Complete order documentation without manual assembly

Half past one, the courier is due at three, and three orders are waiting in the warehouse to be shipped. The person responsible for the paperwork begins the ritual: labels are printed from a separate template, the quality certificate is filled in from the inspection results, the operating manual is dug out of the folder next to the product, the raw material certificate from an email from the supplier, plus the delivery note and the declaration of conformity, because this client requires one. Then everything has to be put together into one set: some pages scanned, some stitched together in a free PDF tool. A dozen or so minutes of clicking for a single shipment. And all over again for the next one.

Why manual assembly fails

The first problem is time. Assembling the documents grows in direct proportion to the number of shipments, and it is usually done by someone who also has transport bookings, goods issue and phone calls on their plate. On a day when shipments pile up, the documents are put together in a hurry, and in paperwork a hurry costs the most.

The second problem is version errors. The set ends up containing a declaration from before the template changed, a certificate with the results of a different batch, or a raw material certificate from the neighbouring order. The client spots it at goods receipt and holds up unloading or payment until the missing piece arrives. The goods have arrived, yet formally there is no delivery.

The third problem is knowledge locked in one head. The fact that client A requires a declaration of conformity, client B wants the certificate in two copies, and products from one particular group get a raw material certificate added, is remembered by a single person. When that person is on holiday, a “red light” comes on in the company: someone assembles the set by guesswork, and only an email from the client reveals what was missing.

A solution in three parts

Part one: the set is defined once, next to the data. Against the product and against the client, the system stores the list of documents that make up a shipment, together with the order of the pages. The client's requirements are recorded in the data, as permanently as a product code or a price. The system remembers, so a new person in dispatch assembles the correct set from day one.

Part two: a mechanism that produces a single file. The system creates a PDF document itself and keeps adding pages to it until the set is complete. The file takes in both pages generated straight from the order data (labels, delivery note, quality certificate) and ready-made files from outside, for example the raw material certificate stored against the material batch at goods receipt, or the operating manual kept against the product. One run produces one multi-page PDF, laid out exactly according to the definition from part one.

Part three: launch without a person, and an archive against the order. The mechanism starts the moment the order is released for shipment, and it can run on the server, so the document is produced even when nobody is sitting at a computer. The finished file is saved against the order record and goes wherever it is needed: to the warehouse for goods issue, by email to the client, to the archive. A year after delivery, the complete set of documents is found with one click against the order, with no searching through binders.

What this means for the business owner

Do the sums on your own numbers. How many shipments leave the company in a week? How many minutes does it take to assemble the documents for one of them? The product of those two numbers is the hours of work spent every week on an activity the system performs in the background. Then there is the second line of the calculation, harder to measure and usually more expensive: the cost of a shipment held up at the client by a missing or incorrect certificate, the cost of re-sending documents, and the impression such a slip makes on a customer who audits their suppliers. The mechanism described above removes both lines: assembly time drops to practically zero, and version errors disappear, because every document is produced straight from the order data.

Where to start

Start with an inventory of the current state: which documents accompany a shipment, where each of them comes from, who assembles them and how long it takes. Such a list takes an hour to put together and immediately shows which documents the system can generate itself and which only need to be stored against the data (such as certificates from suppliers). Then it is worth walking the path of a single order from release to shipment and marking every point where a person retypes, prints or stitches something together. That produces a ready specification on the basis of which a simple application automating the whole process can be built.

If you would like to see what such a mechanism would look like in your company, write to me or book a meeting.

I describe more mechanisms of this kind in the Knowledge section.