Order to Cash: The Four Handoffs Where the Cash Actually Stalls.
Updated: 2 days ago
Walk the order-to-cash path in most mid-market companies and every application on it is doing its job. The CRM produced the quote. The order system accepted the order. The warehouse shipped it. Finance invoiced it and eventually got paid. Ask why the cash took sixty days to arrive on a thirty-day term and nobody can point at a system that failed.
That is because nothing failed inside a system. The delay lives in the four places where the record leaves one system and has to arrive in the next, and those handoffs are where order to cash automation should start and almost never does.
The enemy is the handoff, not the application.
Order to cash is usually drawn as a straight line: quote, order, fulfil, invoice, collect. In practice it is five systems and four joins. The quote lives in the CRM. The order lives in an order entry or inventory system. The shipment lives in the warehouse or a 3PL's portal. The invoice lives in the books. The payment lives in the bank and has to be matched back to the invoice.
Each of the four joins is a handoff, and in most companies of 30 to 300 people each one is done by a person: an email with a PDF attached, a re-keyed order, a shipping confirmation someone forwards to accounting, a remittance somebody matches by hand. The applications are automated. The joins are not.
Every day a record waits at a join is a day added to your days sales outstanding, and none of those days shows up as a cost on anyone's report. They show up as a line of credit you did not expect to need.
The four handoffs, and what breaks at each one.
Here is where the cash actually stalls, in the order it happens.
Quote to order.
The quote was priced from a CRM that does not know current cost or current stock. When it converts to an order, someone re-keys it into the order system, and the two records are already different: a discount that was approved in a conversation, an item that was substituted, a ship-to address that changed. The order that fulfils is not the order the customer agreed to, and the dispute arrives after the invoice.
What breaks: the record of truth for price and item is split between two systems. What it costs: every disputed invoice adds the length of the dispute to the payment date, and the dispute is usually discovered when the customer refuses to pay.
Order to fulfilment.
The order exists, but the warehouse or the 3PL learns about it through an export, a portal upload or an email. Partial shipments, back orders and substitutions happen on the warehouse side and are recorded there, if anywhere. The order system still shows the original order.
What breaks: nobody owns the fact that what shipped differs from what was ordered. What it costs: the invoice is raised for the wrong quantity, or raised late because somebody is waiting to confirm what actually left the building.
Fulfilment to invoice.
This is the handoff that stalls the most cash, because it is the one most companies leave entirely to a person. The goods ship on Tuesday. The invoice goes out when accounting gets the shipping confirmation, checks it against the order, and keys it into the books. In a busy week that is Friday. In month-end week it is next Tuesday.
What breaks: the invoice date is decoupled from the ship date by a queue in somebody's inbox. What it costs: every day between shipment and invoice is a day of days sales outstanding you gave away before the customer even received the bill. On thirty-day terms, invoicing five days late is a seventeen percent extension of credit you did not intend to offer.
Invoice to cash.
The payment arrives. It is short by a disputed line, or it covers three invoices in one remittance, or the reference is a purchase order number rather than an invoice number. Someone matches it by hand. Until they do, the customer shows as overdue, collections calls a customer who has paid, and the cash sits unapplied.
What breaks: the payment record and the invoice record have no shared key. What it costs: unapplied cash is cash you cannot report, and a collections call to a customer who paid is a relationship cost with no line item.
The missing piece is one record moving through five systems.
None of the four handoffs is fixed by buying a better application for the step on either side of it. They are fixed by deciding, for each object on the path, which system may write it and how it moves to the next system as data rather than as a document.
The quote is the order. When a quote is accepted, it becomes the order without re-keying, carrying its approved price, its items and its ship-to, so the order that fulfils is the order that was sold.
The shipment writes back. What actually left the warehouse, in what quantity, on what date, is written to the order by the system that shipped it, not forwarded by a person.
The shipment raises the invoice. The invoice is created from the shipment record on the day of shipment, for the quantity shipped, because that is the only version of the order the customer will accept.
The payment matches itself. Remittances carry the invoice reference, and the ones that do not are matched by rule on customer, amount and date before a person ever sees them.
That is what order to cash automation means when it is done for the person paying for it: not a workflow tool bolted onto each application, but the connection layer between them treated as the product. We have written about the general version of this in why you are not buying applications but how they connect, and about the record-of-truth decision underneath it.
What you get when the joins are built.
Invoices dated the day goods ship, every time, which is the single largest lever most companies have on days sales outstanding and costs nothing in terms.
Disputes that happen before shipment, at the quote-to-order join, where they are cheap, rather than after invoicing, where they delay cash.
A collections list that is true, because cash is applied on the day it arrives and nobody calls a customer who has paid.
One answer to "where is that order" that the CRM, the warehouse and the books all agree on.
On Zoho, this is Zoho CRM, Zoho Inventory and Zoho Books connected so that each one owns its part of the record and none of them re-keys another's. The applications are the easy part. The joins are the work.
When not to do this.
If you ship a handful of orders a week and one person runs order entry, shipping and invoicing, the handoffs are that person's memory, and they work. Automate them when that person is about to be two people.
If your order-to-cash path already runs inside one ERP that your team actually uses end to end, you do not have four handoffs; you have one system. Fix the process inside it before adding anything around it.
If you cannot state, for each of the five objects on the path, which system is allowed to write it, connect nothing yet. An integration built over an undecided record of truth produces two systems overwriting each other, which is worse than the inbox.
How we know this.
Business systems integration is what we do, and the order-to-cash path is where we spend more of it than anywhere else, because it is where the money is visibly late. The commercial terms are built for exactly this kind of work: discovery is no-risk, so you pay for it only if you proceed to a build, and the estimate is guaranteed, so a join that turns out to be harder than we scoped is our cost, not yours.
Where this goes.
Order to cash is the first path where the handoffs become visible, because the cash makes them visible. The same four-join structure runs through procure to pay, through returns, and through the professional services version, where hours are the inventory. A company that has fixed the joins on one path has learned the discipline for all of them: one record, one writer per object, movement as data.
That is the operating model we build toward, and it is the reason we start with the path where the cost of waiting shows up on the balance sheet. If you want to know which of the four handoffs is holding your cash, a no-risk discovery is where we begin: we walk one order from quote to cash with the people who touch it and mark where it waited.
Find out where your orders, inventory and invoices stop agreeing.
In a no-risk discovery we follow an order from sale to shipment to invoice across your systems and show where it breaks and what one connected system would change. You pay only if you proceed. Or see how we approach it.
More on the same problem:
Comments