A cashier holding a card terminal showing an approved $27.84 payment while a queue waits behind a plastic divider.

CEDX Checkout · Finance

The funnel tells you
where it hurts.

320 sessions this month: 92 completed for $117K, 76 still open, and a completion rate of 28.7% against a 62% target. Checkout measures the drop step by step — and scores the abandoned carts worth going after.

The biggest drop is the last step: 48% of sessions that reach payment never complete. The funnel names it; the recovery queue prices it at $84K.

checkout.cedxsystems.com — live build
CEDX Checkout overview: 320 sessions, completion 28.7% against a 62% target, recoverable $84K across 91 high-score carts, auth success 56.4% and complete GMV $117K.

Runs on demo data — Northline Outfitters' hosted checkout, with the experiment results printed on the overview.

28.7% completion92 of 320 sessions · target 62%
48% drop at the last steppayment → complete · the funnel's biggest leak
$172K in abandoned carts152 in pool · 91 high-score · $84K expected back

What it is

Measure the drop, then work the recovery.

A funnel that names the leak

Reach by step: 320 at cart, 266 at contact, 221 at shipping, 176 at payment, 92 complete. The drop-off leaderboard ranks the wounds — 48% at the final step, 20% at payment, 17% at contact — so the argument is about which step, not whose number.

  • Step funnel · by channel · by device tabs
  • Channel chips: Online · Marketplace · Embedded · Payment link
  • Cumulative reach printed per step
checkout — screen-3
A funnel that names the leak

Recovery scored, not blasted

152 abandoned or failed sessions in the pool, 91 scoring 85+, each with an expected recovery value and a play: SMS inside 30 minutes for a $312 cart at the payment step, email plus wallet link for a $559 cart. Expected recovery across the pool: $84K.

  • Score, E(recover) and play per cart
  • Recovery simulator and scoring model on tabs
  • Min-score chips: ≥40 · ≥55 · ≥70 · ≥85
checkout — screen-4
Recovery scored, not blasted

Every session, step and dollar

320 sessions with status, the step they reached, amount and method. A $377 session abandoned at info, a $347 session failed at payment on BNPL, an $886 session expired at payment — the drop-offs are rows, not estimates.

  • All · Open · Complete · Abandoned · Failed chips
  • Step reached on every row
  • $366K filtered total in view
checkout — screen-2
Every session, step and dollar

Product tour

Four screens, captured from the running build.

Not a mockup and not a concept deck. This is what opens at /app/checkout.

checkout.cedxsystems.com
CEDX Checkout Overview screen.CEDX Checkout Funnels screen.CEDX Checkout Abandon recovery screen.CEDX Checkout Sessions screen.

01 — Overview

Completion as a managed number

Completion 28.7% against the 62% target, auth success 56.4% on payment attempts, complete GMV $117K. The do-this-next board prices its own advice: send SMS on the top 25 abandon scores for $25K of impact, effort 2.

  • Guest-default experiment won +6.8pp — printed
  • Completion by channel: 43–48%
  • Health ring at 29% and honest about it

02 — Funnels

Where the 62% target dies

The step funnel shows 100% at cart decaying to 29% at complete, with each step's drop from the one before: 16.9% at contact, 16.9% at shipping, 20.4% at payment, 47.7% at complete. By channel and by device are one tab away.

  • 320 → 266 → 221 → 176 → 92
  • Biggest-drop card: 48% at complete
  • Weekly started-vs-complete trend above

03 — Abandon recovery

The queue worth working

91 high-score carts sorted by expected recovery: a $1.4K cart at the info step expecting $747 back, a $4.9K cart at shipping expecting $2.6K. Each row names its play — SMS < 30m, email at 1h, wallet deep link.

  • 152 in pool · $172K gross cart value
  • E(recover) base $84K with no levers pulled
  • Plays assigned per cart, not per batch

04 — Sessions

The raw material

320 sessions searchable by token, customer or email. The step chips — Cart, Info, Shipping, Payment, Complete — double as filters, so “everyone stuck at payment” is one click and 39 failed rows.

  • 76 open right now · 92 complete
  • Method per session: Card · BNPL · Bank · Wallet
  • Failed 39 — its own chip

Who runs it

Three roles move completion.

Roles, not references. We have no named customers yet, so nobody in these photographs is quoted, credited or claimed as one.

Store operations

Watches live sessions and payment-step failures — 76 open right now, 39 failed — and sees the counter-side reality of the funnel.

76 open · 39 failed

Ecommerce manager

Owns the 28.7%-against-62% conversation, reads the funnel by channel and device, and ships the experiment that removes the next account wall.

+6.8pp · last experiment

Recovery

Works the 91-cart queue top-down with the play already assigned per row — SMS inside 30 minutes where the score says so.

91 carts · $84K expected

The shape of it

What the demo workspace actually looks like.

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.

320sessions76 open · 92 complete · 39 failed
28.7%completionagainst a 62% target
$117Kcomplete GMVpaid sessions · auth success 56.4%
$84Kexpected recovery91 high-score carts of $172K
Reach by funnel stepall channels · from the funnels capture
  • Cart320
  • Contact266
  • Shipping221
  • Payment176
  • Complete92
Sessions completed92 of 320 · the 28.7% on the card
  • Complete · 92 · $117K GMV
  • Open, abandoned or failed · 228
Recovery expectation$84K expected of $172K abandoned
49%
  • Expected back · $84K
  • Written off · $88K of cart value

How it runs

A session's life, in the order it actually happens.

01

Start

A cart becomes a session with a token: 320 this month across online, marketplace, embedded and payment-link channels, 76 still live right now.

02

Measure

Each step's reach is counted and the drops ranked — the 48% final-step drop is a leaderboard position, not an anecdote.

03

Experiment

Changes ship as experiments with printed results: removing the account wall won +6.8 points of info reach, and the card says so.

04

Recover

Abandoned and failed sessions are scored; the top 91 carry $84K of expected recovery with a named play per cart.

One record

The session is one record
across the estate.

These captures come from one demo workspace — Northline Outfitters — so the sessions and amounts here connect to the same record the rest of the estate keeps.

All 132 applications

Limits

What Checkout does not do yet.

Finding this out on the third call is worse for you than reading it here, and worse for us.

Start

Open it before you talk to anyone.

Pilot

Your store, your funnel

  • Everything in Try
  • Session-import plan
  • Recovery-play workshop
  • Estate map
Talk to sales

Estate

Checkout with the rest of it

  • Checkout with Payments, Store and Subscriptions
  • One identity, one bill
  • CEDX delivery
Book an estate map

Questions

Before you pilot Checkout.

Is the software on this page real?

Yes. Every screenshot is a capture of the running build and you can open the same build at /app/checkout. It runs on demo data.

Why is completion only 28.7%?

Because that is the demo storefront's honest number, and the funnel shows where it dies: 48% of sessions that reach the final step never complete. The point of the page is that the leak is located, not hidden.

How are recovery scores set?

Each abandoned or failed session gets a score and an expected recovery value — the model is a tab on the recovery screen. The top of the queue, score 95, expects $163 back on a $312 cart with an SMS inside 30 minutes as the assigned play.

What did the guest-default experiment do?

Removed the account wall at the info step. The result is printed on the overview: +6.8 points of info reach. It is the demo's experiment, shown as evidence of the mechanism, not as a promise for your store.

Is Checkout audited or certified?

No certification has been issued. What we can evidence about hosting, encryption, tenant isolation and production access is written up on the security page.

The funnel is live. Go and look at it.

Live build, demo data, no card. Then find your own final-step drop and compare the number.