Talent Acquisition
AI Resume Screening
Screens applications against role criteria and produces reviewer shortlists.
Intake & Extraction
AI Document Extraction for Recruitment
CVs, registration forms and client job briefs turned into clean ATS records, so consultants spend their time on people, not data entry.
Your consultants should be talking to candidates and clients, not retyping CVs. This build reads the CVs, registration forms and job briefs your desk handles every day and writes the fields you choose straight into your ATS. Anything it is unsure about waits in a review queue for a person. Your team still makes every judgment about who fits a role.
- Candidate records created in your ATS without retyping CVs
- Every extracted field carries its own confidence score
- Unclear fields held for a consultant to check first
$5,400
20–30 working days
Optional maintenance $270/mo
What it is
AI Document Extraction for Recruitment reads the documents a recruitment desk handles in volume, such as candidate CVs, registration forms and client job briefs, and extracts the fields you define into your ATS, each with a confidence score. It does not rank or screen candidates; it gets accurate data in.
Why businesses buy it
Keying CVs into an ATS by hand is slow, dull and easy to get wrong. A mistyped phone number or a missing work history can mean a lost candidate or an awkward call with a client. After this build, new CVs and forms become clean candidate records on their own, and consultants only touch the fields the system flagged. That frees time for sourcing and placements. The Scope Card spells out the document types, the field list and the ATS connection before you order, so you know exactly what you are buying.
Problems this solves
Consultants spend hours typing CV details into the ATS
The system reads incoming CVs and extracts the fields you define, such as contact details, job titles and dates. Each field carries a confidence score and checked data goes into your ATS.
Registration forms arrive incomplete and compliance chases them later
Recruitment validation rules flag missing mandatory fields on registration forms. Gaps show up in the review queue at intake, not weeks later during a compliance check.
Client job briefs sit in email instead of the ATS
Job briefs can be one of the three document types set up. Role, location and requirements are extracted into your ATS. The system does not rank or screen candidates.
A good fit if
Consultants or resourcers copy names, contact details, job titles and skills from CVs into your ATS by hand.
Hundreds of CVs arrive in every format, from tidy PDFs to phone-scanned pages.
Client job briefs come in by email and someone has to set up each role in the ATS from them.
You use an ATS such as Bullhorn, Greenhouse or Workable that can accept records through an API.
Not the right fit if
You need applications screened against role criteria with shortlists for reviewers. See AI Resume Screening for Recruitment.
You need candidate and client meetings booked from chat and email. See AI Appointment Setter for Recruitment.
Candidates and clients go quiet, and you need them followed up. See AI Follow-Up Engine for Recruitment.
CVs arrive mixed with contracts, timesheets and other paperwork in one inbox. Sort them first with AI Document Classification.
How it works
- Documents arrive in one place. CVs, registration forms or client briefs land in the applications inbox or upload folder you connect.
- Fields are read from any CV layout. Name, contact details, current title, employer, location, skills and other fields you define are found whether the CV uses columns, tables or plain paragraphs.
- You control what is captured. Up to 25 fields per document type, and you can leave out anything you do not want stored, such as date of birth or photos.
- Rules catch incomplete records. A CV with no email or phone, or a job brief with no location or rate, is held as an exception rather than creating a half-empty record.
- Clean data creates or updates ATS records. High-confidence fields are written to the candidate or job record in your ATS, with the original document attached.
- Doubtful fields wait for a resourcer. They review queued values beside the source page, correct them, and the record is then completed.
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.
- At least 20 real samples of each document type, such as CVs in varied formats, redacted if you prefer.
- The fields to capture from each type, what to leave out, and the rules for a complete record.
- Access to the applications inbox or upload folder.
- API access to your ATS, such as Bullhorn, Greenhouse or Workable.
- The resourcer or administrator who reviews queued fields.
- Admin access and API keys for both systems, never sent by email.
- One decision-maker who approves the field list and the go-live.
- 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 CVs, forms or briefs and field list confirmed.
Weeks 1–2
CV and brief extraction rules built from the Recruitment template.
Week 3
Applications inbox and ATS connected, review queue configured.
Weeks 3–4
100 sample documents processed, accuracy reported for every field.
Week 4
Your review and two rounds of changes.
Go-live
Live on incoming CVs, recorded handoff, support begins.
| Option | Delivery time | Price |
|---|---|---|
| Standard | 20–30 working days | $5,400 |
| Express | 15–25 working days | $6,200 |
| Rush | 10–20 working days | $7,000 |
What you receive at handover
- AI Document Extraction for Recruitment configured on one intake route and connected to your ATS
- Recruitment field sets and validation rules for up to three document types, such as CVs, registration forms and job briefs
- A human review queue for low-confidence fields (no candidate ranking or screening)
- A test report using recruitment scenarios, 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 give consultants more time for calls and less for data entry.
- Helps keep candidate records complete from the start.
- Makes new job briefs searchable in the ATS sooner.
- Time from CV received to ATS record
- Compare receipt and record-created timestamps for a month before launch and a month after.
- Candidate records missing mandatory fields
- Run an ATS report of incomplete records each month and compare with the pre-launch count.
- Job briefs logged the same day
- Compare brief receipt dates with ATS job creation dates, month on month.
Before you start: Count CVs, registration forms and job briefs received per week. Run an ATS report of incomplete candidate records before work starts.
These are things to track, not promised results. Outcomes depend on your data, your team and your market.
Running costs
The build is a single fixed price. Running costs are roughly $30 to $200 a month in your own accounts for OCR and AI usage plus storage, moving with the number of pages processed. Most CVs are only a page or two, so even high application volumes tend to stay well inside that range.
AI usage during build and testing is billed to your accounts too.
Ask about running costsCustom requirements
A fourth document type, more than 25 fields per type, a second intake route, or duplicate matching against existing ATS candidates is outside the Scope Card. Add a note at checkout for a written Change Quote.
Request a custom versionKeep it running
Monthly maintenance covers monitoring, fixes, updates, a health report and change hours each month, enough to add a field, adjust a completeness rule or handle a new client brief format. It starts after the 30 days of support.
$270/mo (5% of the build price a month, 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 CV parsing service usually bills per CV or per seat | An ATS integration partner quotes after discovery | Ours is one fixed price, shown on this page before you buy |
| Field choice | A fixed set chosen by the vendor | Depends on what was built | Ours captures up to 25 fields you choose and skips what you exclude |
| Incomplete records | Created as parsed | Depends on the brief | Ours holds them for a resourcer first |
| Delivery | Quick to switch on, fixed fields | Usually several weeks | Ours has a fixed delivery time shown before you buy, with faster options |
Questions about this product
Does AI Document Extraction for Recruitment score or rank candidates?
No. It extracts the fields you define from CVs and forms into your ATS and leaves every judgment about suitability to your consultants. If you want applications screened against role criteria with reviewer shortlists, AI Resume Screening for Recruitment is built for that job and can follow this build in the same workflow.
Which ATS platforms can the recruitment extraction build write to?
An ATS with API access, such as Bullhorn, Greenhouse or Workable, connected as the destination system. Extracted fields create or update the candidate or job record, with the source document attached. If your ATS does not accept records by API, a sheet can be the destination for your team to import.
Can we stop it storing personal details we do not need?
Yes. You choose the field list, so details such as date of birth, photos or marital status can simply be left out. For stricter handling of what remains, the PII Redaction Layer and Data Retention and Deletion Configuration are add-ons you can buy alongside this product, listed with their own prices.
Will it create duplicate candidates in our ATS?
It can if the same person applies twice, because matching new CVs against existing records is outside the standard build. Many agencies rely on their ATS's own duplicate checks. If you want the extraction to look up existing candidates first, add a note at checkout for a Change Quote.
How soon will CVs flow into the ATS automatically?
Once the build is live, every CV or form that lands in the agreed source is read and written to your ATS without anyone opening it. The delivery time for each speed is shown in the price and delivery panel on this page. It counts from a complete intake, so sending sample CVs, forms and your field list promptly keeps the date on track.
What does CV extraction cost to run at high volume?
Roughly $30 to $200 a month in your own accounts for OCR and AI usage plus storage, depending on pages processed. Short CVs keep the per-document cost low, and there is no per-candidate fee from us. Monthly maintenance is optional.
Let consultants recruit instead of retyping CVs.
Order AI Document Extraction for Recruitment.
Works well with
Intake & Extraction
AI Document Validation
Checks documents for completeness and consistency against rules.