The thirty-six questions a serious investor asks.

The CFO for a multi-ERP enterprise, the operating partner for a fund. Internal Audit champions, the CISO clears it because it is read-only. This is the working hypothesis, and it stays labeled a hypothesis until the evidence retires it.

Whoever owns the number that two systems disagree about: finance for working capital, the fund for portfolio EBITDA. A new category has no budget line, so the entry is an existing pain with an existing owner, not a new line item.

There is no published price, and the reason is the model: EOS is priced as a share of delivered value, with the mechanism and every fixed term published. Specific figures are set at scoping, against the customer's own systems.

The scoping conversation is free and thirty minutes. The assessment that follows is priced at scoping under the published structure. No number is anchored before anyone has found anything.

Read-only connection to systems already running. No migration, no write access, no new system of record. The remaining burden is access provisioning on the customer side. Production validation runs per engagement, and the assumptions register carries the status.

With a request on the assessment page. The thirty-minute scoping call is free, and a scope and a date come back within one business day. The assessment then runs two weeks, read-only, on two systems of record you name, and hands over the finding register, the decision log, the targets and a written outcome on day 14.

The target written with your finance team before Day 0, measured at day 90 against the baseline captured before the work. If the target is not achieved by day 90, the $35,000 balance is refunded and your one-time cost is the $15,000 assessment. What counts as a finding, as validated and as captured is defined in the pilot charter before any engagement.

$0 a year plus a marginal share of value your finance function has verified: 15% of the first $100,000, 10% to $500,000, 7.5% to $5 million, 5% above. There is no seat fee for your employees, no module fee, no per-connector fee and no minimum term. You can end it at any time, for any reason, with thirty days written notice.

By the value attribution rule, published before any engagement. The attributable share is the lower bound of what the evidence supports, against a baseline captured before the work, and a finding passes six gates first. The confounder deduction is deliberately unfavorable to us, and a dispute is answered by reproducing the calculation and showing every input.

Modeled figures are labeled synthetic everywhere they appear, and the commercial model shares value rather than charging for promise. It is named the Outcome Warranty, and its terms are written into the pilot contract before Day 0.

Eight ERP connectors are stated as conformed in an engineering build that is not published, so the results cannot be rerun from anything here, and nine more are declared scaffolds; none of the seventeen is in the live service, which reads files, EDI, bank statements and three warehouses, tested offline. Production validation is precisely the milestone the published pilot charter exists to produce. A production estate behaving like the mock is the assumption the pilot tests first.

The demo object is synthetic and labeled at every appearance; the real one is produced from your own records in the pilot.

The demo tenant is the walkable one, every surface of it; production tenancy is provisioned per engagement under the pilot charter, and this row records the date the first one stands up.

Run at every seal and published: a failing connector is isolated by its breaker, a forged signature dies deterministically, a tampered thread reports its own break. Refusals are recorded rather than hidden. The walkthrough shows them.

Disagreement is preserved, never repaired in place. Quality information travels with the record, and a figure that cannot be traced is presented as exactly that. Throughput on live-tenant data is measured in the assessment and reported as measured.

Each connector carries its own conformance suite, so vendor drift is detected by a failing case rather than a support ticket. The cost of keeping the seventeen engineering-release connectors current against vendor API churn is unmeasured until production, and the engineering allocation of the raise assumes it is real.

Independent assurance is planned: an accredited independent auditor is expected and has not yet been engaged. Until then the verification is the build's own. The shipped artifact is executed cold, offline, with no network and no vendor in the loop, and the method and the results are published in full.

Nobody yet. It is funded inside the security and assurance allocation and scheduled after close. No penetration test result is claimed anywhere, because none exists.

Not started, and marked not claimed on every page that could imply otherwise. The published milestone is a completed observation window at month twelve post-close, funded in the same allocation as the penetration test.

The customer keeps everything that matters. The systems of record were never touched by EOS, which holds no write credential to them. Evidence packs that verify offline without us are specified, not built; continuity terms (wind-down read-only access, data return) go in the master services agreement. The standard behind the claims is published openly, so the verification survives the vendor.

The entity is a owner-held LLC and no priced round has occurred. The exact figure is a data-room disclosure.

Disclosed in the data room before any signature, alongside the cap table itself.

Any outstanding instrument is disclosed in the data room, in full text, before a signature. Nothing is summarized on a public page where a summary could flatter.

A closing condition of any priced round is the Delaware C-Corp conversion, and the clean table that comes with it. Both are in the legal allocation of the capital plan.

Disclosed in diligence with documentation, not asserted on a page that cannot carry the evidence.

Disclosed in diligence. The public commitment is narrower and checkable: the largest allocation of the raise is commercial proof, not compensation.

Disclosed in diligence. The estate is built and operated at its present, lean scale today, which is what makes the current burn survivable and the raise a choice rather than a rescue.

The estate keeps operating at its present, lean scale; the build does not stop, the commercial motion slows. The three-level structure exists partly for this case: the product-level raise on EOS is a smaller instrument aimed at the same three-pilot milestone.

Closed-loop prepaid service credit. Not a token, not a financial instrument. It is purchased in dollars through standard card processing and redeemed for services on the consumer estate. It settles through a single closed ledger operated by the company.

There is none, by design. No smart contract, no chain deployment, no on-chain asset. A search for an SNX token finds nothing because there is nothing to find, and that is the intended result, not an omission.

No token keys exist. The ledger is company-operated infrastructure, with payment flows verified the way the rest of the estate is verified: signature-checked, replay-idempotent, tested.

Not a meaningful quantity for a closed-loop credit. There is no float, no secondary market, no exchange listing, and none is planned. Credits exist when purchased and retire when redeemed.

None, because there is nothing to hold. A credit that cannot be traded cannot be accumulated as a position, by the company or anyone else.

It is operated as closed-loop prepaid credit: it cannot appreciate, cannot be sold on, and confers no claim on the company or its revenue. No legal opinion is claimed on this page; formal classification questions are handled with counsel in diligence, like every other legal instrument here.

Consumer-estate services settle in it. The enterprise product has no SNX surface anywhere in it, which you can verify by searching this site. In any valuation, treat it as consumer-side plumbing, not an asset.

It does not, and the architecture says so structurally. EOS is Web2, read-only toward your systems by default, holding no write credential to them, with no wallet anywhere in it. The Web3 side is the consumer brand's distribution and provenance layer, valued as optionality on the investor page, and held to the same evidence standard as everything else here. If EOS succeeds, Web3 is upside. EOS was built so that it never becomes a crutch.