The request is fair. A founder asked to spend a meeting on something unfamiliar wants to know the stakes, and "what's it worth?" is the shortest way to ask. The mistake is hearing it as a request for your number. Usually it is a request for a reason to proceed, and you can supply that without inventing a price.
Five questions inside "what is it worth?"
The same words cover different needs. If you cannot tell which one you are hearing, ask: "What would the number help you decide?"
| What the founder says | What they usually need | The next step they are asking for |
|---|---|---|
| "What would we get for our records?" | A sense of the dataset's price | An estimate before committing time |
| "Would this change what the company is worth?" | Reassurance about enterprise value, often before a raise or sale | A view on diligence and disclosure |
| "Is this worth my team's time?" | Effort weighed against an uncertain return | A description of the work involved |
| "What's the least we should accept?" | A floor for a negotiation that has not started | Negotiating guidance |
| "I need a number for the board." | Something to circulate internally | A document the board can read |
Only the first is a straight price question. The second concerns the company's own value, which the guide on whether licensing data affects company valuation answers for a sale: the effect depends on the license terms, and nobody can promise that a multiple moves. The third is about effort. The last two are price questions dressed as process questions, and a guessed figure makes neither one better.
The expectation-setting decision guide
Work through it once the motive is clear. Each row gives what you can say today from published sources, what to leave out, who can give the real answer, and your move.
| Motive | What you can say today | Leave out | Who answers, and when | Your move |
|---|---|---|---|---|
| Dataset price | SourceX's application asks a few questions, needs no files and returns an indicative licensing estimate; the exact payout is fixed only in a signed Dataset Deal Confirmation | Any figure, range or comparison with a reported AI deal | SourceX, at application and then in the Confirmation | Offer your partner link so the CEO sees SourceX's estimate firsthand |
| Company value | A license is one contract among many that an acquirer reviews, and its effect depends on its terms | "It will lift the multiple" or "it won't matter" | The company's deal counsel and auditors | Send the valuation guide; proceed only if the CEO still wants the call |
| Team time | No records are shared at qualification; the company lists its systems, years of history and what can be exported | A promise about hours | SourceX, once it has seen the inventory | Describe the first step and let the CEO judge it |
| Negotiating floor | SourceX proposes an indicative range from the inventory, and the company can agree or counter | A floor of your own, which becomes an anchor | The CEO, advised by whoever they choose | Say that the counter belongs to the company, not to you |
| Board number | Any figure before the Confirmation is indicative, and data that does not reach Closing earns nothing | A number in your own words that the board will quote back | SourceX's written estimate, labeled indicative | Let the board read SourceX's document, not your email |
Then decide, once the CEO has heard the answer to the question they actually asked:
- Proceed when the CEO wants the conversation without a number from you. Send the link or make the introduction they approved.
- Pause when the CEO wants a figure before any step and will not apply to get one. Write down what would reopen the subject, such as a planned system migration, and do not chase.
- Close when the CEO says no. A declined referral needs no counterargument; the guide to declining an introduction request covers the reverse case, when you are the one saying no.
Why a figure from you does harm
Nobody has the inputs yet. SourceX's process page lists ten fit signals, from volume, historical depth and uniqueness to clear ownership and Acquisition Partner demand. It labels its sample data summary "Not a valuation" and says final value depends on actual Acquisition Partner demand and due diligence. From board materials, an operating partner sees two or three of those signals at most. The guide on why buyer demand changes the value of a data opportunity explains the demand side.
Data has no settled pricing method. An OECD note on measuring the economic value of data found no consensus yet on best practice for valuing data in national accounts, with theoretical work focused on market-, cost- and income-based approaches. If statisticians measuring whole economies lack an agreed method, a number improvised at a portfolio dinner has even less behind it.
Your number carries your interest. Under the default terms you would receive the greater of $50,000 plus 5% and 15% of Actual Supplier Payout, with no cap, on this one company's first qualifying transaction, with the USD 50,000 floor applied once per company and 15% on repeats agreed within 12 months; SourceX pays it separately, and none of it comes out of what the company receives. A partner who floats a high figure is describing their own upside, and the founder may notice before you do.
| Illustrative actual supplier payout | Default reward |
|---|---|
| $40,000 | None: below the USD 100,000 Minimum Transaction |
| $250,000 | $62,500 |
| $400,000 | $70,000 |
Your number carries the fund's name. A founder hears an operating partner's estimate as the firm's view, and a figure that lands in a board deck can outlive the conversation. The program terms themselves say that estimates in the partner portal or calculator are projections, not promises, and that indicative ranges are not Actual Supplier Payout.
Founder interest, a CEO's sponsorship and a decline (Illustrative)
Illustrative: three generic portfolio companies with invented details; no outcomes are implied.
Interest without authority. The VP of operations at a 160-person claims-processing company asks over dinner what "our ticket history" would fetch. He is curious, and he is not the CEO. The operating partner answers the time question, because that is the one he can act on, explains that the CEO decides whether the company talks to SourceX, and offers a short note he can forward. No figure, and no introduction yet.
Authorized sponsorship. The CEO of a 310-person logistics software company says her board will not discuss anything without a number. That is the board-number motive. The operating partner sends the partner link with one line: the application returns an indicative estimate without files, and the board can read that document directly. The CEO applies, and the board sees SourceX's figure, labeled indicative, in SourceX's own words.
A voluntary decline. The founder-CEO of a 90-person design agency will not take a call unless someone commits to a minimum first. Nobody can, and the operating partner says so: "I can't give you a floor, and I won't make one up." She declines. The log reads "declined, no follow-up", and the subject stays out of the next board meeting.
A curious executive is not the company's request
In this program, the person who can open a licensing discussion and put a signature on it for the company is the CEO. A VP, a CFO or a co-founder may be the most interested person in the room, may own the records or may prepare the case, and their questions deserve straight answers. Their valuation request is still personal interest until the CEO takes it up. The same split runs on your side of the table: your curiosity about the number, the fund's interest in a portfolio win and the company's decision are three different things, and only the last one decides whether an introduction happens.
So answer an executive at the level they can act on, route the decision to the CEO without pressure, and never let price talk become a lever: a fund that hints a portfolio company "should look at this" has mixed its investor role into a referral the CEO is free to refuse.
Check fit before anyone argues about price
A price debate over a company that does not fit wastes everyone's evening. The checks SourceX publishes: a peak full-time headcount of 50 or more, contractors not included; a documented operating history; the right to license the records; and a CEO who wants to have the conversation. Records that belong to the company's clients are usually not a fit. Until SourceX reviews the company, nobody knows whether its records match current demand, what customer contracts allow or what any license would pay. A company that misses one program may fit a later one.
Note what was asked, and pass on only what the CEO approved
When the CEO wants the conversation, write five lines the same day:
- The date, and who asked for a valuation.
- Which of the five questions it turned out to be.
- What you said, including that you gave no figure.
- What the CEO approved you to pass on, usually their name and contact plus descriptive facts such as record types, systems and years of history.
- What stays with you: board materials, financials, any number the founder floated and anything learned under the fund's information rights.
If the founder has a figure in mind, it belongs in the company's own counter to SourceX's indicative range, not in your introduction. Partners never ask for or handle raw records, and the description stays at metadata level. For the other questions a CEO may send that you cannot verify, use the response bank for unverifiable details. The overview for venture platform teams covers the program from the fund's side.
Where this guide is wrong
- The founder has real comparables. A CEO who has licensed data before may know what similar records earned. That knowledge is theirs to use in their own counter; repeating it still turns it into your claim.
- An indicative estimate can disappoint. Pricing stays indicative until the Confirmation is signed, and qualifying does not guarantee an offer. Sending the question to SourceX moves the anchor to a better-informed source; it does not remove the anchor.
- Demand moves. Acquisition Partner demand is one of SourceX's fit signals, so an estimate can age and a company turned down today may fit a later program.
- Some founders will walk. A founder who insists on a committed figure before any step may never apply. No data on how often that happens has been published.
Methods note. The five motives are a classification built from how founders phrase the question, not from published interviews; SourceX has not released founder interview data or counts of how often each motive appears. The "say today" column draws only on SourceX's terms (version 2026-10-22.2, since replaced by version 2026-11-04), its how-it-works page and its who-qualifies page, all checked 2026-10-26. To test the guide, log the motive behind every valuation request for a quarter and record which response led to an application, a pause or a decline.
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.
When a founder asks for a valuation, answer the question they meant, then register for a partner link so the CEO can request SourceX's own estimate, check the company against who qualifies and keep the program terms open while you reply.
Evaluating SourceX itself as a possible investment is a separate decision from any referral; investor materials are available on request 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.