This is for partner growth managers who bring others into the SourceX program: a firm's advisors, a network's members, a fund's venture partners. The population is the introducers, not the companies they introduce. How a company's CEO understands a proposal is a different question for a different instrument.
Finishing onboarding is not understanding it
Completion rates, attendance and time spent are the clicks of onboarding: easy to count and silent on what the partner will do with a CEO next month. The outcome that matters is closer to what Kirkpatrick Partners calls Level 2: "The degree to which participants acquire the intended knowledge, skills, attitude, confidence and commitment based on their participation in the training." Confidence sits inside that definition, and this design measures it separately, because a partner who is wrong and sure is the one who promises a CEO something.
Scenarios beat recall questions here. The U.S. Office of Personnel Management describes situational judgment tests as presenting a work problem or critical situation tied to the job and asking how the person would handle it, built from critical incidents that subject matter experts supply and scored against those experts' judgments of the best and worst answers (paraphrase). For a referral partner, the critical incidents are already written down: they are the moments the terms were drafted to settle.
The three published terms every card is keyed to
| Family | The published text the key rests on | The misreading the cards probe |
|---|---|---|
| Payment trigger | "A lead, meeting or signed agreement alone does not earn a payable reward." (terms, section 5) | A signature, or an Estimated figure on the dashboard, means money is due |
| Credit | "Only one partner is credited per company." and "an earlier link click alone does not decide credit." (terms, section 2) | Whoever's link was clicked first owns the company |
| Who decides | Opening a licensing discussion and signing for the company fall to the CEO (who-qualifies) | An enthusiastic executive, a board or a former owner can say yes for the company |
Payment also needs the Acquisition Partner's final written acceptance of the complete, validated dataset, completed handoff, cleared payment to SourceX, payment of the company's own compensation, finance approval and the partner's tax and payment details, with a deadline of 60 calendar days after the later of acceptance and handoff. The cards test whether partners know that a list exists and where to read it, not whether they can recite it.
Nine scenario cards with their keys
Each card asks for an action, a one-line reason and a confidence rating from 1 (guessing) to 4 (certain). All situations are fictional; nobody answers with a real company.
| # | Family | Scenario | Best action | Key |
|---|---|---|---|---|
| 1 | Payment | A CEO you introduced signed SourceX's agreement on Friday. A colleague asks if you can invoice your reward on Monday. | Not yet; no reward is payable on a signature | Section 5: a signed agreement alone earns no payable reward |
| 2 | Payment | Your dashboard shows an Estimated reward. Your firm's finance lead wants to count it as money owed. | Treat it as a projection; only Paid means a payout was recorded | Section 5: portal estimates are projections, not promises |
| 3 | Payment | A CEO asks whether your reward will reduce what SourceX pays the company. | No; SourceX pays the reward separately | Section 4: "paid separately by SourceX and is never deducted from the supplier's agreed compensation" |
| 4 | Credit | You and a partner at another firm both know the same CEO and both plan to send a link this week. | Agree in writing who asks before either of you contacts the CEO | Section 2: one partner credited; overlapping claims are SourceX's to decide in duplicate review |
| 5 | Credit | A CEO opened your link in March. Another partner introduced her in writing in June, and she applied in July. | Do not claim credit from the click; SourceX decides from dated records | Section 2: credit from dated records; a click alone decides nothing |
| 6 | Credit | The company you plan to introduce emailed SourceX last year about something unrelated. | Expect it to be treated as a duplicate; the company can still apply | Section 3: any earlier email correspondence counts as a relationship |
| 7 | Decides | A CFO offers to submit the application herself while the CEO is traveling. | Wait for the CEO's yes; the CFO can help gather facts | Who-qualifies: the CEO decides; an introduction without the decision maker's knowledge is usually not a fit |
| 8 | Decides | Your fund holds a board seat. A colleague suggests the board approve the introduction. | Ask the CEO; the board seat does not start the discussion | Who-qualifies: a board seat gives a voice in the boardroom, not the decision |
| 9 | Decides | A portfolio company closed its sale to an acquirer last month. Its former CEO, now running a division, says yes. | Ask the acquirer's CEO, who now decides | Who-qualifies: after an acquisition, the acquirer's CEO decides |
Scoring, fixed before the first cohort. Two points when the action and the reason both match the published text; one point for the right action with a missing or wrong reason; zero when the action contradicts the terms. On cards 4 and 5, "SourceX decides this in duplicate review" is a full-marks reason. On cards 1 to 3 and 7 to 9, "I would ask SourceX" earns one point: safe, but the published pages already answer it. Report scores per family, never as one blended total.
The day-30 form. Write card B for each card A: same family, same key, different surface. Card 1B might read: "The company you introduced has delivered part of its data and SourceX is still validating it; can you invoice?" Same answer, different trigger. Two scorers mark every answer separately and settle disagreements against the published text, never by vote.
Two onboarding formats, assigned before anyone joins
| Element | Specification |
|---|---|
| Format A | The partner reads the terms page and how it works, then a 30-minute walkthrough of the three families |
| Format B | Format A plus three worked practice scenarios discussed in the session; none of them is a test card |
| Allocation | Alternate whole cohorts between formats in date order, decided before invitations go out |
| Day-0 measure | Form A of the cards, taken alone within 24 hours of onboarding |
| Primary measure | Form B score on day 30, per family, by format |
| Secondary measures | Day-0 score; confidently wrong answers (scored zero at confidence 3 or 4); appropriate "ask SourceX" answers; the share of each partner's introductions over the next 90 days in which the CEO's written yes predates submission |
| Not measured | Completion, attendance, satisfaction scores, reward amounts |
Both formats teach the same terms truthfully; neither mentions earnings beyond the published reward rule. A curriculum that bounds outcomes from the first lesson is set out in how to train partners without promising they will earn; this page measures whatever curriculum you run. SourceX's status words, used in card 2, are defined on the rewards page and in the status glossary.
Minimum sample, timing and stop rules
Compare formats only after at least 12 partners in each have taken the day-30 form, spread over at least two cohorts per format. Below that, report how many partners got each card right, per format, and no averages. Hold the retest at day 30, give or take three days, so forgetting has the same time to work in both arms.
Stop rules.
- SourceX publishes a new terms version: stop, rekey every card against the new text and start a fresh comparison.
- A card is missed by most partners in both formats: stop using it, check its wording against the published text, and if the published text itself is what confuses people, tell SourceX rather than teaching to the card.
- A partner answers with a real company or client: remove that answer, remind the cohort, and keep going.
- Anyone proposes using scores to rank, pay or drop partners: stop. Scores used that way change how people answer.
- A partner reports being told in onboarding that they would earn: stop both formats until the material is corrected.
Reading scores without inventing a pass mark
No pass rates have been published by SourceX or any partner, and this design sets no threshold from a guess. The reporting shell, filled only from real cohorts:
| Day-30 result | Format A | Format B |
|---|---|---|
| Partners retested | Not yet collected | Not yet collected |
| Payment cards, points out of 6 (median) | Not yet collected | Not yet collected |
| Credit cards, points out of 6 (median) | Not yet collected | Not yet collected |
| Decides cards, points out of 6 (median) | Not yet collected | Not yet collected |
| Confidently wrong answers, all cards | Not yet collected | Not yet collected |
| Partners lost before day 30 | Not yet collected | Not yet collected |
| Pattern | Decision |
|---|---|
| One family lags in both formats | The onboarding content for that family is the problem; rewrite it before changing formats |
| Format B leads on credit cards only | Keep format A and add worked credit scenarios |
| Confidently wrong payment answers appear in either format | Talk each one through before that partner's first introduction |
| No difference anywhere | Choose the shorter format and say the comparison found no gap at this size |
| Day-0 high, day-30 low | Understanding fades; add a short refresher at day 30 instead of a longer first session |
Downstream counts, such as introductions with the CEO's yes on record first, are reported beside the scores but never used to declare a winning format; too many other things move them. A site-level comprehension check on SourceX's headline, with a different three-question instrument, appears in the maximum-payout headline study.
What the assessment never asks
No card asks for a partner's pipeline, client names, company facts, records or systems. The cards are fictional on purpose: a partner learning the terms needs no real company, and a scoring sheet that held one would become a list of companies that never agreed to be on it. The real-world version of each card arrives later, when a partner meets an actual CEO, and the only test that counts then is whether the CEO decided, with accurate expectations about money, credit and timing. Partners see their own referrals' progress on the dashboard (where a partner sees the current stage), and nothing from it belongs on the answer sheet.
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 new partner can explain all nine cards from the published text, they are ready to register and sign the Channel Partner Agreement before submitting anything; keep who qualifies and the current program terms open while you rekey the cards. Other designs for partner teams sit in the guides hub.