Every company that ships a product has had the same afternoon. Someone in accounting notices a customer who has not paid, looks at the order, and finds that no invoice was ever raised. The goods left the building eight weeks ago. The customer never complained, because no one complains about a bill they did not receive.
It is tempting to treat this as carelessness. It is closer to arithmetic. The order to invoice process has several stages and several people, and any process with enough handoffs will, on some proportion of orders, lose one. The useful question is not how to be more careful. It is how to find the losses the day they happen.
An order passes through a few events: it is accepted, picked, shipped, and eventually invoiced. In a small company the invoice is triggered by a person or by a status change, and either can fail in ways that leave no trace.
A partial shipment is the common one. The order ships in two parts, the first part triggers an invoice for the whole order or for nothing, and the second part has no trigger at all. A substituted item ships under a different code and the invoice rule only looks for the original. A drop-ship order is fulfilled by a supplier and nobody tells accounting that it left. A returned item is replaced, and the replacement is a shipment with no sales order behind it. An order is edited after the invoice was drafted, and the draft is never sent.
None of these is dramatic, and each leaves a shipment on one side of the business with nothing matching it on the other. The shipping system knows it left. The accounting system has no record that anything is owed.
We described where cash stalls after an invoice exists in order to cash: the four handoffs. This is the earlier gap, the case where the invoice does not exist at all, and it has a simpler fix.
The check is a single query run every day. It lists every shipment record with no invoice line pointing back at it. That is all. In most systems it takes an hour to write, and it can be run in the shipping tool, the accounting tool or a report that joins the two.
Three details make it work.
It looks from the shipment, not from the order. A check that starts from orders asks whether each order has been invoiced, which is true for a partially shipped order that has one invoice. A check that starts from shipments asks whether each shipment is covered, which is the right question.
It has an age. A shipment from this morning with no invoice is normal if invoices are raised in a batch at close of day. A shipment from five days ago is a problem. Set a limit, such as two business days, and the list shows only what has exceeded it. A list with fifty items a day is ignored. A list with three is worked.
It has an owner. One named person looks at it every morning. Not "accounting". A person, and a second person who covers their absence. The owner's job is not to raise every invoice but to find out why each item is on the list and to say which of the usual causes it was: partial shipment, substitution, drop-ship, replacement, edited order.
For the first two weeks the list is a clean-up. It finds every historic shipment that was never billed within the window you set, and the amount is usually a surprise. After that it becomes a diagnostic, because the causes repeat. If most of the items are partial shipments, the invoice trigger needs to fire per shipment. If most are substitutions, the invoice rule needs to key on the order line and not the item code. If most are drop-ships, someone has to be told when the supplier ships.
That is the real value. Each recurring cause is a rule that can be changed once, so the item stops appearing. After a few months the list is short and the changes behind it are a small set of rules in the shipping and invoicing steps, written down and owned.
The reverse check is worth running weekly as well: invoices with no shipment behind them. Those are the other kind of disagreement, billed but not sent, and they show up later as credit notes and disputes. In quote to cash we covered the case where the invoice exists and does not match what was quoted. The two checks, shipments without invoices and invoices without shipments, bracket the whole gap between the count and the bill.
On Monday the list shows four items older than two business days. Two are partial shipments whose second part never triggered an invoice. One is a replacement sent after a return, with no sales order behind it. One is a drop-ship the supplier fulfilled on Thursday. The owner raises the first two, creates a zero-value order for the replacement so the shipment has a home, and asks purchasing to tell accounting whenever a supplier ships.
By Friday the list has one item. The following month the partial-shipment rule has been changed so the invoice fires per shipment, and the first cause no longer appears. Nobody was blamed, and the number of items on the list tells you how the process is getting better.
A leak found at year end is expensive for reasons beyond the amount. The customer has forgotten the order, so the late invoice is a surprise and sometimes a dispute. The goods may have been returned, resold or written off. The people involved have changed, and the explanation of why it happened is gone. Some of it will be written off because chasing it costs more than the invoice.
The same item found within two business days costs a few minutes. The customer remembers the delivery, the shipment record is fresh, the person who handled it can say what happened, and the invoice goes out as a normal invoice.
An agent can run the check, group the results by cause and draft the invoices for review. That is a good use of one, because the rule is precise and the action is reversible: an invoice can be drafted and held until a person approves it. What it cannot do is decide that a drop-ship does not need a trigger or that a substitution should be billed at the original's price. Those are rules, and rules have owners.
We build the shipped-not-invoiced check, the age limit and the owner's morning list into the order and accounting systems a company already runs, and trace each recurring cause back to the rule behind it. Discovery is no-risk: we run the check across the last quarter of your shipments and show you what was never billed, and you pay only if you go ahead. The build carries a guaranteed estimate, so if we estimate low, we absorb the difference.
The process has many stages. The control needs only one question, asked every morning: what left the building that nobody billed?
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:
"Something isn't working." "We're outgrowing our tools." "We're afraid to make the wrong move." No-Risk Discovery is a short, practical conversation that gets you clarity before you commit to anything big. We'll tell you if we're a fit. If we're not, we'll tell you that too.
Say who you are and ask. A person replies within one business day.
