Intake & Extraction
AI Document Extraction for Recruitment
Extracts structured data from documents into your systems, adapted for Recruitment businesses that key candidate CVs and compliance documents into their ATS…
Intake & Extraction
AI Document Classification
Every captured document is sorted by type, client and priority automatically, instead of someone opening each one to find out.
Before anyone can act on a document, someone has to open it and work out what it is, who it belongs to and how urgent it is. This build does that sorting. It reads each document's content and layout, tags it by type, client and priority against your categories, and routes it. Uncertain documents wait for a person.
- Documents tagged by type, client and priority automatically
- Confident results routed straight to the right place
- Uncertain documents held for a person to check
$3,900
20–30 working days
Optional maintenance $249/mo
What it is
AI Document Classification reads each document's content and layout, then tags it by type, client and priority against the categories you define, routing confident results and holding uncertain ones for review. Because it reads content rather than filenames, a vaguely named file still lands in the right place.
Why businesses buy it
Sorting documents by hand looks like a small job until you count it: every inbox check, every misfiled contract, every urgent letter that sat behind fifty routine ones. Mistakes here ripple forward, because the wrong team works on the wrong file. This build reads content rather than filenames, so sorting is consistent, and it never guesses on the uncertain ones. The written scope fixes the categories, rules and connected systems before you order, so the outcome is clear.
Problems this solves
Every incoming document has to be opened just to know what it is
Classification reads each document's content and layout, then tags it by type, client and priority. Your team knows what each one is and who it is for before opening it.
Badly named files end up in the wrong folder
Because it reads content rather than filenames, a file called 'scan001' still lands in the right place. Up to three document types are set up against your categories.
Urgent documents wait behind routine ones in a shared inbox
Priority tagging brings urgent items to the front of the line. Confident results route on automatically, and uncertain ones wait in a human review queue rather than being guessed.
A good fit if
Documents already land somewhere, an inbox or a folder, but nobody sorts them by type until someone opens each one by hand.
You handle a mix of document types, such as invoices, contracts and applications, that need to be told apart before anything else happens.
You want each document tagged with its type, and where relevant its client and priority, automatically rather than manually.
You're comfortable with uncertain classifications going to a person you named, rather than being guessed at and filed wrong.
Not the right fit if
You want document data read into your system, not just sorted. Start with AI Document Extraction, which covers intake, validation rules and review in one build.
Your documents aren't captured anywhere consistent yet, they arrive scattered across inboxes and desktops. See AI Document Intake, the stage before this one, built to capture and file them first.
You already know each document's type and just need the data pulled out of it. See AI Document Extraction, which reads content once sorting is already done.
You need documents checked for completeness or errors, not sorted by type. See AI Document Validation, a later QA stage in the same pipeline.
Your documents are specifically contracts needing key-term analysis, not general sorting. See AI Contract Intelligence, built for that document type alone.
How it works
- A document arrives through the intake route you connected, such as Google Drive, SharePoint or an email inbox.
- Classification reads the document's content and layout, not just its filename, to work out what kind of document it actually is.
- It's tagged by type, and where relevant by client or priority, against the categories and rules you defined during setup.
- Confidently tagged documents are filed or routed straight to the folder, sheet or system that category is set to reach.
- Anything the model isn't confident about lands in a review queue for the person you named, instead of guessing at a category.
- We test classification against 100 real sample documents before launch and hand you an accuracy report broken down by category.
What you provide within 72 hours
Your portal lists these the moment you order. Delivery starts the working day after everything is in and your access is checked.
- Your business information (company basics, hours and brand voice), shared during intake, or confirmation we can reuse what you gave us on an earlier order.
- At least 20 real sample documents for each type Classification should learn to tell apart, redacted if needed.
- The category, client and priority labels you want documents tagged with.
- Access to the destination system or systems tagged documents should be filed or routed into.
- The name of the person who reviews anything the model isn't confident about.
- Admin access and API keys for every connected system, shared by adding our team as a user or creating a key you control, never by email.
- One decision-maker who can approve the category definitions and sign off at launch.
- An API key for your own AI model account (for example OpenAI or Google Gemini), created for this build with a spending limit.
The 72-hour rule. Send everything on your checklist, including access, within 72 hours of ordering so your build starts on time.
Timeline
Start
Intake verified, sample documents and category definitions confirmed.
Weeks 1–2
Classification rules and tagging logic built for your document types.
Week 3
Source and destination systems connected and routing tested.
Weeks 3–4
100 sample documents processed, accuracy report shared with you.
Week 4
Your review and two rounds of changes.
Go-live
Live, recorded handoff session, support starts.
| Option | Delivery time | Price |
|---|---|---|
| Standard | 20–30 working days | $3,900 |
| Express | 15–25 working days | $4,600 |
| Rush | 10–20 working days | $5,250 |
What you receive at handover
- AI Document Classification configured on one intake route and connected to your destination system
- Categories for type, client and priority, covering up to three document types
- Validation rules and a human review queue for uncertain results
- A test report from 100 sample documents, with an accuracy report
- An admin guide and a recorded 60-minute handoff session
- 30 days of post-launch support
Expected benefits and how to measure them
- Can cut the time spent opening and sorting documents.
- Helps urgent paperwork reach the right person sooner.
- Makes misfiled documents less likely.
- Classification accuracy
- Start from the accuracy report on 100 test documents. Spot-check a weekly sample after launch.
- Share of documents sent to review
- Compare reviewed documents with total documents each week, using the queue counts.
- Time to handle urgent documents
- Compare arrival and action times for urgent items before launch and after.
Before you start: Count documents received weekly by type and note how they are sorted today. Record how long urgent items take to reach the right person.
These are things to track, not promised results. Outcomes depend on your data, your team and your market.
Running costs
The build price is one-time. Running Document Classification costs roughly $30 to $200 a month in your own accounts, covering OCR and AI usage plus storage, and it moves with how many document pages pass through classification each month.
AI usage during the build and testing is billed to your own accounts too, so the AI model relationship stays yours, not ours.
Ask about running costsCustom requirements
If you need more document types, more tag fields, or a second intake route than the Scope Card lists, add a note at checkout. You get a written Change Quote, and that work only starts once you accept it. Your original order runs on its normal timeline either way. Larger or bespoke projects go through the Custom AI request form instead.
Request a custom versionKeep it running
Monthly maintenance covers monitoring, fixes, updates, a health report and change hours each month, enough to add a new category or adjust how a document type is recognized. It starts after the 30 days of support end, bills monthly, and can be cancelled any month.
$249/mo (the minimum fee, includes 2 change hours)
Optional. Starts after the post-launch support period. Cancel any month.
How maintenance worksHow this compares
| Do it yourself with a SaaS tool | Hire a custom agency | Upmark Labs | |
|---|---|---|---|
| Price | A classification SaaS charges per document processed, and costs climb with volume | A custom agency quotes after discovery calls | Ours is one fixed price, shown on this page before you buy |
| Scope | A classification SaaS is a general tool you build and retrain categories in yourself | An agency agrees scope during the project | Ours is written on the Scope Card before purchase |
| Delivery | A classification SaaS is live once you finish training each category | An agency build usually takes several weeks | Ours has a fixed delivery time shown before you buy, with faster options |
| Sorting accuracy | A classification SaaS trusts its own default model unless you tune it | An agency build depends on what was scoped | Ours is tested against 100 of your real documents with a written accuracy report |
Questions about this product
How much does AI Document Classification cost?
One fixed price covers the Scope Card: your category rules, the routing logic, the review queue and the 100-document accuracy report before launch. You see the figure for each delivery speed in the price panel on this page, and it is confirmed once your Scope Card is agreed. Adding categories later is either maintenance work or a quoted change.
How is AI Document Classification different from AI Document Intake?
Intake captures a document from your inbox or folder and files it, with no sorting involved. Classification takes over once a document is already captured, reading its content to tag it by type, client or priority against categories you define. We suggest running Intake first, then adding Classification once capture is working reliably.
Does AI Document Classification extract data from the document too?
No. It tags and routes each document by type, client and priority, but it does not pull out structured fields. If you want data in your system, start with AI Document Extraction instead. It already includes intake, validation rules and a review queue. Classification suits teams that only need documents sorted and filed.
How much does AI Document Classification cost to run each month?
Roughly $30 to $200 a month in your own accounts, covering OCR and AI usage plus storage. The figure scales with how many document pages pass through classification each month. There is no per-document fee from us. Monthly maintenance is optional and mainly useful when you add categories or change routing rules.
What happens when AI Document Classification isn't confident about a document's type?
It's held in a review queue instead of being filed under a guessed category. The person you named checks it, corrects the tag if needed, and that correction can inform how similar documents are handled later through monthly maintenance. Nothing uncertain is routed automatically without a human check.
Is our document content safe with AI Document Classification?
Yes. Every credential and connected account is shared by adding our team as a user or creating a key you control during setup, never by email, and documents move only between the systems you connected. Classification reads content to assign a category; it doesn't store or share it elsewhere, and sensitive samples can be redacted before we see them.
How long does AI Document Classification take to launch?
The delivery time for each speed is in the panel on this page, counted from the day your intake is complete and access is verified. Most of the build goes into writing the category rules and testing them against your real documents, so a good spread of samples helps the accuracy report reflect what actually arrives.
Can I add a new document category to AI Document Classification later?
Yes. Adding a category within your existing three document types fits inside the monthly change hours that come with maintenance. Adding a fourth document type entirely is a bigger change and gets a written Change Quote first, with pricing confirmed before any work begins.
Stop opening every document just to find out what it is.
Order AI Document Classification and it sorts every document by type, client and priority on a fixed delivery date. The scope above is what you get, in writing, before you pay.
Works well with
Intake & Extraction
AI Document Extraction for Real Estate
Extracts structured data from documents into your systems, adapted for Real Estate businesses that re-key data from leases, sale agreements and property…
Intake & Extraction
AI Document Extraction for Construction
Extracts structured data from documents into your systems, adapted for Construction businesses that lose time keying subcontractor invoices, delivery tickets and site…