The deal pipeline behind data rooms.

Every deal tool on the market can draw a kanban board. What none of them can do is make the record underneath it provable. Meridian starts from the other end: the signatures are quantum-resistant, the signing events are attested on a blockchain, and the pipeline is the layer that reads that record and turns it into judgement. This is the difference between a CRM that tracks what people said happened and a pipeline that knows what actually did.

01The record comes first

Three properties separate a Meridian deal from a spreadsheet with stages in it.

• Every signature is quantum-resistant. NDAs, term sheets and closing documents on Meridian are signed, or countersigned, with ML-DSA-65, the lattice scheme NIST standardised as FIPS 204. Conventional e-signature platforms rest on classical cryptography that quantum computers are expected to break, and a record harvested today can be attacked then. A record signed on Meridian does not have that problem: the mathematics underneath it is chosen to survive.
• Every signing event is attested on-chain. Each acceptance is anchored to Armature L1, producing a tamper-evident, timestamped record outside the platform's own database. Proving that a document was signed, by whom and when, does not require taking KXCO's word for it.
• It already carries serious transactions. Meridian rooms are running live deals today, among them a THB 800 million resort development financing and the data rooms of a multi family office investment programme. The pipeline described below is not a demo drawn over sample data; it is the desk view of rooms doing real work.

A pipeline is only as honest as the record it reads. Ours reads signatures that will still verify in thirty years.

02The deals desk, live today

The deals desk shipped inside Meridian at meridian.kxco.ai earlier this month and has grown fast. What runs in production now:

• Deals, companies and contacts with a five-stage pipeline: Prospect, NDA Signed, In Diligence, Term Sheet, Closing. Stages carry default probabilities the owner can override per deal, and accounts can define their own pipelines and stages.
• A drag-and-drop board with a deal drawer, deal- and company-level sharing on viewer, member and administrator roles, and an audit entry for every change. Sharing a deal never grants data room rights; room access is always checked against the person acting.
• A timeline that cannot drift. A deal linked to its data rooms reads its history live from the room event stream. Nothing is copied, so the deal record and the room record can never disagree.
• Reporting that is arithmetic, not opinion: weighted pipeline by stage, win rate, forecast by expected close month, per-stage conversion from actual stage history, and close reasons on every won or lost deal.
• Connectors and alerts. HubSpot and Pipedrive sync today, Salesforce is built and awaiting activation. Alerts ride Telegram and Slack now, with WhatsApp to follow, and self-serve webhooks feed anything else. Every message is content-free by construction: it says what happened, never what a document contains.

03The pipeline workspace in build

What we are building now is the daily workspace over that engine, designed for a founder or a deals desk that lives in it. In build, honestly labelled as such:

• A metrics row across the top of the board: open deals, pipeline value, weighted value, win rate, average cycle, and the current quarter's forecast, each with its trend.
• Four views of one truth: the board for working, a list for scanning, a table for sorting, and a forecast view that stacks total pipeline against probability-weighted value by close month.
• Signature state on every card. Each deal card shows the envelope it is waiting on, who has signed, who has only viewed, and how many days it has been outstanding, with stalled envelopes flagged in red.
• Suggested next actions, ranked by urgency and drawn from the same activity record: the signature to chase, the export to approve, the counterparty whose reading has gone quiet.

04NDAs tracked to the signer

Most deals die politely, in the gap between "I sent the NDA" and "they signed it." Meridian closes that gap by treating the NDA as a live object with a state, per signer, rather than an attachment with a hope.

Three behaviours follow from tracking at that grain.

• Stalls surface themselves. An envelope out beyond a week turns red on the card and rises to the top of the signature tracker. Chasing is one action, aimed at the named signer who has not signed, not a group email into the void.
• Coverage is a number. The desk reports NDA coverage across the book: how many open deals stand behind a fully signed NDA and how many are running on goodwill. It is the single most clarifying number a deal owner can see on a Monday.
• The gate stays honest. Inside a data room the NDA is mandatory and nothing is readable before it is signed. On the board the owner stays in charge, a card can be dragged forward, but the gap is flagged the moment it happens and the coverage number reports it until it closes.

05Deal health from real activity

Deal stages are what the owner believes. Deal health is what the counterparty is doing. Because the pipeline sits on the data rooms, every deal linked to a room inherits a health score computed from real reading behaviour: who is coming back, what is holding attention, whether the engagement is broadening or narrowing.

The live activity feed runs beside the board with the events that move deals: an NDA signed, a room entered, an export requested, a consistency scan coming back clean, a signature completing. When a counterparty's document traffic cools, the deal is flagged before it goes cold, while a call can still change the outcome. The desk cannot make the call. It can make sure the owner knows to.

06Use case: an equity sale

A founder selling a 30 percent stake, two competing buyers at the table.

1. The deal is created on the board, one card per buyer, and the NDA goes out from the deal itself.
2. Each NDA is signed with a typed name, a Web3 wallet or a KXCO Identity key, countersigned in ML-DSA-65, and anchored to Armature L1. Signature verified, the buyer's data room access turns on in the same moment.
3. The rooms report which documents each buyer actually reads and for how long. Before the first negotiation call, the founder knows which buyer read the model twice and which one skimmed the deck. Deal health scores make the comparison explicit.
4. The term sheet goes out for signature from the same deal, then the closing documents. Every view and every signature sits on one record, from first NDA to close, provable without trusting anyone's memory.

07Use case: a property development

A resort development raising construction finance from private investors.

1. One data room holds the title documents, permits, drawings and the financial model. Investors are invited individually and each signs their own NDA before seeing anything.
2. Every investor is a card on the board. The activity feed shows who opened the feasibility study and who only said they would, and the consistency check has already read the model against the loan documents so the sponsor finds discrepancies before an investor does.
3. Term sheets move the serious investors forward on the board; the rest remain visible, and honest, in the earlier stages. The forecast view weights the raise by stage probability so the sponsor always knows the realistic number, not the hopeful one.
4. This is not hypothetical: a THB 800 million resort development financing runs on Meridian rooms today.

08Use case: debt funding

An operating company raising a term facility from three private credit funds.

1. Each fund signs an NDA, then receives a diligence room with the financials, security documents and covenant history, tiered so each fund sees exactly what its process requires.
2. The desk tracks lender engagement side by side. When one fund's analysts go quiet in the room, the deal health score says so, and the borrower chases that credit committee while momentum still exists.
3. Questions arrive against documents the room has already prepared answers for, because the room drafted the likely questions when the documents were indexed and the borrower approved answers in advance.
4. The facility agreement is signed where the diligence happened. The audit trail is one record from first NDA to drawdown, every signature quantum-resistant, every signing event anchored on-chain.

09Use case: NDAs on their own

A firm that sends twenty NDAs a month before any conversation starts.

1. Every NDA goes out through Meridian and appears in the signature tracker with its per-signer state: sent, viewed, signed, declined.
2. The viewed-but-unsigned list is the daily working list: these are counterparties who opened the undertaking and hesitated, and they are named, not guessed at.
3. Coverage reports the whole counterparty book at a glance, and stalled envelopes escalate on their own. Reminders go to the specific person holding things up.
4. Each signed NDA is a permanent, quantum-resistant, chain-anchored record. Two years later, proving who agreed to what, and when, is a lookup, not an argument.

10Questions people will ask

Is the pipeline live today?
The deals desk is live inside Meridian at meridian.kxco.ai today: deals and stages, the drag-and-drop board, deal health, weighted forecasting, sharing with roles, CRM connectors, and alerts over Telegram and Slack. The pipeline workspace shown in this post, the metrics row, the four views and the signature tracker, is in build now and ships on the same live foundation. This post says which is which, plainly.

What makes the signatures quantum-resistant?
Every signature on the platform is made or countersigned with ML-DSA-65, the lattice scheme NIST standardised as FIPS 204, at Category-3 parameters. That is what quantum-resistant means here: the signatures rest on mathematics that quantum computers are not expected to break, so the record stays provable after they arrive. Classical methods such as wallet signatures are always wrapped in an ML-DSA-65 countersignature.

What does the on-chain attestation add?
Each signing event is anchored to Armature L1, KXCO's blockchain. The anchor is a tamper-evident, timestamped record that exists outside the platform's own database, so proving a document was signed, by whom and when, does not require taking KXCO's word for it.

Do connected tools ever see documents?
No. Events sent to CRMs, Slack, Telegram or webhooks carry facts about activity: who entered a room, what was signed, which documents held attention and for how long. Document names and identities travel; page content never does.

Can a deal advance without a signed NDA?
Inside a data room, no: the NDA gate is mandatory and a reader sees nothing before signing. On the pipeline board the owner stays in charge, a deal card can be dragged forward, but the gap is flagged the moment it happens and the NDA coverage number reports it until it is closed.

The deals desk and data rooms run inside KXCO Meridian at meridian.kxco.ai, with the product overview at kxco.ai/meridian, the intelligence layer described in the previous engineering post, and the product as a whole documented in the user guide. Transactions are described without naming counterparties. KXCO Meridian is operated by Knightsbridge Financial Ltd, trading as KXCO. KXCO is a software company and is not a party to any transaction conducted through a data room, nor a law firm, and nothing in this post is legal advice. Cryptographic posture reflects NIST FIPS 203/204/205 alignment at Category-3 parameters; KXCO does not claim CNSA 2.0.