docRefinery vs Reducto

What each product is for

Both turn PDFs into structured data through an API, so the shortlist is reasonable. The orientation differs. Reducto is a document platform built around parsing: Parse, Extract, Split, Classify and Edit APIs, plus Studio for building pipelines without code. Its breadth of file types follows from that focus on preparing documents for retrieval and AI. docRefinery is built for extraction into business processes. It returns checked, structured data, and each vendor's layout is learned once and remembered. Both handle long documents, and both run inside your own network.

What docRefinery does that Reducto does not, or does differently

docRefinery suits flows where a few hundred layouts cover tens of thousands of documents: invoices from regular vendors, bank statements, purchase orders, loan and claims packets. Ten differences matter most against Reducto.

• Layout memory. Each sender's layout is learned once and remembered, and re-trained when a form changes, so a supplier's hundredth invoice is recognised on sight and a redesigned template is caught as new.
• You define the output. Each field has a plain-language instruction, a "Return as" name you choose as its output key, a type and a required flag, so results land in your system under your names, not ours.
• One output format for every document. Every field returns text, alias, field_confidence and location, and every table returns rows, so one parser handles invoices, statements and IDs alike.
• Long files, segmented. Long, multi-page files are handled at scale and segmented into the documents inside them, each classified into its type, so a loan packet arrives as separate, typed documents.
• Governed fields. Owner, admin and member roles decide what may be extracted, from 20 document types and 785 field definitions, and a caller cannot widen that.
• Null, never a guess. A field that is not on the page comes back null, and a value below the threshold you set per field waits in a review queue for a person.
• Evidence on by default. By default, every field and table cell carries a confidence score and its page and box unless you turn scoring off for a layout, so nobody has to remember to ask for it.
• Learns from corrections. A reviewer's correction becomes a suggested fix, tested on that layout's own documents before anyone applies it.
• Plug and play. API keys with expiry, revocation and an activity log; a workflow designed from one sample document; export to JSON, CSV or XLSX that always reflects corrections; and a Chrome extension, in early access and not yet in the Chrome Web Store, that runs a document and fills the web form in the open tab.
• Custom types at no charge, inside your network. Document types outside the catalogue are trained, hosted and re-trained for you at no charge, in our cloud, your private VPC or air-gapped on your own servers.

Where Reducto leads today

Reducto is more mature, and it is ahead on several things that decide deals. Where we are building the answer, we say so; none of it ships yet.

• File coverage. Reducto reads 30+ file types, slides included, a breadth that follows from its focus on parsing. If presentations or a long tail of office formats are a primary input today, pick Reducto. Wider office formats, presentations among them, are in development at docRefinery.
• Retrieval output. Reducto's parsing output is designed for language-model and agent pipelines. Our retrieval pipeline, Parse, is planned and not yet available; the roadmap shows where it stands.
• Parsing track record. Parsing is Reducto's core product. We publish no accuracy benchmark of our own, so we will not claim to beat it on parsing quality.
• Security review. Reducto says it is SOC 2 Type II audited, with the report on request, and signs BAAs for HIPAA workloads on Growth and Enterprise plans. It describes zero data retention options too. docRefinery's SOC 2 Type II audit is in progress and there is no report yet, so a buyer who needs a report today should pick Reducto. Ask us for the latest status.
• Published list prices. You can model a budget from its pricing page without talking to anyone. Past our free Starter plan, our rate comes from sales.
• Writing back to documents. Its Edit API fills finished PDF and DOCX files. We do not offer an equivalent.

How the pricing differs

Reducto charges by the page, per operation. Its list rates are $10 per 1,000 pages for Parse, $20 for Extract and $40 for Deep Extract, with parsing included in Extract, and a 20% discount on batch Parse and Extract jobs. Its docs add that dense extractions of over 100 fields a page may incur a per-field surcharge. Contracted customers may have different rates. docRefinery is free up to 100 pages a month, and paid plans are custom, quoted to your volume.