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
- The difference between an LP referral and fundraising solicitation: an activity scope comparison that sorts twelve things a fund might ask of an LP into company introductions, the fund's own capital raising, or mixes to split before anything is sent
An LP referral asks an LP who knows a company to put its CEO in touch, and that CEO then decides whether the company talks to SourceX. Fundraising asks an investor to put money into a fund. The two differ in who decides, which documents govern and who pays whom, so keep them in separate messages, never place an investment ask inside a company introduction, and remember that a SourceX reward is paid only for introducing companies under the published terms, never for introducing investors. How rules on fundraising apply to a fund is for that fund's own counsel to say.
Read → - The difference between applications and companies referred: an entity-deduplication bridge that turns every submission into one row per company before a fund reports a number
Applications count submissions; companies referred count the businesses a fund actually put in front of SourceX. The two numbers drift apart whenever a company applies twice, a second contact at the same business applies, a trade name and a legal name are logged separately, or subsidiaries apply on their own. Report companies, not applications: run every submission through a deduplication bridge, roll entities up to their group, then split the companies into credited, not credited and pending.
Read → - The difference between data inventory and referral metadata: seven fields that each answer one screening check, and why the rest of the inventory stays with the company
A data inventory is the company's own working document: its systems, tables and fields, the people who administer them, retention settings and often hosting and security details. Referral metadata is the small, CEO-approved subset a partner hands to SourceX so it can decide whether a first call is worth having. Keep a line only if it answers a screening check: size, location and language, record families, span of years, origin and control, the CEO's position and open questions. Everything else stays with the company.
Read → - The difference between finding archives and licensing them: a ten-step responsibility map that shows where a fund's discovery work ends and the company's licensing work begins
Finding an archive tells you a company may have something worth discussing. It gives nobody the right to license it, and it gives you no part in doing so. A portfolio operations lead's scope ends once the archive is described at metadata level, the holder of each copy and the CEO who decides are named, and that CEO has agreed to a conversation. Rights, custody, scope, preparation and delivery then belong to the company, its counsel and SourceX.
Read → - The difference between introducing a client and recommending a transaction: a seven-step boundary ladder that marks where a mention turns into a call to act, what each step asks your counsel, and three cases that end in a referral, an unpaid introduction and no part at all
An introduction tells a client that SourceX exists, that the company's CEO decides, and how the company can reach SourceX; everything else stays with the client. A recommendation tells the client what to do about the transaction: talk, apply, license, accept terms or sign. What moves you from one to the other is what you say, how closely it fits this client and how the client would reasonably hear it, not the label you give it. Stay on the introduction side unless your counsel, compliance officer or ethics adviser clears more.
Read → - The difference between partner location and supplier eligibility: a two-location diagram that settles who may refer and which company may be introduced as two separate questions, and never lets one answer stand in for the other
Two locations, two questions, two answers. Your own location decides whether you can take part as a partner; SourceX's who-qualifies page says partners can join from most countries, apart from sanctioned places and restricted parties, and because the terms require an approved partner account, SourceX settles it, not you. The company's location, together with the language of its data for places outside the U.S., Canada and the EU, decides whether it can be introduced, and SourceX settles that in review. Neither answer carries over to the other, so record both for every pairing.
Read → - The difference between preserving evidence and monetizing an archive: a preservation-first decision guide that keeps why records were kept apart from whether they could be licensed, so a licensing hope never becomes a reason to keep or discard anything
Preservation and licensing are two decisions with two owners. Records kept as evidence are kept because a legal duty or counsel says so, for a scope counsel sets. Whether a company's records are ever licensed is a commercial decision for its CEO, or for whoever controls the estate if the company has closed. Keep them apart: an introducer's licensing hope is never a reason to keep records past their date or to discard them early, and a preservation file is never an inventory of what could be sold. Settle preservation first, then ask the decider once.
Read → - The difference between program membership and endorsement: a badge-and-name language guide that keeps a signed partner agreement from reading as SourceX's seal of approval, and keeps your own recommendation from hiding its reward
Membership is a contract fact: your firm holds an approved partner account and has signed the SourceX Channel Partner Agreement, so it may introduce companies and may be paid a reward under the terms. It is not SourceX vouching for you, and nothing SourceX publishes makes it a certification or a rank. The unsigned sample agreement grants no trademark license, bars either side from publicly naming the other as a partner without written authorization, and makes the partner an independent contractor, not an agent. Say less than you could, and put the reward beside every recommendation.
Read → - The difference between skepticism and a genuine no: four tests that tell a founder's question from a founder's decision, so you answer the first once and close on the second
Skepticism asks for something: a fact, proof, a person or a condition. A genuine no asks for nothing; it states a decision about the company, and when the CEO says it, it is the company's answer. Sort every reply with four tests (does it ask, does it name a condition, who said it, has it been said before), answer an objection once with facts you can source, and treat anything that is neither as a no. Because your fund holds influence, lean toward reading a reply as a no.
Read → - The difference between technical feasibility and licensing eligibility: a two-stage review comparison showing why an export you could run this afternoon settles none of the questions that decide whether a client's records can be licensed
Technical feasibility asks whether records can be pulled out of a system in usable form. Licensing eligibility asks whether the company fits SourceX's criteria, holds the rights, has never licensed those records for AI training and has a CEO who wants to go ahead. An MSP is the expert on the first and has no vote on the second. SourceX settles eligibility first, from the company's own description and with no files, and extraction comes last, after the company signs. So an export run "to see if they qualify" tests the wrong thing at the wrong time.
Read → - The email test that measures qualified replies instead of opens: a four-check reply rubric scored blind, a baseline coded from replies you already hold, and a send list limited to CEOs you know
Count replies, not opens, and count only the replies that pass four checks: the CEO who decides wrote it, it shows the reader understood the offer, it asks for a next step, and it carries no sign of pressure or misreading. Send only to CEOs you know, compare your current note with one changed note assigned by coin flip, and have two readers score each reply without knowing which note produced it. No partner has published results; this page is the method.
Read → - The escalation path for a disputed investor introduction: five rungs from a paused inbox to arbitration, one evidence packet that travels up them, and a CEO nobody asks to referee
SourceX decides which partner is credited, from dated records, and an argument between investors does not change that. When two funds claim one introduction, a VC ecosystem lead's job is to keep the dispute small: stop all outreach, separate what SourceX decides from what the funds owe each other, send SourceX one evidence-first packet, and treat the partner agreement's 30-day notice and arbitration clause as the last rungs. Nobody asks the company's CEO to referee.
Read → - The ESOP sale process and where a data license fits before trustee valuation
An ESOP sale usually moves from a feasibility study and plan design to engaging an independent trustee, valuation and price negotiation, financing and closing. A data license fits best before the trustee's valuation work starts, so the appraiser can treat the one-time proceeds and any exclusivity term correctly and the trustee sees it fully disclosed.
Read → - The event follow-up that records what the founder requested: a read-back template that quotes each request, says what was done and what was not, and leaves every introduction to the company's CEO
Follow up on what each founder asked for, not on the fact that they came. Note every request at the event in the founder's own words, then send one read-back that quotes it, says what you did, says what you did not do and names the one step still waiting for the company's CEO. Someone who is not the CEO can ask for information but cannot ask for the company to be introduced. Anyone who asked for nothing hears nothing.
Read → - The evidence behind a claim of rapid partner growth: a dated cohort file that has to exist before the sentence is published, and five growth sentences tested against it
A claim of rapid partner growth holds up only if its evidence existed on the day the sentence went out: a frozen, dated file that counts partners by the quarter they signed, shows how many in each cohort proposed a company, and counts each fit company once. Sign-up growth supports a sentence about sign-ups. "Rapid growth" implies the network produces more, so it also needs the activity and unique-company columns. SourceX publishes no partner counts, so no such claim can lean on its name.
Read → - The evidence to include in an attribution question: a seven-field, one-page submission, and the attachments that should never be in it
Send less than you think. SourceX decides attribution from dated records and already holds most of them: portal submissions, link visits, applications. Your packet adds only what you hold, on one page: the company's legal name and domain, your written introduction with its timestamp, the application date as the company reports it, the company's own account of any earlier contact with SourceX, and one sentence stating what you want decided. Leave out company records, screenshots of its systems and anything about another partner.
Read → - The executive council session about business data opportunities: a 45-minute agenda for a room of functional executives, a parking lot for the questions it must not answer, and a take-back note each member can hand to their own CEO
An executive council exists for peer learning among people who do the same job, so a session about licensing business data has to serve that purpose: explain how direct licensing works from published facts, keep every member's company unnamed and undescribed, and leave each decision with that company's CEO. In a council organized by function, members are finance, technology or operations heads rather than CEOs, so the useful output is a short note a member may carry to their own chief executive, never a sign-up sheet.
Read → - The executive in residence role: what EIRs do and how they make portfolio introductions
An executive in residence (EIR) is a seasoned operator who joins a private equity or venture firm for a defined period to advise portfolio companies, evaluate deals or prepare for a CEO or chair seat. Because EIRs meet many portfolio CEOs, they can bring a data licensing screen to qualifying companies through the firm's own process, with conflicts disclosed.
Read → - The exit data book: how KPI preparation also makes a company licensing-ready
An exit data book is the sell-side set of reconciled KPI data, such as revenue by customer, cohorts, margins and pipeline, that bankers and buyers analyze in a sale process. Building it forces a map of source systems, date ranges and data owners, which is the same groundwork a data licensing review needs.
Read → - The exit story and the value creation plan: where a completed data license fits
The exit story is the value creation plan retold as evidence: what the owner set out to change, what was delivered and what a buyer can still capture. A completed data license fits it as proof of a rights-cleared, monetizable records asset, presented as a documented one-time item rather than as projected EBITDA that buyers would discount.
Read → - The experiment behind a better partnership landing page: a frame test that keeps every sentence identical, shows it either as a portfolio perks listing or as a plain explainer, and scores qualified conversions beside the founders who think the fund won them a better deal
Test the frame, not the words. A platform may be tempted to list SourceX in its portfolio perks catalog, beside negotiated discounts, and a perks catalog tells readers the fund bargained for something. Publish the same facts in two settings, a catalog listing and a plain explainer, send each founder to one by a logged coin flip, and judge the result on qualified conversions (a CEO's own request that becomes a verified application) and on how many founders come away thinking the fund got them a better deal.
Read → - The experiment that separates reward interest from company fit: two recruitment invitations that differ only in where the reward sits, a motive question kept out of the comparison, and qualified introductions per person invited as the count that decides
Reward interest and company fit get tangled because the people who sign up for a reward may also be the people with the widest, loosest networks, so comparing them after the fact proves nothing. Separate the two by changing only the invitation. Send half your network an invitation that opens with the reward and half one that opens with the fit facts, both stating identical facts, then count qualified introductions per person invited. A motive question at sign-up describes who joined; it never decides. No results exist yet.
Read → - The family business introduction through a venture relationship: a family authority and sponsor map that finds whoever holds the CEO role, keeps the relative you know as a sponsor and leaves the fund's business out of it
Find out who holds the CEO role before you ask anyone for anything. A venture tie usually reaches a family business through one relative, an LP, a founder in your portfolio or a co-investor, and that relative rarely decides. Under the program the company's CEO decides whether it discusses licensing its records: whoever holds that role, family member or hired executive, with the wider family involved only if the CEO chooses. Ask your contact to sponsor the question, and keep every fund matter out of the conversation.
Read → - The financial questions to ask before a large-network partnership: a sixty-minute finance agenda, and four Illustrative outcomes from $0 to a $120,000 reward
A large network's referral economics are set per company, not per member. On a company's first qualifying transaction SourceX pays the greater of $50,000 plus 5% and 15% of Actual Supplier Payout, with no cap, to one credited partner, and only after the dataset is accepted, handed off and paid for; further qualifying transactions first agreed within 12 months pay 15%. Network size raises the number of introductions, the duplicate rate and the cost of running the program, not the reward formula. Before signing, finance should settle who the partner of record is, how members share any reward, who absorbs a later reduction and what a year with no paid reward costs.
Read → - The first AI data conversation can create value before a deal: what a founder learns in one hour, even when the answer is no
Run as discovery rather than a pitch, one permissioned, metadata-only conversation can tell a founder who controls each system, how many years of records survive, whether a retention setting is quietly deleting them and whether customer contracts limit reuse. Those answers help whether or not a license follows. They are learning, not financial value: only a completed, paid license creates a payout. SourceX has not published an approved real no-deal case, so the case below is Illustrative.
Read → - The first partner manager hire: a role and handoff map that gives them the screen, the duplicate register and the intake cap, and leaves the ask, the decision and the money with others
The first partner manager should own the work that falls between people: activating introducers, running a metadata screen and a duplicate check before anyone contacts a CEO, keeping one dated record per company, routing company questions to SourceX and closing intake when hours run out. The ask stays with the relationship owner, the decision with the company's CEO, and rights review, attribution and payment with SourceX. The role never touches client records and never forecasts anyone's income.
Read → - The first ten conversations for a fund exploring data partnerships: a three-phase learning agenda, with the one question each conversation must answer
Treat discovery as a learning agenda, not a campaign. A platform lead's first ten conversations should run in a fixed order: three inside the fund to settle scope, conflicts and which companies may be discussed; four with SourceX and the people who own CEO relationships; and three that test the idea with CEOs and close with a decision. Each conversation answers one question, and any of them can end the agenda. What is being tested is an authorized company relationship, not portfolio membership.
Read → - The follow-up experiment that honors no-contact preferences: a pre-announced reminder test for unanswered first asks, in which the CEO can cancel the only follow-up in advance, every no binds the whole network and nobody on a no-contact list is ever drawn
A follow-up experiment honors no-contact preferences when those preferences are settled before randomization and outrank it: anyone who has said no or asked not to be contacted leaves the list before the draw, every arm sends at most one follow-up, and any no ends contact from the whole network at once. Inside those limits this design tests one thing: whether announcing that single follow-up in the first ask, with a one-word way to cancel it, changes qualified introductions, clear answers and objections. No result exists yet.
Read → - The follow-up that adds a useful answer: a question-led library that replaces 'just checking in' with one sourced fact and one easy choice
A follow-up is worth sending only if it answers something the company actually wants to know, about its own situation, and leaves an easy way to say no. Replace "just checking in" with one useful answer: what the first step involves, whether the company keeps ownership, how long the steps take, or what you earn and why it never comes out of the company's payout. Each answer in the library below comes from SourceX's published terms or process page and ends with one voluntary next step.
Read → - The founder has the hard drive: who owns the records? A custody-versus-ownership case worksheet for six ways a founder ends up holding company files, showing who decides for those records today and why nobody copies the drive to find out
The drive settles who holds a copy, not who owns the records or who could license them. SourceX licenses from the company, which has to confirm it holds the rights, and the company's current controller decides: its CEO while it operates, the acquirer's CEO after a sale, and the estate's controller once it has wound down or entered a formal process, where the CEO may not be enough. Whether the founder's copy is company property is for the company's counsel. No partner asks for, collects or copies the drive.
Read → - The founder is interested but the CFO is not ready: an eight-concern worksheet that sorts what the terms settle, what only review can answer and what is a real reason to wait
If the founder is also the CEO and wants the conversation, the company's authority is already in the room; the CFO's readiness is something the company settles, not something you resolve for it. Help the CEO list every concern the CFO raises, sort each one by who can answer it, answer only what the program publishes, and treat a CEO who decides to wait for the CFO as a complete answer. Never lobby the CFO, and never route around a hesitant CFO through the board.
Read → - The founder leaves but the relationship remains: a relationship continuity protocol for the six roles a departed founder can take in your network, showing what each role can carry into a referral, what stays with the old company and who decides for it now
When a founder leaves the company, your relationship with them survives and their authority over the company does not. From the exit onward the company's CEO decides about licensing, and the founder decides only for themselves. What the relationship can still do depends on the role the founder takes next: shareholder, adviser, board member, new founder, introducer or colleague at your fund. In every role it can carry trust and introductions to other companies, and never what the founder learned as CEO.
Read → - The founder roundtable that reveals data questions worth answering: a research-first runbook that writes its hypotheses down before the invitations, collects questions instead of company facts and answers in public only those that pass a four-part test
Run the roundtable as research, not as a pipeline event. Before inviting anyone, write down which data questions you expect founders to raise. Invite six to twelve of them, make every seat optional, and collect questions rather than company facts. Afterwards, answer for everyone only the questions that recur, would change a founder's next step, have a named source for the answer and need no company's private facts. A founder who wants more asks privately, later, and only that company's CEO decides whether anything reaches SourceX.
Read → - The founder wants a buyer promised before an introduction: four levels of demand evidence, the strongest sentence each one supports, and the low-cost test that replaces a promise
You can't promise demand, and you don't need to. Before review, nobody can tell a founder that a buyer wants its records: SourceX's own process says qualifying does not guarantee an offer and that positive interest is not yet a purchase. What you can offer is a graded answer: what other companies have announced, what SourceX publishes as fit, what its no-file assessment would tell this company, and the signed document at which demand finally becomes a fact.
Read → - The founder who returns to an old company as chair: a role-and-authority verification guide that confirms what the returning founder holds today, keeps the licensing decision with the company's CEO and treats the chair as a possible door, never the yes
A founder who comes back as chair can open a door: they may know the CEO, the company's history and where its old records were kept. They do not decide. The company's CEO authorizes any licensing discussion and signs for the company, and if the company has been acquired since the founder left, the acquirer's CEO decides. Confirm the founder's exact role and its start date, ask them only whether they would introduce you, and let the new arrangement settle before anyone puts the question to the CEO.
Read → - The fractional CFO's first 90 days with a new client, block by block
A fractional CFO's first 90 days should run in three blocks: days 1-30 secure access, map the systems and stabilize cash; days 31-60 fix the close calendar and reporting pack; days 61-90 set the forecast and board rhythm and take a records depth snapshot, which also shows whether the client meets the baseline for a data license introduction.
Read → - The future of AI training data: what comes after the public web
After the public web, AI training data is likely to come from four growing sources: licensed enterprise records, purpose-built human data, synthetic data and agent interaction logs. None is certain to dominate. For companies holding years of operational records, licensing is the route that applies, because that data is scarce outside company walls.
Read → - The gap between data interest and data readiness: a paired-reading study that has each company score six readiness items the week its CEO shows interest and again at day 60, seals the partner's forecast for comparison and publishes no gap figure until the cohort closes
When a CEO says a data licensing conversation sounds interesting, you learn that the conversation is welcome, not that the company can act on it. No published study has measured that gap, SourceX has released no figure on it, and this page reports none. It sets out the study that would: six readiness items the company scores for itself in the week its CEO shows interest and again at days 30 and 60, a sealed partner forecast, and a denominator that keeps every interested company, including those that stop, go quiet or do not fit.
Read → - The governance review before activating a senior network: a six-switch pre-activation review that clears each LP, executive and advisor relationship on purpose, obligations, conflicts, capital timing, the CEO's path and a free no before anyone is asked
Activate relationships one at a time, never the network as a whole. Before a fund asks an LP, a senior executive or an advisor to introduce a company, check six things about that one relationship: what it exists for, which obligations already bind the person, whether the fund's interest in a reward needs its governance body, whether a capital conversation is open, who can reach the company's CEO, and how a no will be honored. Any relationship that fails a check is left alone, and the company's CEO still decides.
Read → - The handoff checklist after a company accepts an introduction: a three-parcel packet that sends the introduction record to SourceX, the next steps to the CEO and the proof to your own file, with nothing else packed
Once a company's CEO accepts an introduction, the partner's work shrinks to one packet in three parts. SourceX gets the written introduction record: the company, its domain, the CEO, who is introducing it, for what, how and when. The CEO gets SourceX's own account of what happens next and a plain statement of what you will and will not do. Your file keeps the CEO's written yes and the submission reference. No files, samples, logins, figures or price talk go in any part, and then you step back.
Read → - The hidden business data network behind the next AI wave: a permissioned sampling protocol to measure it before anyone claims it exists
Public AI market maps chart funded technology startups and leave out the established operating companies a fund reaches through alumni, portfolio customers, advisors and owner-led relationships, which are the companies most likely to hold years of their own records. Whether those relationships add up to licensable supply has not been measured, by SourceX or anyone else. A research director can test it with a permissioned sample: stratify relationships by type, record operating history and an authorized sponsor, and count every elimination. No results exist yet; this page is the protocol.
Read → - The hidden friction between innovation and legal teams: a stakeholder handoff diagnostic that scores the six facts every request to legal should carry, so a data introduction stops bouncing between the two
The friction is rarely a disagreement about SourceX. It is a handoff problem: the innovation team asks legal something broad, such as whether the corporate can partner with SourceX, without the facts legal needs, and legal replies with conditions nobody owns. Score each request on six facts: the company, the corporate's seat, the partner of record, the money, the draft words and where each fact came from. Then send one note carrying all six. However well the two teams align, the operating company's CEO still decides alone.
Read → - The hidden operating history in a founder alumni network: a ten-question survey that finds the established companies your founders already know
A founder alumni network's value for data referrals is rarely the startups themselves, which are usually too young or too small, but the established companies those founders know from the inside: former employers, family businesses, companies they now run after an exit, boards they sit on. Each counts only through an authorized relationship with that company's CEO. Ask alumni a short, opt-in questionnaire about those companies and relationships, screen the answers against the published fit facts, and rank conversations without guessing proceeds.
Read → - The industry experts who can find AI data opportunities first, and the three gates that cancel their head start
A specialist's head start is recognition. An ERP implementer or a claims auditor can hear an offhand remark about an old system and know it implies years of company-owned records, before a generalist network notices. The head start counts only if three gates also open: the specialist can reach the CEO, their professional rules allow a referral, and the records belong to the company rather than its clients. SourceX has not published interview results on which specialists do this best; this page gives the test and the interview method.
Read → - The introduction brief for an English-operating international company: eight lines that keep what the fund saw in English apart from what the company states about its records
Write the brief so that nothing you observed passes for something the company said. What you see of an English-operating company (leadership, board reporting, website, head office) goes in one line labeled context. The language of the records the company might offer is stated by the company itself, approved by its CEO, or the line reads "not yet stated". You never open, sample or export records to fill it. Quote SourceX's geography sentence, add what only the CEO can confirm, and send nothing else.
Read → - The introduction economy behind AI data: map the people who see the records
The relationships that surface company-owned records are the ones that combine two kinds of access: trust with the owner who can authorize a licensing discussion, and working knowledge of the systems where operational history sits. Outside accountants, fractional CFOs, ERP and IT partners, business lawyers, M&A advisors and private equity operating partners each hold one or both. AI data companies now pay referral rewards to reach them, which is the introduction economy this page maps.
Read → - The introduction opportunity after a client knowledge transfer: a services firm that has just handed a client's work over knows, for once, exactly what it kept, and a seven-family owned-versus-client content review its own leaders answer before the CEO is asked
When a services firm finishes a knowledge transfer, handing a client's work back to the client or on to a successor provider, it has just had to decide, item by item, what belongs to the client and what it keeps. That makes the weeks after sign-off a fair moment for one question to the firm's CEO: is the material that stayed the firm's own, such as its procedures, training and internal quality history, worth a conversation with SourceX? Everything that went to the client stays out.
Read → - The introduction path for a cooperative's member businesses: a cooperative-versus-member consent workflow in which the co-op sends one notice, each member's CEO decides alone and nothing about any member passes through the co-op
Control in a cooperative runs from the members to the co-op, never the other way, and the introduction path has to follow it. A cooperative can tell its member businesses that SourceX exists. Each member is still its own company whose CEO alone decides whether to talk, and the co-op never gathers member records, lists or headcounts, never suggests that members fit and never answers for one. The co-op's own operating records are a separate question, for the co-op's own CEO.
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