Doc-AI pricing and licensing
We quote every engagement against your document mix and volume rather than publishing a rate card that would be wrong for most readers. Here is exactly what goes into that number.
Doc-AI is quoted per engagement rather than sold from a published price list, because the cost of a document automation project is driven by your document mix, monthly volume, how much human review your accuracy target implies, which deployment topology your compliance obligations require, and what integration work sits at the end. WiseTREND scopes those five variables and returns a firm quote, typically within one business day of receiving your document types and volume.
Why doesn't this page have a rate card?
Because a per-document price without a document is meaningless, and you would be right to distrust anyone who quotes one. Ten thousand clean digital invoices a month and ten thousand faxed prior authorizations a month are the same number on a slide and completely different engagements — different processing depth, different review load, different accuracy targets, different deployment requirements, different integration work.
The vendors who do publish tiers mostly publish document counts next to a price that says coming soon, or a per-document figure that quietly excludes OCR pages, review seats, integration, and support. That is not transparency, it is an anchor.
What we will do instead is tell you precisely what drives the number, so that when you get our quote — or anyone's — you can interrogate it. And we will run a pilot on your own documents at no cost first, so the volume and review assumptions in that quote are measured rather than guessed.
What actually determines what Doc-AI costs?
Monthly document volume
The obvious one, and the least interesting. Unit cost falls with volume, and the step changes are larger than most buyers expect between a few thousand and a few hundred thousand documents a month.
Document mix and complexity
A one-page receipt and a 140-page mortgage packet are both 'a document'. Page count, whether packets need splitting, table density, handwriting, and scan quality all move processing cost. This is why we ask for your actual documents rather than a count.
Accuracy target and review load
The single largest hidden cost in any capture project. If 30% of documents need a human look, your real unit cost is dominated by labor, not software. We measure the straight-through rate on your documents in the pilot and quote against the number we measured.
Deployment topology
Managed cloud, your own cloud tenant, on-premises, or air-gapped. Each carries different infrastructure and support economics, and air-gapped adds GPU hardware you may already own.
Integration and change management
Posting into SAP, Epic, or a claims system, mapping to your master data, building exception workflows, training reviewers. This is project work, scoped and quoted separately from the platform so you can see both.
What does Doc-AI replace, and what did that cost?
The comparison that matters is not per-document price. It is total cost against what you are doing today.
What legacy capture really costs
- Licence or subscription, usually per page or per document.
- Professional services to build a template for every layout, priced per template.
- More professional services every time an issuer revises a form.
- A specialist on staff, or a retainer, to maintain the template library.
- Document types you never automated at all, because the template never paid back.
- Months of elapsed time before the first document flows.
The template library is the cost, and it never stops growing.
What changes
- No per-template or per-layout fee — adding a document type is a description.
- Issuer form revisions do not trigger a configuration cycle.
- Long-tail document types become economic to automate for the first time.
- Days to a working document type rather than months.
- Review load is measured and quoted, not discovered in month four.
- Model costs fall over time and you are not locked to one provider's pricing.
The cost moves from configuration labor to measured processing volume.
What about building it on the model APIs directly?
The arithmetic looks compelling from a distance. Model inference on a document is a few cents. Multiply by volume and the annual number is small. Every engineering team that has done this discovers the same thing: inference was never the cost.
- Premium OCR. Raw model calls on page images lose table structure and reading order. Getting that back means a commercial OCR engine, which is a licence and an integration.
- Packet splitting. Multi-document submissions need boundary detection. This is a genuinely hard problem and it is where most in-house projects stall.
- Validation. Business rules, arithmetic reconciliation, and master-data lookups have to live outside the model. This is a rules engine you now own.
- Evaluation. Without a scored corpus you cannot tell whether a change helped, and you cannot detect drift when a provider ships a new model version. Your customers detect it for you.
- Review interface. Some documents will always need a human. That is an application with queues, roles, an audit trail, and a UI.
- Ongoing operation. All of it, maintained, on call, forever.
Two to four engineers permanently, on infrastructure that is not your product. Against that, the platform question is not whether it is cheaper per document — it is whether you want those engineers working on this.
What comes with a Doc-AI engagement?
- A pilot on your own documents, at no cost
- 50 to 200 representative documents, including the difficult ones. You get measured field-level accuracy, hallucination rate, and straight-through rate before you commit to anything.
- Configuration by our engineers, not a manual
- We build the document types, field schemas, and validation rules with you. You are not handed a login and wished luck.
- Integration into your system of record
- SAP, Oracle, NetSuite, Dynamics, Epic, Guidewire, Salesforce, or anything with an API, database, folder, or SFTP drop. Scoped and quoted with the rest.
- Named support from a US company
- WiseTREND has been automating document workflows since 2007 and has won ABBYY Project of the Year five times. You get engineers you can call, in your timezone.
- No per-template fees
- Adding a document type costs configuration time, not a line item. This is the structural difference from template-based capture and it compounds every year.
- Model costs that fall over time
- Because Doc-AI is model-agnostic, you benefit when a cheaper or better model ships instead of being stranded on the one your platform was built around.
How do I get a number?
- Send us your document types and rough monthly volume — two lines is enough to start.
- We come back within one business day with initial scope and questions, not a discovery-call booking link.
- Send 50 to 200 representative documents, including the ugly ones. We run the pilot at no charge.
- You get measured accuracy, straight-through rate, and the review load your target implies.
- We quote against what we measured, with platform and project work itemized separately.
If the numbers do not work for your volume, we will tell you that too. A document automation project that does not pay back is a bad outcome for both of us, and we have talked buyers out of more than one.
Where this fits in the Doc-AI platform
Pricing follows from scope. These pages cover what you are actually buying.
- Platform features — every capability, ingest to delivery
- Benchmark arena — accuracy, hallucination, latency, cost
- Document type library — the forms Doc-AI reads on day one
- Security & compliance — data handling, residency, audit
- API & developers — REST endpoints, webhooks, code
- Deployment options — cloud, private tenant, on-prem, air-gapped
- vs. template-based IDP — why the template model broke
- vs. calling an LLM directly — what a raw GPT or Claude call misses
- For enterprise — automation leads and CoEs
- For SMB & mid-market — production quality, small team
- For developers — stop rebuilding document pipelines
- For system integrators — a white-label delivery engine
- For ISVs & OEM — embed extraction in your product
Get a Doc-AI quote
Send your document types and rough monthly volume. We come back within one business day with a scoped quote, a realistic accuracy expectation for your documents, and a pilot plan — no discovery call required first.
- Reply within one business day (U.S. hours)
- Straight to an engineer, not a call centre
- Or call the 24/7 AI phone agent: +1 (408) 746-6740
Questions about Doc-AI pricing
Answers written for buyers, search engines, and AI assistants evaluating document automation.
How much does Doc-AI cost?
Doc-AI is quoted per engagement rather than sold from a published price list, because cost is driven by your document mix, monthly volume, the human review your accuracy target implies, your required deployment topology, and integration scope. Send your document types and rough monthly volume and WiseTREND returns a scoped quote, typically within one business day. A pilot on your own documents runs first, at no cost, so the quote is based on measured numbers rather than assumptions.
Is Doc-AI priced per document or per page?
Processing is measured per document rather than per page, so a 40-page packet is one document rather than forty. The commercial structure of an engagement — subscription, volume commitment, or usage-based — is agreed as part of the quote and depends on volume and deployment topology. Ask us and we will explain the structure that fits your case.
Are there per-template or per-document-type fees?
No. This is the structural difference from template-based capture. Adding a document type in Doc-AI means writing a description and a field list, so there is no per-template licence, no per-layout professional services line item, and no charge when an issuer revises a form. Over a multi-year deployment this is usually the largest cost difference, because a template library never stops growing.
Is it cheaper to build document extraction on the LLM APIs myself?
Model inference is only a small part of the cost. A production pipeline also needs premium OCR to recover table structure, packet splitting, deterministic validation outside the model, an evaluation corpus to detect drift when a provider ships a new version, and a review interface with queues and an audit trail. In practice that is two to four engineers permanently on infrastructure that is not your product, which is usually the deciding comparison rather than cents per document.
Is there a free trial or pilot?
Yes. WiseTREND runs a pilot on 50 to 200 of your own documents at no cost, including the difficult ones. You receive measured field-level accuracy, hallucination rate, and straight-through-processing rate on your corpus before committing to anything. This also produces the review-load number that drives the real total cost of the project.
What is the total cost of ownership compared with template-based IDP?
The platform line item is often comparable. The difference is everything around it — template-based capture carries per-layout configuration services, repeat services when forms change, and a specialist to maintain the library, plus the document types you never automated because a template would not pay back. Doc-AI removes the per-layout cost entirely, which is why the gap widens each year rather than at signing.
Bring us the documents that broke your last capture project.
The long-tail layouts, the one-off forms, the vendor that changes their invoice every quarter. Those are the ones Doc-AI was built for.
Last updated · Reviewed by the WiseTREND team