It is written for a head of channel partnerships whose bench of introducers is outgrowing one inbox. Program facts come from SourceX's terms (Version 2026-11-04) and who-qualifies page, read on 4 November 2026; the taxonomy, routing and worked numbers are this page's own and Illustrative.
An accurate reply can still be a generic one
A partner writes: "Does a 70-person logistics company qualify?" The fast reply recites the rule: an introduced company must have reached 50 or more full-time employees at its peak, be able to document its operating history, hold the rights to license and offer an authorized contact. Every word is correct, and the partner has learned nothing, because the question was never about the rule. It was about this company, and the answer turns on facts the reply never asked for: how many of the 70 are full-time, whether the records belong to the company or its clients, and whether the CEO knows.
A generic answer is one you could send unchanged to every partner who asks the same words. For some questions that is right; for most, it is a way of not answering. Scaling support means knowing which is which before anyone replies.
What an active partner and a useful introduction look like from the support desk
Support load, counted by kind, is an early reading of partner health. An active partner is one who has signed the SourceX Channel Partner Agreement and asks about specific companies. Questions about named situations ("this company, this CEO, this record system") are a good sign; a stream of general questions with no company attached usually means the partner has not yet found anyone to introduce.
A useful introduction is one that reaches your screen with the facts needed to answer it: full-time headcount at peak, when the company's own records start, who created them, where the company is and whether the CEO wants the conversation. 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 question taxonomy and expert routing table
Six families cover nearly every partner question, each with a different answerer. One test assigns the family: what would I need to know to answer this partner and only this partner?
| Family | Example question | The answer depends on | Facts to collect before replying | Who answers | Reusable |
|---|---|---|---|---|---|
| 1. Program mechanics | "How is the reward calculated?" | The published terms, or the partner's signed Schedule A | Which terms version applies, and whether the partner signed an addendum | Your versioned answer library | Yes, with the version on it |
| 2. Company fit | "Does this 70-person company qualify?" | The company's own metadata | Full-time headcount at peak, records' start year and origin, location and language, CEO's knowledge | Your screener, applying the published checks | The rule yes, the verdict never |
| 3. Claims and duplicates | "Has anyone already introduced this group?" | SourceX's dated records, and your own claim log | The partner's dated introduction evidence | SourceX decides; you report only your own log | No |
| 4. The partner's own situation | "Which tax form do I send?" | The partner's residence, employer and professional rules | Nothing; you are not the right holder of these facts | The partner's own tax or compliance adviser | No |
| 5. Conversation craft | "How do I raise this with a CEO who said not now?" | Experience of real CEO conversations | The relationship and what was already said | A peer partner with a record of such conversations | Patterns only |
| 6. Not yet settled | "Does introducing one subsidiary affect credit for its sister company?" | A decision nobody has published | The dated facts, kept for when SourceX reviews them | Logged as open; SourceX decides in review | Only the word "open" |
Family 3 deserves its own warning. The terms say partners never receive access to other partners' records, and credit is decided from dated records, including earlier applications and relationships that only SourceX can see. So the honest reply has two halves: what your own claim log shows inside your network, stated without naming who logged it, and that SourceX alone decides credit. When two partners collide, seven evidence checks for one company and two partners sets out what each should keep.
Family 6 currently holds three known questions: credit when a parent or sister company was already in contact or already introduced, what happens if a partner turns a reward down, and introductions still in review when a partner leaves the program. None is answered in the terms. The right reply names the question as open, records the facts, and leaves the decision to SourceX, which settles related-company credit when it reviews duplicates.
Reusable answers need an environment line and a version
The Consortium for Service Innovation's KCS method gives a shape for the answers you do reuse. Its article structure, paraphrased, records the issue in the requester's words, the environment, documented precisely and including versions, the resolution, and a cause where one applies. The environment is the field support teams skip, and the one that makes an answer specific.
For a referral program, the environment of a family 1 answer has three parts: the terms version it rests on, whether the partner is on the standard terms or a signed Schedule A or addendum (the terms say a signed Schedule A prevails for the reward rate and base), and the date the answer was checked. Versions matter because the terms changed five times in October 2026 and again on 4 November 2026, when Version 2026-11-04 removed the cap.
| Field | Example: the reward question |
|---|---|
| Issue, in the partner's words | "If the company gets paid USD 200,000, what do I get?" |
| Environment | Standard terms, Version 2026-11-04; no signed addendum; checked 4 November 2026 |
| Resolution | the greater of $50,000 plus 5% and 15% of Actual Supplier Payout, with no cap, so USD 60,000 on a first qualifying transaction of USD 200,000 (Illustrative); the USD 50,000 floor applies once per referred company, and repeats within 12 months pay 15%; paid separately by SourceX and never deducted from the company's payout |
| Limits | Payable only after the conditions in section 5 of the terms; a lead, meeting or signature alone triggers nothing |
| Expiry | Re-check when a new terms version is published or the partner signs an addendum |
Family 2 answers get a different treatment. Keep the rule reusable and make the verdict depend on a short intake: the five facts above, each marked pass, unknown or fail with a date, collected from a conversation with the relationship holder and never from files, samples or logins. The reply then says "may be a fit; SourceX decides after review" or names the check that fails, never "qualifies". Which eligibility questions you can answer at all is sorted in the three-owner eligibility table.
One quarter of partner questions, with a struggling cohort and a claim query (Illustrative)
Illustrative: an invented bench of 40 partners sent 96 questions in a quarter. One partner manager had three hours a week for questions, about 39 hours in the quarter. Before routing, every question landed in that inbox and took the time shown in the "before" column, about 52 hours in total, so replies slowed and grew shorter. Shorter replies were the generic ones.
| Family | Questions | Answered by | Minutes each, before | Minutes each, after | Hours after |
|---|---|---|---|---|---|
| 1. Program mechanics | 30 | Versioned answer library | 20 | 5 | 2.5 |
| 2. Company fit | 34 | Screener, with the five-fact intake | 45 | 25 | 14.2 |
| 3. Claims and duplicates | 9 | Own log, then SourceX | 30 | 10 | 1.5 |
| 4. Partner's own situation | 8 | Partner's adviser | 25 | 5 | 0.7 |
| 5. Conversation craft | 10 | Peer partner | 40 | 10 | 1.7 |
| 6. Not yet settled | 5 | Open log | 30 | 10 | 0.8 |
After routing, the partner manager's share fell to about 21 hours, inside the 39 available. Two cases explain most of the gain.
The underperforming cohort. Ten software implementation partners sent 21 of the 34 company-fit questions, 17 with none of the five facts attached, each needing two or three emails. Requiring a headcount at peak and a records start year on the question form cut the back-and-forth, and showed that most of the cohort's companies kept records that belonged to their clients: a criteria lesson, not a support one.
The duplicate-risk case. A partner asked whether a 300-person logistics group had already been introduced. Your log showed that another partner in the network had logged one of its subsidiaries a month earlier. The reply said the group was already logged in the network and should not be approached separately, did not name the other partner, and noted that only SourceX's records show any earlier dealings with SourceX itself. Whether the earlier subsidiary introduction affects credit for the group is one of the open questions above, so it went into the open log with both partners' dated evidence.
The capacity limit. Peer partners are a capacity too. The two partners who took conversation-craft questions agreed to at most two calls each a month; when a third request arrived, it waited for the next month rather than going to someone untested.
Expert hours, peer load and colliding claims
Read the routing table monthly alongside four measures. Each one tells you whether support is scaling or just moving.
| Measure | Where it comes from | Warning sign (Illustrative) | Response |
|---|---|---|---|
| Expert hours per active partner | Time logged against families 2, 3 and 6 | Rising two months running | Tighten the intake before recruiting |
| Fit questions arriving with all five facts | The intake form | Below 80% | Make the fields mandatory |
| Questions from the top three askers | Question log | Above half of all questions | Check whether they are your most active partners or your most confused |
| Claim questions per 100 | Family 3 count | Above 10 | Overlapping circles; make the claim-log lookup a step before outreach |
Heavy askers can be your best partners, working many live companies, or partners applying rules they misread. The families tell you which: the first group asks family 2 and 5 questions about named CEOs, the second repeats family 1 questions.
Gates before the bench grows, and how a wrong answer gets corrected
Recruit more partners only when support can absorb them without falling back to generic replies. These gates are Illustrative; set your own in advance.
- Every library answer shows a terms version no older than the current one.
- At least 80% of fit questions arrive with all five facts.
- Expert hours used stay under 70% of hours available for two months.
- No family 3 or family 6 question has been answered as if it were family 1.
Answers will sometimes be wrong. Log which partners received which library answer; when one changes, replace it with a dated note of what changed and why, and send the correction to everyone who had the old version. Let partners flag answers they believe are wrong; a flag from a partner holding a signed addendum often means the environment line was missing. For questions that founders, rather than partners, bring to a platform team, the founder-question backlog applies the same idea to a different asker.
Where routing by family breaks down
- Mixed questions. "Does this company qualify, and what would I earn?" spans families 2 and 1. Split it and answer the fit half first; the earnings half has no honest answer until a payout exists.
- The library can drift from the source. Paraphrased terms age. Link the program terms and the FAQ in every library answer so partners can check the original.
- Peers can be confidently wrong. A partner with many conversations may carry habits from an older terms version. Give peers the current library before they advise anyone.
- Routing can feel like deflection. Sending a tax question to the partner's adviser is correct, but say why: you do not hold the facts that decide it.
- Open questions can sit forever. Review the open log monthly and send SourceX the facts for any item a live introduction depends on.
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 partner's question has been answered with that partner's own facts, point them to partner registration and the qualification criteria, and put the program terms version on every answer you send. For what the partner role covers, see what a SourceX referral partner is, and for more playbooks, the guides hub.