-offs logged | Number of people between the introducer and the CEO |
| 6. CEO answer | Yes to a conversation, no, none by the cutoff |
| 7. Days to answer | Introduction date to the CEO's answer, or censored at the cutoff |
| 8. Disposition | One code from the ledger below |
Variable 4 matters because the CEO alone may not be enough in some structures. Where two equal owners, a bankruptcy trustee or a court controls the company, the study flags the case and reports it separately.
The sponsor-reach disposition ledger
Every case ends in exactly one of eight codes. The logic is borrowed from survey research, where AAPOR's Standard Definitions sort each sampled case into one of four groups: completed interviews, eligible cases not interviewed, cases of unknown eligibility and cases that are not eligible.
| Code | Disposition | Group | In the reach denominator? |
|---|---|---|---|
| 1.1 | CEO reached and agreed to a conversation | Reached | Yes |
| 1.2 | CEO reached and declined | Reached | Yes |
| 2.1 | Stalled at a sponsor | Not reached | Yes |
| 2.2 | Stalled at a custodian | Not reached | Yes |
| 2.3 | Introducer withdrew, or the relationship owner declined to ask | Not reached | Yes |
| 3.1 | No answer by the cutoff; company known to pass the fit facts | Unknown | Yes, as a range |
| 3.2 | No answer by the cutoff; fit facts unknown | Unknown eligibility | Partly, through the estimate e |
| 4.1 | Company fails a published screening criterion: under 50 full-time employees at peak, based outside the program's listed regions, or records owned by its clients | Not eligible | No; reported separately |
Three rates come out of the ledger, always with their counts:
- Reach, lower bound: codes 1.1 and 1.2, divided by all cases coded 1.1 to 3.2.
- Reach, upper bound: codes 1.1, 1.2, 3.1 and 3.2, divided by the same total.
- Reach, central estimate: codes 1.1 and 1.2, divided by codes 1.1 to 3.1 plus e times the 3.2 count.
AAPOR's outcome-rate equations carry the same term: e is the estimated proportion of unknown-eligibility cases that are in fact eligible, and AAPOR's guidance is that it must rest on the best available evidence and must never be chosen to raise the rate. Here e is the eligible share among cases whose fit facts were established in the same window, and the value used is published with the results. The agreement rate is reported separately: code 1.1 divided by codes 1.1 and 1.2.
Method note. The ledger, variables and rates are built from SourceX's published fit criteria, screening guidance and attribution rule (terms and who-qualifies page, checked 2026-10-26), AAPOR's disposition logic and the STROBE reporting checklist. The study measures whether introductions reach the CEO and how fast. It does not measure demand for any company's records, SourceX's review decisions, completed handoffs or payouts.
Keep stalled and declined introductions in the denominator
Drop them and the rate measures the method, not the introductions. An introduction that stalled with an IT director is the most informative case in the study, and it is exactly the case a tidy spreadsheet deletes as "inactive".
Illustrative case: enthusiasm without authority. Generic company, invented facts. A partner who implements maintenance systems introduces a 350-person logistics company through its IT director, who is keen and offers to pull an export. The partner declines the export, since partners never handle records, and asks the IT director to raise the idea with the CFO. Ninety days later nobody above the IT director has heard of it. Code 2.2. It stays in the denominator.
Illustrative case: reached and declined. An outside accountant raises the idea with the owner-CEO of a 140-person engineering firm during year-end planning. The CEO says no: a sale process is starting and nothing new happens until it closes. Code 1.2. For reach, a success; for the conversation, a no. The note records the event that would justify asking again.
Results stay descriptive. Introducers choose their paths, and those who already know the CEO will choose the direct path, so a higher reach rate on direct paths could reflect stronger relationships rather than the path itself. Two checks narrow that gap without closing it: compare paths within the same introducer, and compare paths within the same band of relationship age. Neither turns an observational study into an experiment, and the report should say so.
What the published report must contain
STROBE, the reporting checklist for observational studies, asks for the numbers at each stage of a study and the reasons for non-participation at each stage, and suggests a flow diagram. The report follows it.
| Section | Contents | Status |
|---|---|---|
| Methods | Frame, entry window, cutoff, consent, variables, the value of e and how it was set | Published on this page |
| Flow | Counts at each stage with reasons, drawn as a flow diagram | Not yet collected |
| Results by path | Reach range, agreement rate and days to answer, each with counts | Not yet collected |
| Counterexamples | Cases that cut against the main result, coded and described | Not yet collected |
| Limitations | Selection, coding, small cells and program changes, as listed below | Drafted below |
How long each stage takes after a verified application belongs to the stage duration benchmark; whether introductions end in a completed, paid handoff belongs to the conversion benchmark. Mapping which relationships a fund has in the first place is the subject of the hidden business data network.
Limits, and what would make the conclusion wrong
- Selection. Partners who volunteer their logs are likely the more careful ones, so reach could be overstated for partners in general.
- Small cells. One quarter of introductions will leave some paths with few cases. Report counts beside every rate, and do not rank paths with a handful of cases.
- Coding. Whether a CEO "heard" the proposal comes secondhand from the introducer. A second coder recodes a random subset and the agreement is reported.
- Control structures. Cases with two equal owners, a trustee or court control are reported apart, because the CEO alone may not be enough there.
- Rule changes. The program terms are versioned (this page uses Version 2026-10-22.2, since replaced by version 2026-11-04). A change in attribution or authorization rules during the window splits the cohort.
- Distance from outcomes. Reach is early. A high reach rate says nothing about review acceptance, rights or completed handoffs.
The hypothesis is wrong if sponsor and custodian paths reach the CEO as often as direct paths once the age of the relationship is held equal. In that case, the effort of finding the CEO path first is not worth it, and partners should introduce through whoever they know best.
A measurable next step. Run one quarter with every participating partner logging every attempt. Publish the flow and the eight-code ledger with the cutoff date, even if the counts are small.
Before an introduction enters the cohort
Each company must plausibly fit before anyone is asked: 50+ full-time employees at peak (contractors excluded), documented operating history and rights to license. 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. Discovery covers metadata and nothing more: which records exist, their rough volume, the years they span, the system that holds them and their owner. Nobody asks for or handles records, samples or logins.
Only one partner is credited per company, so agree who submits before anyone approaches the CEO, and sign the SourceX Channel Partner Agreement before submitting; a copy signed under the agreement's older name (Channel Partner Non-Circumvention and Non-Disclosure Agreement) carries on in force. Three questions remain open until SourceX's review: whether the records match current data needs, whether the company's contracts limit reuse, and what a license would cover.
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.
Once a path ends at a CEO who has said yes to the conversation, register for a partner link, set the company beside who qualifies and work through the program terms ahead of any submission. More research guides are in the guides hub.
This protocol studies introductions, not SourceX as a company; readers deciding whether to invest in SourceX can request investor materials 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.