This hub offers practical guides and best practices for advisors looking to successfully introduce companies in the U.S., Canada, the EU, the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia (outside the U.S., Canada and the EU, at least 70% of the data must be in English) to SourceX, a direct licensing partner for approved business data for AI training and evaluation. Here, you'll discover step-by-step instructions and strategic insights for introducing companies to SourceX, which licenses data directly from companies and manages the agreement, rights review, data preparation, delivery and payout, subject to acceptance at Closing and agreed terms. These resources are designed to help you navigate the process efficiently, focusing on companies with 50+ full-time employees at peak. Unlock the potential of valuable data partnerships by leveraging these comprehensive guides.
Practical guides to successful company data introductions
Short, task-focused guides on identifying suitable companies in the U.S., Canada, the EU, the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia (outside the U.S., Canada and the EU, at least 70% of the data must be in English), recognizing valuable proprietary data, and making an introduction the decision maker will welcome.
- Step 1Share your linkSend your personal link to a company you know.
- Step 2Company appliesThe company applies itself at /apply.
- Step 3Company signs and is paidThe company signs, delivers its data and is paid by SourceX under its agreement.
- 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
What kind of companies should I introduce to SourceX?
You should focus on introducing companies in the U.S., Canada, the EU, the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia (outside the U.S., Canada and the EU, at least 70% of the data must be in English) that have 50 or more full-time employees at peak. These are the companies that can unlock the potential for valuable data partnerships with AI labs and data buyers.
What is the purpose of these guides?
These guides offer practical instructions and strategic insights for advisors to successfully introduce companies in the U.S., Canada, the EU, the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia to SourceX. They help navigate the process efficiently and identify valuable proprietary data for AI labs and data buyers.
Who benefits from SourceX data introductions?
AI labs and data buyers benefit from approved business data that SourceX licenses directly from companies in the U.S., Canada, the EU, the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia for AI training and evaluation. Advisors can also benefit by introducing companies to SourceX as their direct licensing partner.
What kind of information do these guides provide?
These guides provide step-by-step instructions and best practices for identifying suitable companies in the U.S., Canada, the EU, the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia. They help in recognizing valuable proprietary data and making introductions that decision makers will welcome, facilitating data transactions for AI.
Related pages
- How to model a one-time receipt, like a data license, in a 13-week cash flow forecast
- 2021-vintage private equity funds: what to do with companies bought at peak multiples
- 3PL and logistics M&A in 2026: what buyers want and what sellers should preserve
- 3PL valuation multiples in 2026: what buyers pay for in a logistics business
- A/E industry M&A trends in 2026, and what selling firms should prepare
- Acquiring a business from a retiring owner: assessing decades of records before they go
- Referring US companies from Canada: a guide for Canadian advisers
Canadian advisers can earn a reward by introducing companies in the U.S., Canada, the EU, the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia (outside the U.S., Canada and the EU, at least 70% of the data must be in English) with years of operational records to SourceX. Introduce only companies in the U.S., Canada, the EU, the UK, Australia, New Zealand, India, Singapore, South Africa, Latin America, the UAE and Saudi Arabia with 50+ full-time employees at peak, never share records, and check provincial professional rules, CRA reporting, GST/HST and CASL before you register.
Read → - Refund abuse review exception example: a valid return that looked like abuse, step by step
The records that explain a refund abuse exception are the ones that let someone follow a held refund to a decision and then to a checked outcome: the order and return events with their reason codes, the rule log showing which threshold fired, the support ticket, the reviewer's note and approver, the refund transaction and any later rule change. The most informative cases are valid returns that looked like abuse. Partners confirm such records exist, at metadata level, then ask the customer operations director to sponsor an introduction to the CEO, who can authorize a licensing discussion.
Read → - Refund abuse review history: what to inventory before a returns platform migration
Before a returns platform migration, inventory the record classes that explain each refund abuse review: the order and return events, the rule or flag that held the refund, the support ticket with its comments, the reviewer's decision and the confirmed outcome. Note the years each class covers, the system and owner of each, and the date the old returns tool stops giving access. A partner collects that metadata, never the records, and introduces the person who can authorize a licensing discussion.
Read → - Refund abuse review records across systems: the IDs that join store, returns app and helpdesk
Reconstructing a refund abuse review takes five joins: order to return, return to the rule hold, hold to the support ticket, ticket to the reviewer's decision, and decision to the refund. A commerce platform export gives the first and last; the hold, the conversation and the decision live in the returns app and the helpdesk. Each join depends on an identifier, such as an order number in a ticket field or a refund ID in a case note. Partners map which keys exist, never the records themselves.
Read → - Registration is a sign-up, activation is a definition: a worksheet and ten edge cases so two analysts count the same partners
A partner registration records that someone created an account. It does not show that the person signed the SourceX Channel Partner Agreement or introduced a company, or that any company applied. "Activated" has no definition in SourceX's terms, so it means whatever your fund writes down. Pick the triggering record, the date rule, the unit and what reverses it, test the definition on edge cases until two analysts agree, and report registered and activated as separate counts with their denominators.
Read → - Regret selling my business: what owners wish they had checked before closing
Owners most often regret what they failed to check before closing: weak earnout terms, vague transition roles, tax structure chosen late, and assets they did not know they owned. One overlooked asset is the company's records, which an asset sale can move to the buyer along with any option to license them to SourceX and data users.
Read → - Relationship partner vs CAS lead: who makes a client introduction
In multi-service accounting firms the relationship partner should ask the owner and register the referral, while the CAS lead records structural observations such as years of history and systems. Clear independence and referral-fee rules first, and give the client one disclosure and one point of contact.
Read → - Remnant asset sales: why books and records are carved out and what trustees can do next
A remnant asset sale is a trustee's end-of-case sale of whatever the estate still holds, such as unknown claims, refunds and residual rights, to a buyer who pays one lump sum so the case can close. Books and records are usually excluded, so a debtor's operational records may still be available for a separate fit check and license.
Read → - Restored at the fuse, still out at the transformer: a worked outage verification exception
The records that explain an outage exception are the ones that contradict each other: a work order marked restored, then a callback or meter reading showing some customers still dark, then the nested event and second crew action that fixed it, then the check that confirmed everyone was back. A company that kept all four can show how it caught and corrected a closure that came too early. This page works through one Illustrative storm case step by step, names the person who made each call and the record that kept it, and ends with how to spot clients holding similar cases.
Read → - Restructuring outlook 2027: what the 2026 signals mean for your intake
The 2026 signals, including debt maturities, private-credit stress, tariff pressure and sector disruption, point to a busy 2027 restructuring pipeline without a reliable timeline. Add a records-and-rights screen to every intake so licensable operating records reach SourceX before systems are decommissioned.
Read → - Result-to-approved-correction record chain: the sample, deviation and report keys that join LIMS, QMS and document control
A lab connects result-to-approved-correction records by anchoring everything on the sample ID and checking that two more keys were written beside it: the investigation number in the result's audit reason, and the report number with its revision. Inside the building, almost every hop has a key. The hop that breaks is the one leaving it: nothing links the corrected report to the earlier copies a client already received, so a unit correction can be absent from downstream copies.
Read → - Result-to-approved-correction records in LIMS, QMS and controlled documents: a sample metadata packet for an introduction
A referral partner can describe a lab's result-to-approved-correction history without reading a single result. The useful metadata is which system holds the instrument result, where the deviation and its approval live, how corrected reports are versioned and issued, how many corrections occur each year, which identifier links them, and who owns each step. The case to test first is a unit correction that was approved and recorded at the source but never reached the copies people outside the lab actually use.
Read → - Retiring with no buyer or successor: options that preserve value
A retiring business owner with no successor can transfer the company to managers or employees, keep owning it with a hired leader, sell later or for less, or wind it down in an orderly way. Whichever path wins, the company's operational records may still produce a one-time license payment, provided they are preserved before systems shut down.
Read → - Return-to-refund records in Shopify, Gorgias and Loop: joining a partial bundle return from ticket to refund
An ecommerce brand connects return-to-refund records by carrying the Shopify order number through every step: the customer's message in Gorgias, the return and inspection result in Loop, and the refund on the Shopify order timeline. The test that shows whether the chain holds is a bundle returned with pieces missing. If the brand can say which record explains the reduced refund, the chain is intact; if the refund stands alone, the reasoning was lost.
Read → - Revenue acceleration in PE portfolio companies: the levers that move the top line
Revenue acceleration in PE portfolio companies usually comes from five levers: pricing, retention and cross-sell, sales effectiveness, new products or markets, and add-on acquisitions. A newer, separate option is licensing historical operational records to SourceX, a one-time payment that needs no new customers, product build or AI engineering hires.
Read → - Revenue allocation corrections across systems: a map of billing, revenue tool, ledger and memo keys
Reconstructing a revenue allocation correction across systems needs four keys that survive the change: a contract ID carried by the CRM, billing system, revenue tool and ledger; an obligation or line ID that persists through the amendment; a version number on every calculation; and a journal reference that points back to the version that produced each entry. A fifth, softer link ties the adjustment memo to its contract and version. Billing and the ledger can agree on totals without any of these, but only the keys explain the totals.
Read → - Revenue allocation corrections history: what to inventory before a revenue tool migration
Before a revenue tool migration, inventory every record that explains a corrected allocation: the contract schedule and each amendment, every calculation version with its stand-alone selling price inputs, the approved adjustment memo, and the journal entries it produced. Note the years covered, who administers each system and when old-tool access ends. Keep it at metadata level, then introduce the head of revenue accounting, who can bring in the controller, counsel, the CFO and ultimately the CEO.
Read → - Revenue leakage audit: contracts, billing and the records behind them
A revenue leakage audit compares signed contracts, orders, usage and invoices over several years to find unbilled or uncollected revenue. It also shows how many years and systems of records a company holds and whether customer contracts limit data use, which are early questions in a SourceX licensing fit screen.
Read → - Revenue synergies in buy-and-build: cross-sell, pricing, reach and combined records
Revenue synergies in buy-and-build come from earning more from the combined customer base than the businesses could alone: cross-selling services, harmonizing pricing, reaching new territories and winning larger accounts. A further source sits in the records, since several add-ons' histories can form a broader licensable dataset, but treat that as one-time upside with documented rights, not a modeled synergy.
Read → - Reward hacking in a month-end close: when the total ties and the work is wrong
Reward hacking is when work satisfies a narrow scoring rule while missing the actual task. In a business, it looks like a bank reconciliation whose total ties to the statement while hard-to-match invoices were never matched. For an introducer, the question is whether a company kept the evidence that separates the two: item-level matches, adjustments, reviewer notes and the corrected outcome, and whether someone can authorize a license discussion about those records.
Read → - Reward hacking in business AI: why correct totals can hide missing invoices
Reward hacking appears in a reconciliation when an AI agent is scored on whether the difference reaches zero and gets there with an entry no document supports. The total looks right, the check passes, and an invoice is still missing. Business records that keep each statement line tied to its source document and reviewer sign-off let an evaluator write a check the shortcut cannot pass, which is why those archives can matter to AI developers.
Read → - RFI-to-change-order exception example: a late-notice hard case, decision by decision
The records that explain how an exception was handled are the ones that capture each decision with its date: the RFI answer, the revised drawing, the daily report recording the choice to proceed, the cost code opened for the extra work, the written notice, the priced proposal and the signed modification. In the Illustrative case below, notice went out late, so part of the cost fell outside the federal 20-day lookback. Each step left a record, and lined up in order they let a reviewer check every judgment the contractor made.
Read → - RFI-to-change-order system map: where each record lives and the keys that join one case
To reconstruct an RFI-to-change-order decision across systems, you need two keys carried from record to record: the RFI number and the drawing's system version. They have to reach the notice letter, the cost code for the extra work and the executed change order. Each of five systems holds one piece under its own label, so the map below shows where each record sits, which key reaches it, who owns it, and what one fully joined case looks like.
Read → - RFI-to-paid-change record chain: the identifiers that join Procore, Bluebeam and Viewpoint, hop by hop
A company connects RFI-to-paid-change records by carrying one reference, usually the RFI number or a change event number, from the question through the drawing revision to the approved change order and the bill, and by recording where that reference drops out. The project management tool makes the question end explicit and the ERP makes the money end explicit. The drawing in between, commented on and reissued, is the weakest hop, because nothing forces anyone to write the RFI number on it.
Read → - RFI-to-paid-change records: a sample metadata packet for introducing a contractor
A partner can note, without opening a single record, which systems hold each step from design question to reviewed drawing to change payment (here Procore, Bluebeam and Viewpoint Vista), how many projects and years each covers, which object types exist, who owns each one, and whether every drawing revision can be matched to a commercial approval. That is the whole packet. The records stay with the contractor; whether they can be licensed is settled later, after SourceX reviews them.
Read → - RIA aggregator integration: keep client NPI out and sort the firm records worth a look
During RIA aggregator integration, client records are nonpublic personal information and stay out of any data licensing discussion. Firm-level records, such as research memos, model portfolio rationales and operating procedures with personal data removed, may be reviewed with counsel, and a platform with 50+ full-time employees at peak can then be introduced to SourceX.
Read → - Roadside repair authorization exceptions: the records behind an out-of-network approval
The roadside repair records that explain an exception are the ones that capture the decision, not just the fix: the driver's report, the service desk ticket, the outside vendor's quote, the approval against a spending limit, the work order, the invoice and any variance decision, and the check that the repair held. Together they show the problem, the authorized action and the result. A fleet maintenance consultant describes them at metadata level and introduces the executive who can authorize a licensing discussion.
Read → - Roadside repair authorizations across systems: a map of unit, ticket and PO keys
Reconstructing a roadside repair authorization across systems needs three keys to appear in more than one place: the unit number, a breakdown or ticket ID, and the purchase order number. A fourth, the repair code, links the original fix to any repeat failure, and many fleets use the industry's standard maintenance coding for it. Follow the keys from telematics to the service desk, the vendor quote, the approval, the work order and the vendor invoice. Where one system drops a key, the case breaks there and someone has to guess.
Read → - Role change to access, before and after: what a ServiceNow-only export loses about one employee transfer
A single-system export loses the reason behind every access change and the evidence that old access was ever taken away. Read alone, a ServiceNow request log for a transferred employee shows new access being granted and nothing being removed, so access appears only to grow. The job change that prompted the requests sits in Workday, the group moves and any direct app assignments sit in Okta, and the time between a move and the removal of old access exists only when the three are read together.
Read → - Role-change-to-access record chain: how the employee ID joins Workday, Okta and ServiceNow for one transfer
A company connects role-change-to-access records through the employee ID: issued by the HR system, mapped onto the identity profile, and pulled into every access ticket through the requester field. The join is strong where each system reads the ID from the one before it, and weak where people type names. For someone who changes jobs rather than leaving, the weakest link is the last one: proof that access tied to the old role was removed, which often has no record at all.
Read → - Role-change-to-access records: a sample metadata packet for Workday, Okta and ServiceNow
A referral partner can describe a company's role-change-to-access history using only metadata: where the employee event is recorded, how it reaches the identity system, where access requests and removals are ticketed, how long each system keeps its events, and who owns each step. The part worth testing is the mover, someone who changes jobs rather than leaving, because that path is the one most companies never fully automated, and its records show where judgment entered.
Read → - Roll-up exit strategy: how to present a platform's combined records footprint
A strong roll-up exit strategy shows buyers what the platform has become, including the records every add-on brought with it. Map each system and archive to the legal entity that owns it, note years of history and export access, and confirm who can sign for each entity before any sale or data licensing process starts.
Read → - ROT data cleanup: do not mark old history as obsolete too fast
In a ROT data cleanup, hold back old project, ticket and email history that scripts tag as obsolete until the owner has reviewed it. Years of connected, outcome-labeled records can be licensable. A four-question hold screen keeps deletion from becoming irreversible.
Read → - Royalty reconciliation dispute records: the evidence trail when a refund crosses a reporting period
The royalty records that explain a disputed reconciliation are the ones that let someone recompute both sides: the franchisee's sales report as submitted, the POS transactions and refunds behind it, the exclusion and period rules in the franchise agreement version that applied, the adjustment request, and the approved credit with the note that cites the clause. Together they show the problem, the authorized action and the checked result. A franchise accounting advisor describes them at metadata level and introduces the franchise finance head.
Read → - Royalty reconciliation disputes across systems: a map of the keys that join POS, royalty platform and ERP records
Reconstructing a royalty dispute needs five joins: the refund to the payment it reverses, the payment to the location and reporting period, the period's report to the agreement version that set the rule, the short payment to its dispute ticket, and the ticket to the credit memo that settled it. The first join usually lives in the POS. The rest cross into the royalty platform, the franchise ERP and the agreement files, which is where most histories break.
Read → - Rubric failure: when an invoice check scores 4 out of 4 and misses the damage note
Rubric failure is when the checklist used to grade work leaves out part of what the work actually requires, so wrong work can score as perfect. A business example: an accounts payable quality check confirms the PO, price, quantity and approver, but never asks whether every invoice attachment was opened, so an approval that missed a damage note passes. For an introducer, the question is whether a company kept its checklists, the graded cases and the later evidence that exposed the gap, and who can authorize a license discussion.
Read → - Rubrics and QA scorecards: why they are useful as AI evaluation data
Rubric-based evaluation grades open-ended AI output against written criteria instead of a single right answer. Companies already hold this material as QA scorecards, audit checklists and review templates with scored examples, which makes them a distinct kind of potentially licensable record when they come with real work and real scores.
Read → - Rule 2004 examinations: how to locate a debtor's records, systems and custodians
Rule 2004 examination document requests work best when they ask for the map before the records: every system the debtor paid for, who administers each cloud tenant, where archives sit and which outside custodians, such as bookkeepers and managed IT providers, hold copies. That map shows a trustee what exists, what can still be exported and what may be licensable.
Read → - RWI underwriting call questions: data, privacy and AI topics to prepare
RWI underwriters usually ask how the company collects and uses personal data, who owns its code and content, what open-source and third-party material it relies on, and how it uses AI. A data license, signed or planned, should be described plainly and with a dated status so it is not treated as a known issue.
Read → - SaaS contract data export rights: can a client get its full history out?
Whether a company can retrieve its full history from a SaaS vendor depends on its contract: export formats, timing, termination assistance, retention after exit and who owns the customer data. Rules vary by contract and state, so check the signed terms and ask counsel. This is general information, not legal advice.
Read → - SaaS renewal management: using the calendar as a data checkpoint
SaaS renewal management for portfolio companies can double as a data checkpoint: before a tool is cancelled, downgraded or consolidated, confirm how many years of history it holds and whether it can be exported. Tools with multi-year records of outcomes may merit a review for licensing directly to SourceX before access ends.
Read → - SaaS valuation multiples for private companies in 2026: what moves the number
Private SaaS companies in 2026 are valued on a multiple of ARR or of EBITDA, and the multiple moves with growth, net revenue retention, gross margin, profitability and scale. Published ranges vary by dataset and deal size, so test them against comparable sales. A one-time data license adds cash once and is not capitalized at the ARR multiple.
Read → - Sales tech stack consolidation: check the data before tools are cut
Sales tech stack consolidation should include one step before any license lapses: inventory the history each tool holds, such as sequences, replies, call recordings, forecast submissions and deal outcomes, decide what to export, and flag portfolio companies whose multi-year records merit a licensing review. Cutting the tool saves money; deleting the history can destroy an asset.
Read → - Sales-to-delivery handoffs: a clue to operational context, and a four-seam discovery guide for asking which team wrote what before anyone is introduced
A handoff from sales to delivery is a clue, not a dataset. It suggests a company writes down what it promised and what it then had to build, the kind of operating context AI developers look for, but it proves nothing about rights or demand. Use it to ask four teams, sales, solution design, delivery and support, which records each one writes, where and for how long, who wrote each layer, and whether the CEO wants a conversation. Ask without requesting a single file.
Read → - Salesforce data archiving: what history to keep before you purge
Before archiving or purging Salesforce data, keep whatever tax, legal and contract rules require, archive externally rather than deleting, and ask the owner whether years of activities, cases and closed opportunities have licensing value. Consultants flag the question and introduce SourceX without opening or exporting any record.
Read → - Same alert, two causes: a worked wind turbine performance investigation, step by step
The records that explain a performance alert exception are the telemetry around the alert, each investigation step with who took it, any work order raised, the confirmed cause and the check that proved the outcome, kept together per case. The hard case is two alerts with identical text, one caused by weather and one by failing equipment, where only the investigation tells them apart. This page works through one such Illustrative case on two wind turbines, names each decision and the record it left, and ends with how to recognize clients who hold cases like it.
Read → - SAP integrators: the data introduction hidden in an S/4HANA transition
In most SAP estates, the decisions behind process exceptions sit outside clean master data: in documents attached to business objects, in workflow approval logs, and in emails and spreadsheets referenced from them. An S/4HANA transition tends to carry master data and open items forward and leaves that exception layer to a separate decision. SAP integrators can ask the company owner which exception decisions live there and, if the company appears eligible and wants to proceed, introduce its decision-maker to SourceX.
Read → - SBA Form 159 and referral agents: does it cover data introductions?
SBA Form 159 is a fee disclosure form tied to compensation connected to SBA loans, so an introduction to a separate data licensing service is a different activity. Whether your institution treats it that way depends on its policies and the current SBA rules, so confirm with compliance before registering as a SourceX partner.
Read →
Free resources
- Business succession planning assessment — Ten questions on successor, transition and documentation.
- NPV calculator — Net present value with a discounted cash flow table.
- Time value of money calculator — Future and present value with optional regular payments.
- All free tools · MCP resource center
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