Mapping referral friction from a CEO's yes to a submitted application: a twelve-code friction taxonomy and a consented journey study, with stages, logs and reporting rules set before any company is followed

Short answer

Friction research follows companies from the moment a CEO agrees to a conversation until a verified application, or until the company stops, and gives every obstacle a stage, a family and a verdict on whether it should be removed at all. SourceX has published no such study, and this page reports no results. It sets out the design: six stages defined by dated events, twelve friction codes in six families, a short journey log kept with consent by the company's own contact, an exit interview for stalled cases, and analysis rules that keep every stopped or missing company in the count.

Visual briefFull-size diagram
A path of six stations from CEO's yes through first look, inside check, started and submitted to verified; a pin dropped between inside check and started opens an Illustrative log entry reading day 19, inside check, rights doubt, code R1, verdict: protective, beside a legend of protective: keep and avoidable: fix and a box of stops that stay in the count: paused, stopped, out of fit and silent; six family chips run along the bottom: understanding, effort, authority, rights, trust and context.
A study design with one Illustrative log entry. No study has run and no results are published.

The original brief called this page "the referral friction map". No map exists yet; what follows is the method that would draw one.

Why the study starts at a yes and stops at day 90

Messages that never earn a reply belong to a different instrument, built for owners who left an offer unanswered. This study begins where interest is on the record: a CEO's written agreement to a conversation, followed by the partner's dated introduction. Starting there makes the denominator countable and consent straightforward, because every company in it has already said it wants to hear more.

It ends at a verified application or at day 90 after the introduction, whichever comes first. Ninety days is the window SourceX's terms use: the first valid introduction "receives credit when that company submits a verified application", within the attribution window (section 2). One last read at day 120 records any later application as an outcome for the company, though it would not be credited.

Six stages, each opened by a dated event

StageOpens whenRecorded by
S0 YesThe CEO agrees in writing and the partner submits the introductionPartner log
S1 First lookThe company's contact first reads SourceX's public pages or writes to SourceXJourney log
S2 Inside checkThe CEO brings in anyone else: finance, operations, counsel, the boardJourney log
S3 StartedThe company begins SourceX's applicationJourney log
S4 SubmittedThe company submits the applicationJourney log
S5 VerifiedSourceX verifies the applicationThe partner dashboard's stage, or the company's report

Four endings other than S5 stay in every count: paused (the contact says the company will resume), stopped (the company decides not to go on), out of fit (the company concludes it falls outside SourceX's screen) and silent (no entries or replies for 30 days). Out of fit is not friction. A company that finds it is too small, or that its records belong to its clients, has reached a correct answer early.

A link click is never a stage. It shows nothing about what a company decided, and SourceX's terms say a click alone does not decide credit.

The friction taxonomy: six families, twelve codes

CodeFamilyWhat the contact describesDefault verdictWho could act on it
U1UnderstandingUnclear what SourceX does or what the company is being asked to doAvoidableThe partner's wording; SourceX's pages
U2UnderstandingUnclear who decides, who pays whom, or whether the partner's reward reduces the company's payoutAvoidableThe partner, quoting the terms
E1EffortTime spent assembling facts the application asks for, such as peak headcount or years of recordsPartly protectiveThe company
E2EffortSteps, logins or forms that slowed the contactAvoidableSourceX
A1AuthorityThe CEO was unavailable, or the contact was unsure whether to proceedProtective while the CEO has not decidedThe company
A2AuthorityA control question: a parent, an acquirer, a board or a formal processProtectiveThe company and its counsel
R1RightsDoubt whether records are the company's own or its clients'ProtectiveThe company's counsel; SourceX's review
R2RightsConfidentiality or privacy concerns about the recordsProtectiveThe company's counsel; SourceX's review
T1TrustDoubt about SourceX or about the partner's motiveAvoidable when it comes from missing informationThe partner; SourceX
T2TrustConcern about how licensed data would be usedProtectiveSourceX's published materials
C1ContextA competing event: an audit, a migration, a sale, a busy seasonRecord onlyNobody
C2ContextChanged circumstances: the contact left, the company was acquiredRecord onlyNobody

The verdict column is the point of the taxonomy. Protective friction does work that ought to happen: counsel checking whether service records contain client material, or a CEO taking a week to decide. Removing it to lift a completion figure would produce applications that fail in review or should never have been filed. Avoidable friction is confusion or effort that serves nobody, such as a finance lead unsure whether the partner's reward comes out of the company's payment, when the terms answer that in one sentence: the reward "is paid separately by SourceX and is never deducted from the supplier's agreed compensation." Two coders assign the verdict case by case; the default is only a starting point.

The journey study protocol

InstrumentCompleted byWhenNever collected
Partner logThe introducing partnerAt the introduction and whenever a status changesAnything from the company's records
Journey logThe company's named contact, with the CEO's consentA two-minute entry at each stage event, plus a weekly prompt for 90 daysRecords, samples, system contents, figures
Exit interviewThe contact or the CEOWithin two weeks of a stop, or at day 90Anything the company has not chosen to say
Stage datesSourceX, if it runs the studyFrom its own dated records, reported in aggregateOther partners' records

A journey log entry has six fields: the date, the stage, what happened in one sentence, any obstacle in the contact's own words, the roles involved, and whether it was resolved. The format follows the diary study, which Nielsen Norman Group describes as a qualitative way to gather behavior over time and in context, with participants reporting as things happen over days to a month or more (our paraphrase). Here the diary covers one decision for 90 days.

The exit interview asks five questions, in order: "Walk me through what happened after your CEO agreed to talk." "Where did it slow down?" "Was there a point where you nearly stopped, and what decided it?" "What did you need that you didn't have?" "Is there anything you would want the person who introduced you to do differently?"

Enrollment and consent. Every introduction made by participating partners during one enrollment quarter is eligible. The CEO agrees for the company and the contact agrees personally; either may stop at any time; partners see only aggregate results; and taking part has no bearing on SourceX's review. Companies that decline to be studied are counted, never followed.

Measures, with every company kept in the denominator

  • Completion. The GOV.UK Service Manual divides completed transactions by all transactions started, partly completed and failed ones included (our paraphrase). Here: verified applications within 90 days, divided by every company enrolled at S0, including the paused, stopped, out-of-fit, silent and withdrawn.
  • Stage progression. The share moving from each stage to the next, with every company that entered the stage in the denominator.
  • Time between stages. Companies still inside a stage at day 90 are counted as open at that point, never dropped, as in the stage-duration benchmark.
  • Friction. Codes by stage, by family and by verdict, and how often each appears among companies that reached S5 compared with those that stopped.

Reporting rules. Every table shows the number enrolled. Cells with fewer than five companies are folded into "other". Anything not in the dated pre-registration is labeled exploratory. A code reported more often by companies that stopped is associated with stopping, nothing more: stoppers may simply have more to say. A causal claim needs a test that changes one thing at a time.

One journey logged, one that stopped (Illustrative)

Illustrative. Generic companies and invented days; no outcome is implied for any real company.

Case 1, a 210-person facilities services company.

DayStageLog entryCode and verdict
0S0 YesThe CEO agrees; the partner submits the introductionNone
5S1 First lookThe contact reads SourceX's terms and screening pageNone
8S2 Inside checkThe finance lead asks whether the partner's reward comes out of the company's paymentU2, avoidable
19S2 Inside checkCounsel asks whether service records include client materialR1, protective
37S3 StartedApplication begunNone
41S3 StartedThe contact is unsure whether contractors count toward headcountU1, avoidable
49S4 SubmittedApplication submittedNone
60S5 VerifiedVerification shown on the dashboardNone

Two of the three codes were avoidable and could be answered from published pages: SourceX's terms on who pays, and its screening page, which excludes contractors from the headcount. The rights question was protective and took eleven days; a partner should expect it, not shorten it.

Case 2, a 160-person distributor. The CEO agrees on day 0 and the contact reads SourceX's pages on day 12. On day 23 the owner announces a sale process (C1), and the contact notes that the buyer's CEO will decide once the deal closes (A2, protective), which matches SourceX's screening page. The company stops on day 30. The exit interview confirms nothing in the introduction would have changed the outcome.

What a partner could change because of a result

Each row is conditional. No result exists.

If the study findsA partner mightA partner should never
Avoidable understanding codes clustered at S1 and S2Rewrite the introduction and resource page, then test the rewriteAnswer with figures or promises about review
Avoidable effort codes at S3Pass the pattern to SourceXFill in or prepare the application for the company
Protective rights or authority codesSay early that rights are reviewed and that the CEO decidesTreat them as obstacles to argue away
Context codesWait, and note the date to revisitChase a company that stopped for its own reasons

Testing the application path before sending more companies to it is a separate exercise, set out in testing the referral form before sending more traffic.

Which companies enter the study

Only companies that already passed the partner's screen against SourceX's published profile: at least 50 full-time staff at their high point, contractors left out; years of records they produced themselves; the right to license that material; and a CEO who asked for the conversation. Whether the records match current demand, and whether customer contracts allow licensing, stay unknown until SourceX's review, and the study does not try to answer either. The partner never sees the company's records: the how it works page says partners need not export, upload or describe confidential material, and the journey log describes events, never contents. Partners who enroll companies have signed the Channel Partner Agreement before making any introduction.

Where the friction study can mislead

  • Logging changes the journey. A contact who writes entries may move faster than one who does not.
  • Missed weeks. Weekly prompts reduce gaps; missing weeks are shown as missing, never filled.
  • Who agrees to be studied. CEOs willing to be followed may differ from those who decline, so the decline count is published beside the results.
  • One process, one period. If SourceX's application changes during enrollment, the change is dated and results are split at that date.
  • Context that wording cannot touch. In case 2 the company stopped for a sale; counting it as friction to fix would mislead.

How this study joins a yearly census of outcomes is described in the annual research blueprint, and what may be cited about results in the meantime in the partner results research guide. More designs are in the guides hub.

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.

If your next introductions are the ones you would study, screen each company against who qualifies, read the program terms so you can answer who pays whom in one sentence, and register as a partner before the first CEO's yes.

  1. Step 1Share your linkSend your personal link to a company you know.
  2. Step 2Company appliesThe company applies itself at /apply.
  3. Step 3Company signs and is paidThe company signs, delivers its data and is paid by SourceX under its agreement.
  4. Step 4You get your rewardthe greater of $50,000 plus 5% and 15% of the actual supplier payout, with no cap. A reward becomes payable after SourceX has paid the referred company under its signed agreement and the payout conditions in the program terms are met, including finance approval and your tax and payment details.

Common questions

Should a partner try to remove every obstacle the study finds?

No. Some friction does necessary work: counsel reading a client contract, or a CEO taking time to decide. The taxonomy marks each obstacle protective or avoidable, and only avoidable friction is a target. Removing protective friction would produce applications that fail in review or should never have been made.

Why does the study start at a CEO's yes rather than at the first message?

Because a yes is a dated, consented starting point, so every company in the count can be followed and asked. What happens to messages that never earn a yes needs a different instrument, built for owners who did not reply.

Do the journey logs ask companies about their records?

No. A log entry records a date, a stage, what happened in one sentence, any obstacle in the contact's words, the roles involved and whether it was resolved. Records, samples, system contents and figures are never asked for.

Sources

  1. refer.sourcex.si/terms
  2. refer.sourcex.si/who-qualifies
  3. refer.sourcex.si/how-it-works
  4. gov.uk/service-manual/measuring-success/measuring-c…
  5. nngroup.com/articles/diary-studies

Free resources

Know a company with valuable proprietary data?

Become a referral partner, get your link and introduce the company’s CEO (in an owner-run company, usually the owner). Referral partners can join from most countries. We can't work with partners in countries or regions under U.S. sanctions or on U.S. restricted-party lists. The companies you introduce should be based in the U.S., Canada and the EU, or in the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia with at least 70% of their data in English.

Refer a company →

I own a business

Explore licensing your company's data directly to SourceX for AI training and evaluation. Start a short assessment; no uploads needed.

Start an assessment