# Owner referral information needs: study design | SourceX

Canonical: https://refer.sourcex.si/guides/what-owners-want-to-know-before-sharing-a-referral-link-internally
Last updated: 2026-11-04

Nobody has studied this for SourceX introductions yet, and this page reports no answer. It publishes the design of a study that could: consented interviews with CEOs who received a partner's referral link, drawn from four groups, namely owners who passed it to colleagues and applied, owners who passed it on and stopped, owners who applied without involving anyone, and owners who declined or never answered. Each interview rebuilds every internal forward in order and codes what the owner needed settled first. Findings will appear only after the interviews close.

The study needs SourceX and consenting partners, because only their dated records show who received a link. It changes nothing about how any company is reviewed, and no partner learns which of their introductions were sampled.

## The gap between receiving a link and deciding

Under the program, the CEO decides whether the company discusses licensing its records, and signs for it. The people who could answer the follow-up questions often sit elsewhere in the company: records with operations, contracts with counsel, payment with finance. SourceX's own page for CFOs, for example, lists four questions it expects a finance chief to bring: how the value is set, when and how the company is paid, how the income is recorded and what costs come out first. That list is SourceX's expectation, published on its site. It is not evidence of what owners ask before they pass a link on.

The internal forward itself has never been looked at closely: who receives it, what the owner writes above it, what comes back, and what the owner wanted to know before letting it travel. If owners mostly need one thing settled first, a partner's note could settle it in its opening line; if needs differ by company size or recipient, one note may be the wrong tool. Nobody can yet say which.

Forwarding does not move credit. Section 2 of the terms says attribution is decided from dated records and that "an earlier link click alone does not decide credit." The study asks about owners, not about attribution.

## Four research questions, written so no answer is assumed

1. Before the first internal forward, what did the owner want to know, and where did the answer come from?
2. Who received the link inside the company, in what order, and with what words from the owner?
3. Which questions came back from colleagues, and which of them did SourceX's published pages already answer on the day?
4. Do owners who kept the link to themselves, or declined, describe different needs from owners who passed it on?

No hypothesis is ranked in advance. The coding frame further down lists candidate topics so that two coders share definitions, not so the study can confirm them.

| Term | Meaning in this study |
|---|---|
| Owner | The company's CEO, who decides whether it discusses licensing; in an owner-run company the owner usually holds that role |
| Internal forward | Any passing of the link, or of the partner's note that carried it, to someone inside the company before the owner decides |
| Information need | A question the owner says had to be answered before a forward or before deciding |
| Held back | The owner read the link and passed it to nobody |

## Sampling frame, four strata and consent

The frame is every introduction that consenting partners submitted from their dashboards during one entry window of two quarters, plus every application SourceX credited to those partners in the same window. That yields a dated list of companies whose CEO received a link, including companies that never applied, which no study of applicants alone can supply.

| Stratum | Owners in it | Planned interviews | What it adds |
|---|---|---|---|
| A | Passed the link inside the company, then applied | 6 to 10 | The route that reached SourceX |
| B | Passed it inside the company, then stopped | 6 to 10 | Where internal circulation ended |
| C | Applied without involving anyone | 6 to 10 | Whether a forward is needed at all |
| D | Declined, or never answered the partner | 6 to 10 | The group an applicant study never hears |

Strata A to C are assigned from one screening question at consent ("Did you pass the link to anyone inside the company before deciding?"); stratum D comes from the partner's dated log. Sampling inside each stratum is purposive by headcount band, so a 60-person firm and a 600-person firm both appear.

Recruitment rules, fixed before the first invitation:

- **One invitation per owner**, sent by a research lead who works on neither sales nor review. Stratum D gets no reminder.
- **Never invited:** an owner who asked not to be contacted again, and a company SourceX already knew before the introduction, since section 3 treats any earlier lead, account or email with SourceX as a duplicate.
- **Written consent** covers recording, transcription, quotation and withdrawal up to a dated analysis freeze.
- **Timing:** the interview takes place after the company's outcome is recorded, or 120 days after the link arrived, whichever comes first, so it never touches a live decision.
- **Colleagues:** with the owner's permission, one colleague who received a forward may be interviewed separately, with their own consent. The owner still speaks for the company.
- **No incentive tied to outcomes:** any thank-you is the same for every participant and unconnected to any application.

## The internal-forwarding interview guide

Each interview runs about 40 minutes and follows six modules in a fixed order. The interviewer may reorder probes inside a module, never the modules.

| Module | What the interviewer asks | Follow-up rule | Never asked |
|---|---|---|---|
| 1. Arrival | How the link reached you, from whom, and what you took it to be | Keep the owner's own words for SourceX and the partner; correct nothing during the interview | Whether the company would fit |
| 2. Forward map | Each person you passed it to, in order: who, why that person, what you wrote, what came back | Rebuild one forward at a time on a dated line; a forward to the board counts | Employees' names; roles are enough |
| 3. Before passing it on | What you needed to know before passing it on, asked open first | A card sort of candidate topics only after the open answer is recorded | Anything about records, figures or contract terms |
| 4. Holding back | If nobody else saw it: who would have, and what stopped you | "I decided alone" is a complete answer | Reasons the owner marks private |
| 5. Questions returned | What colleagues asked, and where the owner looked for answers | Note any page or person the owner names | Copies of internal emails |
| 6. Close | What the note should have said; consent for each quote | Read back every quote the owner approves | Any offer to reconnect the company with SourceX |

Module 3 carries the study. Asking open first and showing cards second keeps the candidate list from putting words in the owner's mouth, and the report keeps the two kinds of answer apart. Confidential detail an owner volunteers about records or customers is left out of the transcript.

## Coding frame, fixed before the first interview

The unit of coding is one forward, or one decision to hold back. Seven candidate families start the codebook:

| Code family | Defined as a question the owner wanted settled about |
|---|---|
| Who decides | Whether passing the link on commits the company to anything, and whose call it is |
| Who is asking | Who the partner is, and whether the partner could be paid |
| Staff effort | Which people would have to do what, and when |
| What leaves the company | Whether files, records or names would be asked for |
| Money flow | How the company would be paid, and whether a partner's reward changes that |
| Timing | Whether this period suits the company |
| Rights | Whether customer contracts or other obligations allow it |

Two coders independently code the first eight transcripts, reconcile differences and freeze version 1.0 of the codebook. After that, a quarter of the remaining transcripts are coded twice and agreement is reported family by family. A code added after the freeze carries a dated memo explaining why. Coders work without seeing the company's outcome field.

For each coded question, an analyst then runs a **published-answer check**: does a SourceX page live on the interview date answer it as asked? Yes, partly or no, with the URL and date. That separates needs a page could meet from needs only a conversation can meet. For the money-flow family, for instance, section 4 of the terms already says the reward "is paid separately by SourceX and is never deducted from the supplier's agreed compensation"; whether owners found that sentence is an open question.

## Counting the owners who forwarded nothing, declined or never replied

Every introduction in the frame appears in the recruitment ledger, broken out by stratum, whether or not an interview happened.

| Ledger line | A | B | C | D |
|---|---|---|---|---|
| Introductions in the frame | Not yet counted | Not yet counted | Not yet counted | Not yet counted |
| Excluded before invitation, with reason | Not yet counted | Not yet counted | Not yet counted | Not yet counted |
| Invited | Not yet counted | Not yet counted | Not yet counted | Not yet counted |
| No reply | Not yet counted | Not yet counted | Not yet counted | Not yet counted |
| Declined the interview | Not yet counted | Not yet counted | Not yet counted | Not yet counted |
| Interviewed | Not yet counted | Not yet counted | Not yet counted | Not yet counted |
| Withdrew before the freeze | Not yet counted | Not yet counted | Not yet counted | Not yet counted |

Interviews describe; they cannot show cause. If owners in stratum B mention rights more often than owners in stratum A, the report will give the two counts and call it a difference between groups of interviewees, never the reason anyone stopped. Whether a different note would change behavior takes a controlled comparison, such as [the five-factor trust study](/guides/what-makes-a-company-introduction-feel-trustworthy) or the role-by-role comprehension stage in [the landing page test for different decision-makers](/guides/the-referral-landing-page-test-for-different-decision-makers).

## Reporting rules, and why the findings section is empty

As of 2026-10-27 no interview has taken place. The eventual report would follow COREQ, the 32-item checklist for reporting interview and focus group studies by Tong, Sainsbury and Craig that the EQUATOR Network lists.

- Code frequencies appear as counts ("5 of 9 owners in stratum B"), never as percentages of all owners.
- Forward maps are shown by role only; quotes appear only with consent and without company identifiers.
- Partners receive aggregate results only: never transcripts, and never whether their own introduction was sampled.
- Released alongside the results: the invitation text, consent form, interview guide, codebook with its version history, and the full ledger.

## Where a forwarding study can mislead

- **Recall.** Rebuilding a forward weeks later invites tidy stories; dated partner records anchor the timeline, and owners are never asked for their emails.
- **Who says yes.** Owners willing to be interviewed may hold stronger views; the ledger shows the gap.
- **Whose study it is.** Owners may soften criticism when SourceX asks; an interviewer outside sales and review, and a close without any offer, reduce that pressure.
- **The frame.** Owners who never opened the link are invisible to a record-based sample.
- **Size.** Between 24 and 40 interviews describe these owners, not owners in general.
- **Roles.** In a smaller company the "finance chief" may be an outside accountant; code roles as the owner describes them.

## What a partner can do while the study is only a design

Write the note so it still reads correctly after the owner passes it on without you: the CEO decides, SourceX decides fit after its own review, your possible reward stated plainly, and no files wanted from anyone. Then leave the choice of who sees it to the owner. When a finance chief's caution holds the company back, [the founder-and-CFO worksheet](/guides/the-founder-is-interested-but-the-cfo-is-not-ready) sorts each concern by who can answer it; [the founder decision interview protocol](/guides/what-makes-a-founder-say-yes-to-a-data-licensing-conversation) studies the yes or no itself rather than the forward, and SourceX's [process overview](/how-it-works) is what most owners read first.

## 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.

Partners whose introductions should be countable in a study like this can [register](/register), sign the Channel Partner Agreement before submitting anything, and submit each introduction from the dashboard so a dated record exists; check the company against [who qualifies](/who-qualifies) and the [program terms](/terms) first, and browse related designs in the [guides hub](/guides).

## Sources

- https://refer.sourcex.si/terms
- https://refer.sourcex.si/who-qualifies
- https://refer.sourcex.si/how-it-works
- https://sourcex.si/for/cfo
- https://www.equator-network.org/reporting-guidelines/coreq/
