KYC and Onboarding: Clean Document Capture Before Verification

Onboarding a new customer usually means asking for documents: an ID card, a passport, a proof of address, a company registration. The user photographs them with a phone, and those photos enter your verification pipeline.

Identity verification and KYC checks are only as good as the document images they receive. A passport photographed at an angle, with a glare across the chip, or a proof of address half in shadow can fail verification, trigger manual review, or get silently rejected by an automated system. Each failed attempt is a user dropping out of onboarding.

Why document photos fail KYC checks

Verification systems — liveness checks, face matching, document authentication — are trained on clean, flat, well-lit document captures. Real-world phone photos are not that:

• Perspective distortion — the document photographed from an angle instead of head-on
• Reflections and glare — light bouncing off laminated cards and passport covers
• Shadows — the document held under a desk lamp or in a car
• Cropped edges — a corner of the ID outside the frame, cutting off the MRZ or expiry date
• Low resolution — the document occupying only part of the frame

Any of these can flip a verification result. In regulated industries, a "soft fail" often means manual review — and manual review is cost and delay at onboarding scale.

Where a scanning API fits

A scanning API normalizes the document before the verification engine sees it:

Phone photo → scan API → clean, flat document → KYC verification / OCR

The scan step does not verify anything. It makes verification possible:

• Detects the document in the frame
• Corrects perspective so the document is rectangular and head-on
• Removes glare and shadows so text and security features are legible
• Returns a clean, consistent image every time

The result: verification systems receive input shaped like their training data, instead of whatever the user happened to produce.

The integration is one endpoint

ScanKit exposes POST /scan/crop. Your onboarding backend sends the photo and gets a clean document back.

import requests

def clean_document(photo_path: str, api_key: str) -> bytes:
    with open(photo_path, "rb") as f:
        r = requests.post(
            "https://api.scankit.io/scan/crop",
            headers={"X-API-Key": api_key},
            files={"file": f},
        )
    r.raise_for_status()
    return r.content  # clean document ready for verification

Call it before your verification provider, on both mobile app uploads and web uploads. One call, one clean document, and your KYC provider sees what it expects.

What onboarding teams get

• Higher pass rates — clean documents verify on the first attempt more often
• Fewer manual reviews — less "resubmit a clearer photo" friction
• Consistent records — every submitted document is a clean, upright, searchable file
• Faster onboarding — fewer retries means shorter time-to-first-transaction

Privacy and where the data goes

Identity documents are the most sensitive data a company handles. In the EU, the processing location is a compliance question before it is a technical one. When you add a scan step, check three things:

• Is it hosted in the EU?
• Is the upload encrypted in transit?
• Is the document deleted after processing?

ScanKit is built on those three answers: EU hosting, TLS in transit, deletion after processing. Only the clean scan leaves the request — the original photo is not stored.

---

The bottom line: KYC verification is only as reliable as the document images it receives. A scanning API turns a phone photo into a clean, flat, verification-ready document before your provider ever sees it. One endpoint, one call, fewer dropouts.

Try it with 50 free credits, no credit card required: create a free ScanKit account.