Snap a photo of any document — paper receipt, vendor invoice, handwritten list, business card, table from a printout — and get a clean Excel file. iPhone and Android photos work directly, no scanner app required, no quality settings to fiddle with.
Convert your first photo — free
A phone photo of a paper receipt isn't a clean scan. There's perspective distortion (the receipt is photographed at an angle), uneven lighting (one side lit by a window, the other shadowed), motion blur, and the receipt itself is often crumpled or curled. Generic OCR tools produce gibberish on photos because they were tuned for flat-bed scanner output.
Field reps photographing receipts on the road, contractors snapping photos of vendor invoices at the jobsite, and small-business owners shoeboxing receipts as photos all face the same problem: photos go in, garbage data comes out. Manual transcription stays the fallback even with 'OCR' tools.
PDFExcel's image pipeline does perspective correction, lighting normalization, and motion-blur compensation before running OCR. The OCR is tuned for the documents users actually photograph — receipts, invoices, handwritten notes, business cards, paper lists. After OCR, the same AI that reads native PDFs reads the OCR text and extracts structured rows and columns.
iPhone photos (HEIC or JPEG), Android photos, screenshots, and scans from mobile scanner apps (Adobe Scan, Genius Scan, Microsoft Lens) all work. Drop a single photo or a camera-roll selection (as ZIP) for batch processing. Same output structure as native PDFs — Excel with structured rows.
The image pipeline is tuned for phone photos specifically — perspective skew, uneven lighting, motion blur, crumpled paper. Same accuracy ceiling as scanned PDFs once the image is normalized.
Most OCR tools were built for flat-bed scanner output. PDFExcel was built for photos taken in real-world conditions — on the road, at the jobsite, in the warehouse.
Each photo becomes one row with merchant + date + amount + category. Drop straight into your expense report or business bookkeeping.
| # | Source | Date | Merchant | Amount | Tax | Category |
|---|---|---|---|---|---|---|
| 1 | IMG_4218.HEIC | 03/02/2025 | Blue Bottle Coffee #14 | $9.18 | $0.68 | Meals |
| 2 | IMG_4219.HEIC | 03/04/2025 | Uber | $32.45 | $0.00 | Transport |
| 3 | IMG_4220.HEIC | 03/06/2025 | Office Depot | $153.55 | $11.37 | Office |
| 4 | IMG_4221.HEIC | 03/09/2025 | Hilton Garden Inn | $213.57 | $24.57 | Lodging |
| 5 | IMG_4222.HEIC | 03/14/2025 | Whole Foods Market #225 | $57.07 | $4.23 | Meals |
Field sales reps and contractors photographing receipts on the road, small-business owners scanning paper documents with a phone, attorneys with case-file paper documents, students/researchers digitizing handwritten notes.
Photos receipts on the road for a week. Drops the camera roll into PDFExcel on Friday, exports to the company expense template. Files Friday's report in 10 minutes instead of an hour.
Receives a paper invoice from a material supplier on-site. Snaps a photo, converts to Excel, drops into Sage 100 Contractor as a draft AP entry — without going back to the office.
Years of handwritten observation notebooks need to become a structured database. PDFExcel handles each notebook page as a photo, extracts dates + observations into rows. Saves months of manual transcription.
Yes. iPhone HEIC photos upload and convert directly. Same workflow as JPEG. The image pipeline auto-detects format.
Yes. Output from Adobe Scan, Genius Scan, Microsoft Lens, and other mobile scanner apps works directly. They produce cleaner-than-camera-roll images so accuracy is typically higher.
Printed/typed handwriting (block-style) usually works. Cursive handwriting is hit-or-miss — the OCR will attempt it but you should review handwritten fields. Dates, amounts, and proper nouns are typically the most reliable handwritten extractions.
10 photos per month, free, forever. Most personal expense reports (5-15 receipts) fit comfortably in the free tier. Plans from $69/month for 50 documents.
Phone photos are noisier inputs (perspective + lighting + blur) so accuracy is typically a few percentage points lower than clean scans. On critical fields (dates, amounts) accuracy is still 95%+ for clear photos. We recommend a quick visual spot-check on photographed receipts before importing to bookkeeping.