Troubleshooting receipt scanning

The Merchant field is blank after scanning.

The merchant name could not be read from the image. Common causes include low-contrast thermal paper, glare, or the business name being printed very small at the top. Try rescanning with better lighting, or type the merchant name manually.

The Amount field is blank after scanning.

OCR did not detect a recognisable total. This can happen if the total text is very small, the image is blurry, or the receipt uses an unusual layout. Enter the amount manually from the physical receipt.

The wrong amount was extracted (e.g., the subtotal instead of the total).

The scanner reads the most prominently displayed numerical value, which is not always the final total. Compare the extracted amount to your receipt and correct it by tapping the Amount field and typing the correct value.

The merchant name shows a code or legal entity name instead of the store name.

Some receipts print the registered business name (e.g., 'ABC Holdings LLC') rather than the trading name. Correct the Merchant field manually after scanning.

The 'Scan Document' option does not appear.

MileLedger requires camera access to use the document scanner. Go to iOS Settings → Privacy & Security → Camera → MileLedger and enable access. If the option still does not appear, force-quit and reopen the app.

The camera launches but does not detect the receipt edges.

Ensure the receipt is on a contrasting background (e.g., a white receipt on a dark desk). Increase room lighting. Move the phone closer or further from the receipt until the yellow detection overlay appears.

The scan result is blurry or unreadable.

Tap the viewfinder to lock autofocus before capturing. Make sure your phone is held steady. If the receipt is faded, try scanning under direct lamp light or attach a photo taken in better conditions from your photo library instead.

A PDF attachment does not trigger OCR.

OCR is applied to the first page of an attached PDF. Ensure the PDF page containing the merchant name and total is the first page. If the PDF is image-based (scanned), OCR should still run; if it is text-based, the extraction will use the embedded text layer.