CEDX Payables · Finance
239 open bills worth $4.76M, 88 already overdue at $1.44M, and 103 blocked from the pay run waiting on approval. Payables is the board that decides what gets paid, when, and which discounts are worth taking early.
The duplicate review board is the quiet hero here: 41 pairs open, $766K at risk if both sides get paid. The expensive mistakes are the boring ones.
Runs on demo data — the bills screen marks its own auto-coded rows “pre-approved · GL matches vendor history”.
What it is
110 bills sit inside discount windows; the board recommends 25 takes worth $8.5K, each scored with face, saving, days left and terms. BILL-2476 saves $932 on $47K at 2/15 Net 45 with three days to spare — the cash-cap simulator proves the $620K it would use.
48 pairs scanned by multiple signals; 41 remain open and three are high-confidence. The review board shows BILL A next to BILL B — BILL-2200 against BILL-2199 for SecureStack SaaS at $11K — with confidence, tier and the dollars at risk on the row.
41 bills — $308K — were auto-coded and approved without a human. Each row shows the GL suggestion with its confidence and the reason: “GL matches vendor history · PO referenced · invoice # captured”, or “auto-approved · under $5K threshold”. Every one has an undo.
Product tour
Not a mockup and not a concept deck. This is what opens at /app/payables.
01 — Overview
Open AP $4.76M with $1.44M overdue across 88 bills, paid MTD $140K, and 103 bills blocked from the pay run waiting on approval. Alerts arrive priced: 41 duplicate pairs at $766K, and 5 vendors with recent bank-detail changes holding $189K of open AP.
02 — Bills
320 bills in the workspace with status from draft to paid. The touchless block sits at the top — 14 bills auto-coded and approved in this view, each with its GL suggestion, confidence and reason — above the rows still waiting on a person.
03 — Early-pay
The ranked queue shows score, the take decision, face value and saving per bill: Bright Group $44K saving $872, Keystone LLC $34K saving $678. The cash-cap simulator runs the same list against $620K of available cash.
04 — Duplicates
Seven pairs sit on the review board right now worth $151K in view: SecureStack SaaS at $11K, Papertrail Print at $4.1K with near-identical amounts and a shared PO, Ridge Services at $20K. Open, high, medium and low chips filter the pile.
Who runs it
Roles, not references. We have no named customers yet, so nobody in these photographs is quoted, credited or claimed as one.
Works the bills register and the duplicate board before every run — the $766K question is theirs to answer before the money moves.
41 pairs · $766K at risk
Clears the 103 bills waiting on approval — $1.66M blocked from the pay run — with the GL coding already suggested and explained.
103 bills · $1.66M waiting
Decides whether the 25 early-pay takes fit this week's cash — the simulator has already done the $620K arithmetic.
25 takes · $8.5K saveThe shape of it
Every figure below is legible in the captures above. Nothing here is a projection of your estate — it is the state of the demo data.
How it runs
Bills arrive and most get a GL suggestion with a confidence and a reason — 12% never need a person at all, and every automated call carries an undo.
What needs a person queues for one: 103 bills worth $1.66M are waiting right now, and approval SLA lagging is itself a priced alert.
Discount windows are ranked by saving against a cash cap — take, skip or let expire — and duplicate pairs are resolved before the run, not after the second payment.
Paid MTD sits on the overview at $140K against the weekly paid-vs-billed pace, with bank-detail changes on 5 vendors flagged before money moves.
One record
These captures come from one demo workspace — Northline Systems — so the vendors and amounts here are the same ones the rest of the finance apps carry.
Where the PO behind a referenced bill was issued — maverick buying and contract leakage are tracked on the same vendors.
CEDX TreasuryThe cash the pay run draws on: liquidity, runway and whether the early-pay takes fit the week.
CEDX BooksThe ledger the coded bills post to — and the finished reference page for this estate's evidence standard.
Limits
Finding this out on the third call is worse for you than reading it here, and worse for us.
Start
Open it right now
Your vendors, your terms
Payables with the rest of it
Questions
Yes. Every screenshot is a capture of the running build and you can open the same build at /app/payables. It runs on demo data.
Visibly. Every auto-coded row prints its GL suggestion with a confidence score and the reason — vendor history, PO referenced, invoice number captured, or under the $5K threshold — and every decision has an undo button next to it.
By saving, against a cash cap. The board shows face value, discount saving, days left and terms per bill; the simulator runs the same queue against $620K of cash and lands at $8.9K across 37 takes under cap.
Multiple signals — the screen names near-identical amounts and shared POs on the flagged pairs. 48 pairs were scanned, 41 are open, and the three high-confidence ones carry $766K of risk if both sides were paid.
No certification has been issued. What we can evidence about hosting, encryption, tenant isolation and production access is written up on the security page.
Live build, demo data, no card. Then ask how your team would find 41 duplicate pairs.