When one large introduction distorts the entire dashboard: put the top company on its own line, print its share, and read every other company as a count

Short answer

When one introduced company produces most of a fund's paid or estimated referral value, every total, average and per-partner figure on the dashboard describes that company rather than the channel. Put the largest company on its own line, print the share of the total it carries, and read the rest of the cohort as counts of unique companies by stage, with open cases listed. The program has no cap on rewards, so nothing limits how large one row can grow, and one row is often most of the total.

Visual briefFull-size diagram
A single bar of paid referral rewards in which one company fills 87%, split underneath into a top-company card at $100,000 and a card for the other eight companies, with the HHI of 7,656 marked.
Illustrative cohort. Nine invented companies; shares and the HHI follow the method on this page.

This matters most to a fund COO taking a referral channel to a staffing or budget discussion. A dashboard reading "$70,000 paid" invites the question "how many more like that?", and the honest answer is usually "we do not know yet, because one company produced almost all of it."

Why one company can swallow a referral dashboard

Five features of the program terms decide how a dashboard behaves when one company is much larger than the rest. None of them is a flaw; each just needs its own column.

MechanismWhat the terms sayWhat it does to a dashboard
Proportional baseOn a first qualifying transaction the reward is the greater of $50,000 plus 5% and 15% of the Actual Supplier Payout, the compensation SourceX actually pays the referred companyA company paid several times more produces a several times larger reward, so one large payout dominates any sum
Floor, minimum and repeatsThe USD 50,000 element applies once per referred company, on its first qualifying transaction; a transaction below the USD 100,000 Minimum Transaction earns nothing unless SourceX agrees in writing; repeats first agreed within 12 months pay 15%, and there is no capThe top row has no ceiling, so one large payout can be most of the total, while small payouts drop out of it entirely
EstimatesBefore the company is paid, any reward shown is an estimate based on proposed compensation; portal estimates are projections, not promisesOne large estimate added to paid figures can double a "potential" total overnight
InstallmentsWhere the company is paid in installments, the reward accrues as each installment is paidOne large company can show up as paid in several quarters and look like several wins
ReductionsIf the company's payout is reduced or reversed, the reward is reduced proportionatelyThe top row can shrink later and pull the whole total down with it

The terms treat a parent company and all of its subsidiaries as a single referred company, so the USD 50,000 floor applies to the group once. Fold group entities into one row before measuring anything, or a group introduced as three names will look like three moderate results instead of one large one.

Population, value measures and the reporting period

Population. Every unique company introduced in the cohort period with its CEO's agreement, counted once per group. Companies SourceX already knew are listed as not credited, never dropped, because the terms treat any earlier lead, account or email with SourceX as an existing relationship.

Reporting period. Cohorts by introduction quarter, read at a stated cutoff date. A cumulative total since launch is a different report and must say so.

Value measures. Only one of them is money the fund has received.

MeasureUnitWhere it comes fromRule on this dashboard
Companies introducedCount of unique companiesThe fund's introduction recordsThe denominator for every rate
Companies creditedCountPartner dashboard: first valid introduction, verified application within 90 daysDuplicates and lapsed windows shown as their own rows
Companies whose payout startedCountReward status past EstimatedCounted, never valued in the headline
Rewards paidDollars, by companyStatus Paid on the rewards page, matched to the bank recordThe only dollar figure that enters shares or the HHI
Rewards estimatedDollars, by companyStatus EstimatedKept off the dashboard's totals entirely

The concentration-adjusted result view

The view has four blocks, always printed in this order. Counts come first because they are the part a single company cannot distort.

  1. Counts by stage. Unique companies introduced, credited, in review, closed with a reason and still open, each with its denominator.
  2. The top company on its own line. A code, its stage, its paid reward, whether a repeat within the 12-month Repeat Window is still possible, and its share of all paid rewards.
  3. Everyone else. The same measures for the rest of the cohort, so a reader sees what the channel produced without its largest row.
  4. Concentration figures. Top-company share, top-three share, the Herfindahl-Hirschman Index (HHI) computed on paid rewards per company, and the effective number of paying companies, which is 10,000 divided by the HHI.

The HHI comes from antitrust work. The Justice Department's Antitrust Division computes it by squaring each firm's share of a market and adding the squares: it approaches zero when many similar-sized firms share a market and reaches 10,000 when one firm holds everything. This page borrows only the arithmetic. The Division's own bands (1,000 to 1,800 moderately concentrated, above 1,800 highly concentrated) describe markets, not a cohort of nine companies, and should not be quoted as thresholds for a referral channel.

Our trigger rule. When the top company holds more than half of paid rewards, or the effective number of paying companies is below three, the dashboard leads with blocks 2 and 3 side by side and drops every average. That rule is our editorial choice, not a SourceX requirement; set your own before the first quarter, not after a large result arrives.

Nine companies, one large row (Illustrative)

Illustrative. One fund's introductions over four quarters, read at a cutoff date. Generic companies, invented amounts, no real deals implied.

CodeCompany (Illustrative)Stage at cutoffPaid to the companyReward paidShare of rewards paid
A900-person claims administratorPaid in full in two installments$400,000$70,000100%
B160-person freight brokerPaid; below the Minimum Transaction$40,000$00%
C120-person engineering consultancyPaid; below the Minimum Transaction$20,000$00%
D300-person contact center operatorSigned, not yet accepted; status Estimated$0$0Excluded
E210-person field-services group, two subsidiaries introduced as one companyIn review$0$0Not applicable
F140-person property managerIn review$0$0Not applicable
G180-person software resellerDeclined in review$0$0Not applicable
H260-person insurance agency groupNot credited: an earlier email with SourceX$0$0Not applicable
I110-person dental billing serviceIntroduced; no application yet, window open$0$0Not applicable

Company A's first qualifying transaction pays the greater of USD 50,000 plus 5% of $400,000 and 15% of $400,000, which is $70,000. B's $40,000 and C's $20,000 fall below the USD 100,000 Minimum Transaction, so neither earns a reward without a written exception from SourceX, and none applies here. Company D's dashboard shows an estimate on a large proposed amount, and it stays out of every figure below.

What the unadjusted dashboard said. "Rewards paid: $70,000. Average per paid company: $23,333. Average per credited company: $10,000." Every one of those numbers is company A.

What the concentration-adjusted view says.

BlockCompaniesReward paidWhat a reader learns
Top company (A)1$70,000, from one first qualifying transactionOne outcome, which could still grow: a repeat agreed within 12 months pays 15%, with no cap
Every other company8$0; B and C were paid below the Minimum TransactionWhat the channel produced without its largest row
Still open at cutoff4: D, E, F, I$0Cases that can still move either line
Closed without reward4: B and C below the minimum, G declined, H not credited$0Each with its recorded reason
Concentration figureValueHow it was computed
Top-company share100%$70,000 of $70,000 paid
Top-three share100%Only one company has been paid a reward
HHI on paid rewards10,000100², one paying company
Effective number of paying companies1.010,000 ÷ 10,000

An effective number of 1.0 says the paid result is one company. That is the sentence the committee needs, and no average can say it.

Counting companies once before measuring concentration

Concentration figures are only as good as the rows beneath them. Four reconciliations come first.

  • Groups. E's two subsidiaries were logged separately by two colleagues. They are one company for the once-per-company floor, so they are one row here. The terms do not say whether another partner's introduction of a sister entity, or a parent's earlier contact with SourceX, changes credit for E; SourceX settles that in duplicate review, and until then the row reads "pending SourceX's decision".
  • Colleagues introducing the same company. Only one partner is credited per company. If two people at the fund introduced A, it is still one row with one credited account.
  • Known-to-SourceX companies. H had an earlier email exchange with SourceX, which the terms count as an existing relationship. It stays on the page as not credited, with that reason.
  • Installments across periods. A was paid in two installments in different quarters. In a cohort view it is one row; in a period view it appears in two quarters, and the period report must say that both lines are the same company.

Every open row carries the next event that resolves it: D waits on the Acquisition Partner's acceptance, E and F on SourceX's review, and I on an application before its 90-day window closes. The deduplicated pipeline waterfall covers the counting rules in depth, and the VC referral dashboard guide defines each stage's denominator.

The committee note when one row is most of the result

Write the note so that it would still be true if company A had never been introduced.

  • Lead line: "Nine companies introduced; three paid, one rewarded; four open. One company produced 100% of rewards paid; the other two paid companies fell below the USD 100,000 Minimum Transaction."
  • Without the top company: "The other eight produced no reward: two were paid below the Minimum Transaction, four are open and two closed with reasons."
  • What we learned: which ring of the network produced A, how long it took, and whether anything about A's path can be repeated.
  • What we do not claim: no reward per introduction, no yield, no run rate. A paid $70,000 once; multiplying it by four, or by the number of open rows, describes nothing that has happened.
  • What changes next quarter: the trigger rule stays as written, and D's estimate stays off the page until D's payout is paid.

The stage-to-cash reasoning behind keeping signed and estimated amounts out of income lines is set out in why signed agreements are not closed referral revenue.

Where the adjusted view can still mislead

  • Tiny denominators. With one paying company, a second payment swings the HHI by thousands of points. Read the figure as a description of this cohort, never as a trend.
  • Heavy tails may be the business. If a few large companies are what the channel is for, a view that dwells on "everyone else" can understate it. That is why block 2 is printed, not removed.
  • A second kind of concentration. One LP or one board member may have produced A. Measure concentration by relationship owner as well as by company; a channel that depends on one person's network is fragile even when results spread across companies.
  • Reductions. If A's payout is later reduced, the reward is reduced proportionately, and the top share changes with it.
  • Attribution practice is unsettled elsewhere too. In a PartnerStack and Wynter survey of 100 senior B2B SaaS leaders, 14% named poor visibility into partner impact as a blocker and only 42% reported multi-touch attribution. Large software companies with dedicated partner teams still struggle to see what partners produce; a small fund dashboard should be modest about what it can show.

Methods note. The view, the trigger rule and the nine-company cohort were built for this page from the SourceX program terms (Version 2026-11-04), the rewards page's status definitions and the Antitrust Division's HHI arithmetic, all checked 2026-11-05. It measures paid rewards and stage counts only; it does not measure record quality, demand or what any open company will be paid. SourceX has not published the distribution of supplier payouts or rewards across companies, and those figures are not yet published.

Which companies can produce a large row at all

A large row starts with a company that fits and a CEO who wants the conversation. Introduced companies need 50+ full-time employees at peak (contractors excluded), years of records they created themselves, the rights to license them and a CEO who authorizes the discussion and signs for the company. 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. Funding status plays no part.

What nobody at the fund can know in advance is how large a company's payout will be: that depends on SourceX's review of demand, the rights picture in customer contracts and the scope the company chooses to license. Partners screen at metadata level only, asking which systems hold which records, for how many years and who administers them, and never handle files, samples or logins. Platform teams starting the channel can begin with the venture platform overview or browse 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.

Once your dashboard can show its largest company on its own line and every other company as a count, register for a partner link, screen the next introduction against who qualifies and keep the program terms beside the report.

Building an honest referral dashboard is fund operations work; assessing SourceX as a possible investment is a different decision, and readers weighing it can request materials through the SourceX investor overview.

This article is for information only and is not an offer to sell or a solicitation of an offer to buy securities. Any offering by SourceX is made only to verified accredited investors through its offering documents.

  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 fund drop its largest referred company from the dashboard?

No. Report it on its own line with its share of the total, and show the rest of the cohort beside it. Hiding the top company understates the channel as badly as averaging it overstates the channel.

Does the once-per-company floor fix concentration?

No. The USD 50,000 element applies once per referred company, on its first qualifying transaction, and there is no cap above it. Nothing stops one company from being most of a small cohort's paid total, which is the usual case.

Can estimated rewards go into the concentration figures?

No. Before the referred company is paid, any reward shown is an estimate based on proposed compensation, and the terms call portal estimates projections. Compute shares and the HHI on paid rewards only.

Sources

  1. refer.sourcex.si/terms
  2. refer.sourcex.si/rewards
  3. refer.sourcex.si/who-qualifies
  4. justice.gov/atr/herfindahl-hirschman-index
  5. partnerstack.com/resources/research-lab/the-state-of-partners…
  6. sourcex.si/investors

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