SourceX's terms give the reason in a single clause. Section 2 says attribution "is decided from dated records, including written introduction evidence, existing applications and preexisting relationships; an earlier link click alone does not decide credit." A path comparison built on clicks measures something the program ignores.
What each path leaves on the record
| Path | What reaches the CEO | Dated record you hold | What you can count | What you never see |
|---|---|---|---|---|
| A. Personal link | Your note, with your partner link | Your sent email | The dashboard stage for that company | Whether the CEO read it |
| B. Resource page, sent by you | The same note, linking to your page, which carries the same link | Your sent email and any written reply | The dashboard stage for that company | What the CEO read on the page |
| C. Resource page, found alone | Your page, reached with no note from you | None, unless the CEO writes to you | Only companies that contact you | Readers who apply on their own |
| D. Dashboard submission | Your direct submission, after the CEO agreed | The submission and the CEO's written yes | The dashboard stage | Nothing beyond A |
Row D is the route how it works describes next to the personal link: partners get a link and can also submit an introduction from the dashboard. It is not an arm below, but a CEO from arm A or B who asks you to make the introduction for them ends up there.
Row C is where overclaiming starts. A reader who finds your page, clicks your link and applies may or may not be credited to you; SourceX makes that call from the records, and a click does not make it. The company loses nothing either way: under section 4 of the terms SourceX pays any reward separately, from its own share, and never out of what the company is paid, so the company's amount is identical with or without you. Treat path C as a public service you cannot measure well, not as a channel to tune.
The path comparison measurement design
| Element | Specification |
|---|---|
| Question | Per CEO contacted, does a note pointing to your resource page produce more or fewer qualified introductions than a note carrying your personal link? |
| Unit | One CEO you know, contacted one to one, whose company you believe may meet the published fit facts |
| Arm A | Your note with your personal link |
| Arm B | The same note, word for word, with a link to your resource page where the personal link was |
| Assignment | A logged coin flip per CEO before writing |
| The page during the test | One dated version, frozen for the window; no per-company pages and no form asking about the company |
| Window | Sends across one quarter; cutoff 120 days after the last send, covering the 90-day attribution window and some review time |
| Primary measure | Qualified introductions per CEO: companies that reached a verified application credited to you and not closed at the cutoff |
| Secondary measures | Written requests to proceed; questions asked per CEO; days from note to application; CEOs who say they read the page |
| Path C | Companies that contacted you after finding the page alone, logged separately and never compared with A or B |
| Analysis | By assigned arm, whatever each CEO did next |
The log behind each count is the record the agreement already expects. Section 3.3 of the sample Channel Partner Agreement says each introduction record must identify the company, domain, partner, opportunity, introduction method and timestamps. Add one column for the arm and the test log is complete. If your firm keeps introductions in more than one place, the reconciliation worksheet shows how to match them to portal activity.
The resource page stays inside the rules it already follows
The page in arm B is not a new kind of promotion. It follows the publicity and outcome rules in sections 4.5 and 4.6 of the Channel Partner Agreement and the disclosure guidance already written for a community resource page and for a link in an association newsletter. This test changes none of those rules and repeats none of them.
It adds three constraints of its own. The page is identical for every CEO during the window, so arm B is one thing rather than many. It never names a company or shows company facts. And it never invites a reader to send company information to you, because nobody gathers company data for SourceX: the company applies itself, and its CEO decides whether it does.
Understanding and unwanted effects, recorded in both arms
When a CEO replies in either arm, ask one optional question before going further: "Who decides at your company whether to talk to SourceX, and who decides whether the company fits?" The correct answers are the CEO and SourceX. Log right or wrong per arm; a page that explains more should raise the share of right answers, and if it does not, the page is not doing the job you built it for.
Log these unwanted effects with the CEO's own words and a date:
- Endorsement. The CEO reads the page as SourceX vouching for your firm, or your firm vouching for SourceX.
- Approval. The CEO believes the page says the company qualifies or has been accepted.
- Records. Anyone sends files, exports or attachments: keep nothing, tell the sender, and fix the wording before the next send.
- Payment worry. The CEO asks whether using your link lowers what the company is paid. Answer that it does not, and note which arm prompted the question.
- A second voice. Someone other than the CEO writes back on the company's behalf. Record it; the CEO still decides.
Zeros, small counts and the rules that end the test
Many partners will see few qualified introductions in a quarter, and some arms will see none at all. Hanley and Lippman-Hand's 1983 JAMA paper was written for exactly that situation, studies in which none of the events of interest happened, and argued that a zero tends to be read very differently from one or more events (paraphrase). The arithmetic behind the caution is short. If a path's true rate were p, the chance of seeing no successes in n tries is (1 − p) multiplied by itself n times, and that chance drops to 5% once p reaches roughly 3 divided by n. Zero qualified introductions from 20 CEOs is still consistent with a true rate of roughly 14%. A zero does not show that a path fails; it shows you have not yet seen it work.
Set the floor before the first send: compare rates only when each arm holds at least 20 CEOs and the cutoff has passed, and below that publish counts only.
End or pause the test when:
- the page turns out to name SourceX as your partner without SourceX's written authorization: take it down, since section 4.5 of the agreement applies;
- SourceX publishes a new terms version, or the page needs any edit: close the window, date a new page version and start again;
- a CEO asks you to stop writing: remove them permanently and keep them out of both arms;
- records reach you: the records rule above applies, and the test waits until the wording is fixed.
An arm looking better halfway through is never a reason to stop.
What would change your default path
| Result at the cutoff | Reading | Decision |
|---|---|---|
| B at least matches A on qualified introductions, with more right answers to the understanding question | The page explains what a short note cannot | Point notes to the page |
| A beats B on qualified introductions, understanding equal | The extra step costs more than the page adds | Keep the personal link; keep the page for readers who look for it |
| Level on both | The page is reference material, not a conversion tool | Keep whichever path you can keep accurate with less effort |
| Fewer than 20 CEOs per arm, or zeros in both | No answer yet | Report counts and keep your current habit |
Path C never enters this table. If companies keep writing to you after finding the page alone, that is worth knowing, but it is a count without a denominator, and it says nothing about whether the page beats a personal note.
Who belongs in either arm
Only CEOs of companies you believe may meet the published fit facts: a full-time headcount that peaked at 50 or more, with contractors left out; years of records the company made itself; the right to license them; a location and language that fit the sentence on who qualifies; and a CEO who wants the conversation, because the CEO is the one who decides. Ask at metadata level only, about record types, rough years and systems, never for files, samples or logins. Whether the records match a current program, what the rights review finds and what a license would cover remain unknown until SourceX reviews the company.
Remove before assignment any CEO who tells you the company has already been in touch with SourceX: SourceX checks its own records before inviting an introduced company, and a company it already knew is not credited to anyone. If another partner also knows the company, only one introduction can be credited, and SourceX's duplicate review settles it from dated records. You must have signed the SourceX Channel Partner Agreement before the window opens, since introductions can be submitted only after signing. If the application path itself confuses CEOs in either arm, the referral form walkthrough is the place to look before blaming the link or the page.
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.
Before you send either path to a CEO, register as a partner, measure the company against who qualifies and work through the program terms, so every count in your log rests on a dated record. Related test designs are in the guides hub.