An Invoice Held Up Overnight by One Excel Upload

Sales data was complete by the afternoon, yet invoice creation still waited for an Excel upload the next morning.

Illustration of sales data moving from manual upload to an automatic connection with the accounting system. — The invoice follows the sales confirmation without waiting for the morning upload.

Conceptual illustration. The case is real; company details are disguised.

On delivery day, drivers returned the receiving documents until about four in the afternoon. Checking and confirming the documents continued until about six. That work took place at the depo, the site where goods are prepared before they are delivered. Customer, item, price, and proof-of-receipt data were all complete. Yet the sales invoice to the business customer had not yet appeared in the accounting system used to process billing.

The data was only moved at eight the next morning. The IT team exported the sales, then uploaded an Excel file to Accurate through a subscription-based intermediary application. The billing team began processing once the invoice appeared there. Each part finished its work according to the flow, but between the afternoon confirmation and the morning upload there was an overnight gap built into the schedule for moving the data.

Illustration of documents already checked in the afternoon while the data waits to be uploaded to the accounting system. — Complete documents did not immediately set invoice creation in motion.

There was another problem once the invoice existed. The receivables summary showed the amounts still unpaid, but it did not separate invoices that had been sent from those that had not. The work of collecting and the work of waiting for payment sat in the same view. To manage collection, the team needed the sending status of each invoice in addition to the value of open receivables.

As head of the IT Division, I worked with the team to connect the two systems directly. Moving the upload earlier or adding a night shift would keep the same manual step. Issuing invoices from files outside Accurate was not an option at the time either. The IT team changed the data flow so that confirmed sales move to the accounting system without waiting for the next upload.

The IT team built a flow to send item, customer, and sales data from the operational system to Accurate. Item and customer data are provided first so that a sale has the right reference when its invoice is created. After confirmation, the sales data flows through that connection. Invoice creation follows the readiness of the transaction, replacing the daily sequence of exporting and uploading a file each morning.

Illustration of the overnight gap between complete sales documents and the transfer of data to the accounting system. — The upload schedule inserts waiting time between pieces of work that are already done.

The connection also carries billing and payment status back from Accurate to the operational system. The invoice summary by customer group then shows Total Billed and Total Not Yet Billed alongside the receivables value. The billing team can read which work has already been sent to customers and which part still needs processing. The movement of data is followed by news of what happens next to the transaction.

Since it went live in July 2023, the daily Excel upload has left the invoice-creation flow. Confirmed data goes in without waiting for the next morning, and invoices are ready to bill sooner. The subscription to the intermediary application was not renewed. The work of moving files was replaced by a system-to-system connection built by the IT team, while document checking remains a separate step before a sale is confirmed. In July 2023, that summary produced the first figure that could be read: 87.3% of unpaid invoices had been billed. Before the billing status existed, that share was not visible. The summary separates invoices that have been sent from those that still need work. With that status, collection work had a measure that had not appeared in the receivables summary before.

An automatic flow needs a way to follow transactions that have not yet made it across. The IT team maintains five sync flows between the operational system and Accurate, checks the failure log every day, and follows up on invoices that are held back. The system retries data that failed to send. In the 2026 flow, the check runs every three minutes and sends only sales that are ready and do not yet have an Accurate number. Finance Internal, which verifies sales with a ready-to-sync status before they are sent, checks that data before it enters the accounting system.

Illustration of sales data moving automatically from the operational system to the accounting system. — Sales data goes to Accurate; billing and payment status returns to the operational system.

The division of functions between the systems evolved in March 2026. Invoices are issued directly from the operational system, while Accurate is used to exchange invoice files with customers and to record payment settlement. The role of the system-to-system connection changed with it: moving data is still needed for bookkeeping and payment administration. The finance team issues invoices from the operational system, with follow-up status still tied to the sales transaction.

My lesson: waiting time between systems needs to be examined as part of the business process. Once the data is complete, the schedule for moving files should no longer decide when the next piece of work begins.