See in seven steps how PEDIF captures information from PDFs and emails, checks it, and prepares it for the systems you already use. Recurring documents that have been set up can be processed automatically. New, changed or unclear cases are sent to focused review.
Step 1 of 7
Your business partners can keep sending documents as PDFs or email attachments. PEDIF receives the document and captures the information needed for further processing.
The process journey above shows in seven steps how a document travels through PEDIF, from the moment it arrives to the handoff into your systems. The following sections add background to each stage and explain what happens when a document does not follow the standard path.
Each step answers a simple question. Has the document arrived? Do we know its layout? Which values does it contain? Are they correct? Does anything need a human decision? Where does the data go? Can we trace what happened? Together, these steps turn a PDF or email into data that your systems can use without retyping, while keeping control over every document.
Your business partners keep sending documents in the way they are used to, usually as PDF attachments to emails. PEDIF receives them from a defined mailbox or interface and records the sender, the time of receipt and the incoming channel. Nothing changes for the partner, and nothing gets lost in individual inboxes.
PEDIF checks whether a document matches a layout that has already been set up and approved. Known layouts are assigned to their processing route. New or changed layouts are deliberately marked, because processing them with the wrong rules would lead to wrong data. This step is what keeps automation reliable when business partners change their documents.
For known layouts, the relevant header and line-item values are captured into the agreed fields: order or invoice numbers, dates, parties, articles, quantities, prices and references. The data is then checked against defined rules, such as mandatory fields, formats, totals and plausibility, and where applicable against master data. Only data that passes these checks moves on automatically.
Documents with findings are shown for review together with the specific reason, for example a missing reference or an unknown article number. The responsible person corrects or clarifies the case, and the document continues. Because only these cases need attention, the team no longer has to check every document, and exceptions do not disappear in an inbox.
Validated data is handed to the next process in the agreed format, for example as an ERP import file, an EDI message, XML or CSV, via API or as an e-invoice. For every document it remains traceable when it arrived, which layout and rules were applied, which values were captured and where the result went. This makes the process verifiable for internal checks and audits.
Imagine a regular customer sends a purchase order as a PDF attachment in the morning. PEDIF picks it up from the order mailbox, recognizes the customer's known layout and captures the order number, delivery date and twelve order lines. The checks show that all mandatory fields are present and that every customer article number can be mapped to an internal article. The order is handed to the ERP as an import file, and the order desk sees it in the system without having typed a single value.
Later the same day, a new customer sends an order in a layout that has never been seen before. PEDIF marks it as new and shows it for review. The order desk processes it, and because the customer is expected to order regularly, the layout is set up as a new fingerprint for the next time.
The steps are deliberately arranged so that problems are caught as early as possible. A document with an unknown layout is stopped at recognition, before any values are extracted with the wrong rules. Values that do not pass the checks are stopped before they reach the target system, where errors would be much harder to correct. And because every step leaves a trace, it is always possible to find out later why a document was processed in a certain way. This order is what allows automation and control to work together instead of against each other.
Not every document arrives as a neat PDF attachment. Some customers write their orders directly into the email body, others send spreadsheets or Word files, and some emails contain several documents at once, for example an invoice together with a delivery note. Several attachments in one email are separated and each document is assigned to its own route. Inputs without a known PDF layout, such as order text in an email or a spreadsheet, can be prepared with pedif+ and shown for focused review, so they enter the same process instead of being handled outside of it.
This matters because the share of such inputs is often higher than expected. Bringing them into the same process means that the team works with one list of cases, regardless of how a document arrived.
