OCR
Data extraction for procurement: the price paid becomes an argument
Data extraction for procurement turns dormant documents into a searchable price history: item codes, quantities, unit prices, discounts and terms, line by line. The price actually invoiced becomes a negotiating position. The plan covers extraction; setting it against your negotiated terms is the next step.
The problem
The negotiated price is written down. The price paid has never been read back
The price list is signed, the price catalogue is up to date, and nobody can say whether the prices applied last month were the right ones. The prices actually invoiced live in separate documents: one per invoice, one per entity, one per supplier. None of them goes into a pivot table.
Automated reading is the first tool that comes to mind, and rightly so: what OCR extracts from a purchasing document (OCR being optical character recognition) lists the fields expected on each type of document. It delivers clean data. It does not yet deliver an argument.
A price read correctly is not a price checked. The argument comes from the comparison: the invoiced price against the rate schedule, the priced bill of quantities known in French as the BPU; the volume discount band against the volume actually ordered; the year-end rebate against the agreed calculation base.
What extraction makes negotiable
- Contract coverage rate
- What share of spend genuinely runs under a negotiated contract? Answering it means knowing what was bought, from whom, at what price and under which item code. Once the lines are extracted, off-contract purchases can be counted rather than assumed, and tied back to a supplier and a purchasing category.
- Bands and rebates
- Volumes accumulated by item code and by entity, discount bands crossed, year-end rebates to claim, ancillary charges that appeared in no price list: the values were already in your invoices and your rate schedules. They simply could not be read together.
- Preparing a renegotiation
- A renegotiation file is built on history: the spread in the price of the same item code between entities, the drift of a price catalogue across successive amendments, single-source dependency in a sensitive category. That is extraction, then comparison on equal terms.
A supplier does not argue with an impression. A supplier argues with a line, a date, an item code and a quantified variance, all taken from their own documents. It investigates, you decide.
Capture
Gather and consolidate all your procurement data.
Collection by dedicated mailbox, upload, SFTP, scan, API or ERP export. Reads every format (native PDF, scan, photo, spreadsheet, structured feed), then extracts line by line: supplier, references, quantities, unit prices, discounts, taxes, terms.
The end of re-entry. A procurement history you can finally query.
What Capture detects
- Long documents and multi-page tables
- Unknown layouts
- Several languages and currencies
- Poor scans and handwriting
Extract first, then check
The plan covers extraction. Checking your negotiated terms is the next step and is sized with you.
- Extraction: item by item, with the units, the discounts and the terms exactly as they are written.
- The check: every invoiced line set against the framework contract, the rate schedule, the price catalogue and the indexation clauses.
- The output: a history searchable by supplier, by item code and by entity, exportable to your analysis tools.
Nothing is lost. Everything can be checked, everything can be proven.
- The contract clause and the invoice line, highlighted side by side.
- Every extracted value stays linked to the exact place in the document where it was read.
- The same case produces the same decision, today as in six months: the rules are applied deterministically.
- No discrepancy is set aside in silence. Anything that matches no rule is raised, with its reason.
- The agent records what it did, in the order it did it: who, what, how much, when.
Frequently asked questions
Is a rate schedule read as reliably as an invoice?
They are not the same object. An invoice lists what was delivered; a rate schedule lists what was agreed, often across many pages, with units, minimum orders and revision rules. Both are extracted, then matched item code by item code.
Can prices from several entities and several currencies be compared?
Yes, provided the lines are brought back to a common base: units of measure converted, currency and exchange date made explicit, scope of service equivalent. That is the condition for a comparison that holds up in a committee.
Which documents does Zylio handle?
Invoices, contracts and amendments, price lists and rate schedules, quotes, purchase orders, goods receipts, credit notes, statements, as native PDF, scan, photo or structured format, in several languages and currencies.
Does the data need to be perfectly clean?
No. The Capture agent is designed for heterogeneous documents and incomplete reference data; structuring is part of the deliverable.
Measurable impact in every environment
More than 5 million procurement documents analysed
Between 1 and 7% of margin recovered
on the scope analysed
From 15 to 45% of time given back to teams, per FTE
depending on the scope and on data maturity
Zylio fits into your existing ecosystem.
The ERP runs the process. Zylio handles the exception and recovers the value that escapes it: invoices without a purchase order, line-by-line price discrepancies, duplicates and overbilling, off-contract spend.
Your data under high security.
Zylio meets the most demanding standards, and nothing is committed without your approval.
- Certifications
- Hosting
- Encryption
- Access
Read next
- Extract, then check, then proveThe pillar of the cluster: the full chain, from reading to evidence.
- Reading a contract and its rate scheduleTurning a signed PDF into a price grid you can genuinely check.
- Putting quotes on one gridOptions, variants, ancillary charges: comparing what is comparable.
- Several languages, several currenciesDate formats, separators, local taxes: what no longer blocks anything.
- Purchase orderThe reference for what you committed to: line item, quantity, negotiated unit price, lead time, cost coding.
See what this looks like on your own data
Twenty minutes, on a spend category of your choosing. We show you what the agents detect, with the evidence behind it.
- No commitment, on your own data
- Result in 3 weeks
- 20 minutes, no sales pitch
- Your data stays hosted in France
- No change of tool or process