# Referral workshop outcome test: now or later | SourceX

Canonical: https://refer.sourcex.si/guides/how-to-test-a-workshop-s-effect-on-actual-introductions
Last updated: 2026-11-04

Draw everyone you would invite into two groups before sending a single invitation: a now group that attends the workshop at once and a later group that attends after a 90-day counting window. Then compare the groups on introductions whose CEO agreed in writing and that reached a verified application SourceX credited, read on day 180. Compare by invitation, never by attendance, because the people who show up are the ones most likely to introduce anyway. No results from this design are published.

This is for the person who runs referral enablement at a firm or network: an advisory practice briefing its client partners, an association briefing chapter members, a fund briefing operating advisors. Every introduction still goes through someone who has signed the SourceX Channel Partner Agreement, and each one names a single company whose CEO agreed to it.

## Turnout measures interest; the test measures what changed

Workshop reports usually count registrations, attendance and a satisfaction score. None of them says whether anyone later asked a CEO a well-judged question. Kirkpatrick Partners' model calls that Level 3, Behavior: whether participants apply what they learned once back at work, and it ties behavior to required drivers such as job aids, coaching and work review rather than to the event alone (paraphrase). An introduction is a behavior, and a verified application is the closest result a partner can observe.

The trap is self-selection. The colleague who books the first seat already has a CEO in mind. If you compare attendees with people who skipped, the workshop gets credit for that colleague's prior intent. Drawing groups before anyone is invited, and keeping every invitee in the group they were drawn into, removes that problem; the [event-quality funnel](/guides/how-to-measure-event-quality-beyond-registrations) covers what an event can show on the day itself.

## What the workshop may teach, and the promises it must leave out

Both groups end up seeing the same session, so its content must be right before the draw.

| The session may | The session must never |
|---|---|
| Walk through [who qualifies](/who-qualifies) on screen and link it | Tell anyone that a named company qualifies |
| Say the CEO decides whether to talk and signs for the company | Suggest the network, a board seat or the partner can decide for a company |
| Explain that SourceX decides fit and credit from dated records | Say that sending a link secures credit |
| Say the partner may earn a reward from SourceX, disclosed to each CEO | Present the reward as income anyone can count on |
| Show how to describe records by type, years and system | Ask attendees to gather files, samples, exports or logins |
| Say nobody gets preferential access to a company's data | Imply that SourceX endorses the host or its members |
| Invite attendees to raise SourceX only where it fits a real conversation | Set targets, quotas or a "bring one company" ask |

Two clauses of the sample Channel Partner Agreement shape the invitation. Section 4.5 bars either party from publicly identifying the other as a partner without written authorization, so do not bill the session as a SourceX partnership event unless SourceX has agreed in writing. Section 7.1 says the partner is not SourceX's agent and may not commit SourceX to prices, licenses or payouts, so no slide may estimate what a company would be paid.

## Now group, later group: the draw and the clock

| Element | Specification |
|---|---|
| Decision | Keep running the workshop, rework it, or stop |
| Hypothesis | The now group makes more qualified introductions per 10 invitees in days 1 to 90 than the later group |
| Unit | One invited person who could raise SourceX with a CEO they know |
| Allocation | A logged draw before any invitation, done separately within each role list so both groups get a similar mix |
| Now group | Attends on day 0 |
| Later group | Keeps the standard links and attends on day 91 or soon after |
| Counting window | Introductions submitted from day 1 to day 90 |
| Read date | Day 180, so an introduction made on day 90 has a full 90-day attribution window |
| Analysis | By drawn group, whoever attended |

Ask now-group attendees not to forward slides or notes until the later session. Some will anyway; the spillover question below measures it.

## The workshop follow-through evaluation

Each stage has one definition, one piece of evidence and one thing that never counts. Copy SourceX's stage and status words exactly as the dashboard shows them; [how it works](/how-it-works) lists the statuses as Estimated, Eligible, Approved and Paid.

| Stage | Counts when | Evidence | Never counts |
|---|---|---|---|
| Invited | The person was on a list before the draw | The logged draw | People added afterwards |
| Attended | Present for most of the session | Attendance list | Registration alone |
| Asked a CEO | The invitee says they raised SourceX with a CEO | An optional one-line report at day 90 | A forwarded newsletter |
| CEO agreed | The CEO said yes in writing before any submission | The dated email or note | A verbal "sounds interesting" |
| Introduction submitted | One company, submitted by a signed partner | The firm's referral log or the partner's dashboard | A link click or a page visit |
| Qualified introduction (primary) | SourceX verified the application, credited it to the introduction and has not closed it by day 180 | Dashboard stage | An application still unverified |
| Closed or duplicate | SourceX closed it, or found the company already known | Dashboard stage | Your own guess about fit |

The report is one table, by group, per 10 invitees, with every column filled by count first and rate second. Until a run finishes, every cell reads "not yet collected"; nothing is estimated in advance.

| Count at day 180 | Now group | Later group |
|---|---|---|
| Invited | Not yet collected | Not yet collected |
| Attended (the later group, after day 90) | Not yet collected | Not yet collected |
| CEO agreed in writing | Not yet collected | Not yet collected |
| Introductions submitted | Not yet collected | Not yet collected |
| Qualified introductions | Not yet collected | Not yet collected |
| Closed or duplicate | Not yet collected | Not yet collected |
| Side-effect count (next section) | Not yet collected | Not yet collected |

Where members are partners in their own right, you cannot see their dashboards. Ask for counts and copied stage words at day 180, record who did not answer, and report non-response beside the counts rather than filling the gaps.

## Side effects worth more attention than the headline

| Unwanted effect | Signal | Target |
|---|---|---|
| A CEO learned of the introduction after SourceX did | The CEO's own reply, or SourceX telling you | Zero |
| Pressure | A CEO complains about being pushed or asked repeatedly | Zero |
| A promise | A CEO says they were told the company qualifies, or that SourceX vouches for the host | Zero |
| Records offered | An attendee receives or forwards files, samples or logins | Zero |
| Duplicates | Introductions of companies already known to SourceX | Similar in both groups; a jump in the now group means attendees rushed familiar names |
| Spillover | Later-group invitees who saw the slides before day 91 | Reported, so the gap can be read as a floor |

Information an attendee holds through a board or advisory seat may prompt one question to that company's CEO and nothing more. If the session gives anyone the idea of using board knowledge to suggest a different company, stop and rewrite it.

## Floors, length and stop rules

Introductions are rare, so small groups decide nothing. Publish per-10 rates only when each group holds at least 20 invitees; below that, publish counts. Several runs of the same session, same draw rule and same read date may be pooled.

**Stop rules.**

- An introduction reaches SourceX before its CEO agreed: pause new submissions from both groups and fix the consent part of the session.
- Anyone collects or forwards company records because of the session: stop, and fix that slide before the later group attends.
- A CEO reports pressure or a promise: pause and rewrite the session before day 91.
- SourceX changes the terms or who-qualifies before the later session: teach the current version, note the change and treat the two groups as different versions.
- Never end the count early because the now group looks ahead. Run to day 180.

## Reading the difference between the groups

Before the draw, write down the smallest gap that would change your plans, for example one more qualified introduction per 10 invitees, and state uncertainty as a range.

| Result at day 180 | Decision |
|---|---|
| Now group ahead on qualified introductions by more than your gap, side effects at target | Keep the workshop and run it for each new cohort |
| Now group ahead on introductions submitted, level on qualified ones | The session drives volume, not fit: rework the fit and consent parts |
| No clear gap either way | The session is not what moves introductions; test a lighter driver, such as a one-page reminder or a short follow-up call |
| Any side effect above target | Fix the session first; the comparison waits |

The workshop's knowledge effect can be tested separately with [the onboarding scenario cards](/guides/the-partner-onboarding-experiment-that-measures-understanding); this design deliberately measures only what people did.

## Where a workshop comparison misleads

- **Small numbers.** One busy introducer can swing a group. Name outliers in the report instead of deleting them.
- **Spillover.** Shared slides shrink the gap, so a small difference may understate the session's effect.
- **Facilitator effects.** A charismatic presenter is part of the treatment. Running the later session with the same person keeps it constant.
- **Self-reported asks.** The "asked a CEO" stage relies on invitees' word; the primary measure does not.
- **Review is SourceX's.** A qualified introduction is a verified application still open, not an accepted company or a payable reward; a lead, meeting or signed agreement alone earns nothing payable. For running the session itself, [the chamber workshop design](/guides/the-chamber-workshop-that-starts-with-company-questions) shows a question-first format.

## 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 the draw, [register as a partner](/register) and sign the Channel Partner Agreement, build the session's fit slide from the current [who qualifies](/who-qualifies) page and record the [program terms](/terms) version both groups were taught. More test designs are collected in the [guides hub](/guides).

## Sources

- https://refer.sourcex.si/terms
- https://refer.sourcex.si/who-qualifies
- https://sourcex.si/documents/partner/SourceX_Channel_Partner_Agreement_Sample.pdf
- https://www.kirkpatrickpartners.com/wp-content/uploads/2024/04/an-introduction-to-the-new-world-kirkpatrick-model-a-white-paper.pdf
