Purchasing documents
AI standardization: your purchasing documents brought to one structure, line by line, with proof
AI standardization brings all your purchasing documents (quotes, price schedules, catalogues, rate cards, amendments) to one structure: same fields, same units, same references. Capture reads each document in its original format, recognises its structure and brings every line to the standard, keeping the link to where it was read. The result can be compared, checked and queried.
The problem
Your prices already exist. They are locked in documents
A framework contract, its rate schedule in an annex, a supplier catalogue, a price list updated every quarter, quotes every week: the same knowledge, in ten different structures, none of which can be confronted with the others without someone re-keying it first.
A twelve-thousand-line price schedule in PDF
The grid exists, it is signed, it is the reference. But it sits in a two-hundred-page annex and nobody confronts it with anything. 12,000+ price lines per framework contract, pricing rules included.
One article under three references
Manufacturer reference in the catalogue, supplier reference on the quote, internal reference in the ERP. Without alignment, the price history is in three pieces.
Units that do not talk to each other
The catalogue prices per carton, the schedule per unit, the quote per lot. Comparison requires a conversion the spreadsheet does not know.
The mechanism
Read, recognise, align: what the agent does on every document
Capture reads the document in its original format (native PDF, scan, image, spreadsheet, structured feed) and recognises its structure. For a quote: header, lines, options, terms. For a price schedule or rate card: items, units, prices, tiers, rebate rules, effective dates. For a catalogue: references, descriptions, packaging, list and negotiated prices. A hundred-page table, a merged column, a footnote that changes a price are not lost.
Recognition then aligns every line with your repository. The description is attached to an item of your taxonomy and, when it exists, to the article already known under another reference. The unit is made explicit and brought to the reference unit, factor displayed. Currency, effective date and pricing rule (tier, rebate, revision formula) are extracted as data, not as text. A ten-thousand-reference catalogue and a ten-line quote follow the same path.
Everything decided is visible: attachment, conversion, reason, confidence score. What is not certain is flagged, never guessed. The standardized document joins your repository where Compare can confront it with other offers, Compliance with invoices, and where you can find it, line by line, months later. The original stays one click away from every value.
How a contractual price grid is read and structured, rules included.
Which documents
Six purchasing documents, one structure
Quotes
Lines, options, charges, terms: every quote received joins the standard, whatever its supplier or format.
Price schedules and rate cards
Items, units, prices, volume tiers, effective dates: the grid becomes a reference you can confront, line by line.
Supplier catalogues
References, packaging, list and negotiated prices, aligned with your taxonomy and your known articles.
Supplier price lists
Every update read and dated; the old price and the new one coexist, with their validity period.
Contracts and amendments
Price, rebate, revision and penalty clauses extracted as rules, attached to the supplier and the period.
Purchase orders
Orders placed, in the same standard, so that the ordered price compares with the negotiated and the invoiced price.
The journey
Four steps, for every document
The same journey for a one-page quote and a thousand-line price list. You only step in on what is flagged.
- 01
Intake
Dedicated mailbox, upload, SFTP, API or export from your ERP. Each document is detected and attached to its supplier.
- 02
Transformation
The structure is recognised by a recent vision-language model; each line is extracted with its position in the document.
- 03
Standardization
Fields, units, references, pricing rules and dates are brought to the standard of your repository, transformation displayed.
- 04
Approval
Uncertain points are presented to you. You approve or correct; the standardized document is available in your tools.
What changes
What you get, by function
For procurement, a unified view of offers and negotiated prices, comparison possible on receipt, and tracking of contractual prices that no longer depends on who opened the annex. For finance, reliable reference data to prepare entries, check invoices and answer statutory auditors, with full traceability down to the source document. For operations, quick approval and less re-keying, on a single screen rather than across three tools.
Over time, it is a repository that lives: every standardized document enriches the price history, every correction is remembered for the supplier concerned, and no model is trained on your data. A price list update replaces the previous one without erasing it; a new quote is read against articles the catalogue already made known. More than 5 million purchasing documents analysed, all formats, languages and structures combined.
The use case: a signed contract becomes a price reference you can confront.
Your tools
Compatible with your ERP, without touching it
Zylio reads the exports and APIs of SAP, Sage, Oracle, Microsoft Dynamics 365, NetSuite, Cegid or Workday, and returns standardized documents in the format your tools expect: multi-format export, open API, standard connectors. Nothing is written into your ERP without your decision. Hosted in France, ISO 27001, SOC 2 Type II, full audit trail.
What Zylio reads, what it returns, and what it never modifies.
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
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
Which formats and languages are supported?
Native or scanned PDF, image, spreadsheet, email with attachments, structured feed, ERP export. Documents in several languages and currencies are read in their original language and currency; the necessary conversions are displayed with their rate and date. Multi-page tables and annexes are read in full.
Is a price schedule of several thousand lines processed in full?
Yes. The platform absorbs framework contracts of 12,000+ price lines, pricing rules included, and documents of 100+ lines in multi-page specifications and dense tables. Every line keeps its position in the original annex, so that any value can be checked against the source.
How are standardization errors handled?
Every line carries a confidence score. An unknown description, a missing unit or an ambiguous price are presented to you with the original document, so that you judge on the source rather than on the extraction. You correct in one click; the correction is remembered for that supplier and applied to the following documents, without any model being trained on your data.
Where do the standardized documents go?
Into your Zylio repository, where they can be confronted with one another and queried line by line, and to your tools by export or API, in the format they expect. Zylio writes nothing into your ERP on its own: you decide what goes in, when and through which channel, and the original document remains reachable from every value.
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
- Automatic standardization of quotationsEvery quote received, whatever its format, brought to a single, verifiable structure.
- Standardizing contractsBring every supplier contract into one common structure, without rewriting a single word, so you can compare and query them at will.
- Automatic integration of supplier quotationsThe quote received by email or portal, read, checked and returned to your ERP as structured lines, without re-keying.
- PDF extraction of purchase dataTurning PDFs into structured, consistent and sourced purchase data, ready for analysis.
- Supplier price listsThe pillar: rate schedule and price list control, line by line.
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