Business card scanner API

Turning a stack of business cards into CRM contacts is exactly the kind of tedious data entry software should handle. The AIDataParser business card scanner API takes a photo of a card and returns clean contact JSON: full name, job title, company, email, phone, website, and address. It reads the card as an image, so logos, colored backgrounds, and unusual layouts don't trip it up the way template OCR does.

Because you define the schema, the output drops straight into your contact model — no mapping layer, no cleanup. One card, one API call, and you have a structured lead ready to sync to a CRM or enrich further. Multiple phone numbers or emails come back as arrays so nothing gets dropped, and the confidence flag warns you when a glossy or low-light photo needs a second look.

For event apps, sales tools, and networking-agent workflows, this is the fastest path from a physical card to a usable contact record. Send the photo by upload or base64, get JSON back for one credit, and skip the manual typing entirely. Nothing is stored after the request, so contact data stays yours.

Get 50 free credits

Fields you can extract from a business card

namestring

Person's full name.

titlestring

Job title.

companystring

Organization name.

emailstring

Email address.

phonestring

Phone number(s).

websitestring

Company URL.

Example request

POST the business card by URL, upload, or base64, with a JSON Schema describing the output you want.

curl -X POST https://aidataparser.com/v1/parse/document \
  -H "Authorization: Bearer adp_live_..." \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://example.com/business-card.jpg",
    "schema": {
      "type": "object",
      "properties": {
        "name": {
          "type": "string"
        },
        "title": {
          "type": "string"
        },
        "company": {
          "type": "string"
        },
        "email": {
          "type": "string"
        },
        "phone": {
          "type": "string"
        },
        "website": {
          "type": "string"
        }
      }
    }
  }'

Example response

The data field is guaranteed to match your schema.

{
  "object": "parse.document",
  "data": {
    "name": "Priya Natarajan",
    "title": "Head of Partnerships",
    "company": "Skyline Labs",
    "email": "priya@skylinelabs.example",
    "phone": "+1 212 555 0198",
    "website": "skylinelabs.example"
  },
  "confidence": "high",
  "review_needed": false,
  "credits_charged": 1,
  "credits_remaining": 49
}

FAQ

Does it read cards with logos and colored backgrounds?

Yes. The card is processed as vision input, so branding, background colors, and non-standard layouts don't degrade extraction the way fixed-template OCR does.

Can it capture multiple phones or emails from one card?

Yes. Model those fields as arrays in your schema and every value on the card is returned, so a card with both a mobile and office line loses nothing.

What image formats are supported?

PNG, JPEG, WebP, and GIF, sent by upload, URL, or base64. A low-light or blurry photo lowers confidence and sets review_needed rather than returning bad data.