Scan receipts from photos, totals checked
Snap your receipts, drop the photos here, and get their data back with every total checked, in the browser, without an app.
Drop your files hereLet go to start
or paste with Ctrl/⌘+V
Try it 5× without an account. A free account includes 20 AI credits a month. See plans
What it does
Receipt scanner, explained
To scan receipts online, take a photo of each one with your phone (or scan them), drop the images on this page and choose Read. Each photo is made into a one-page PDF in your browser, text recognition puts the printed characters into a text layer, and the AI reads merchant, date, receipt number, payment method, items, subtotal, tax, tip and total, keeping the quote each value came from.
Read more
Photos of receipts are hard to read: thermal paper fades, receipts curl, and a 1 and a 7 look alike. So every receipt is added up: the items plus tax (unless tax is included in the prices) plus tip must equal the total. When a digit is misread, the receipt is flagged with the difference, and you correct it on the page while you can still see the quote.
For the best results, lay each receipt flat on a dark surface, fill the frame, avoid shadows, and take one photo per receipt. Pick the receipt's language before reading so recognition uses the right alphabet.
This works in any modern browser and needs no app. When you're done, download Excel or CSV, or an expense report with a total per currency.
How to do it
Scan receipts from photos, totals checked in 4 steps
1Your PDF
2Ask or summarize
3Answers that cite pages
- 1
Photograph your receipts
One receipt per photo, flat, in focus and well lit.
- 2
Drop the photos here
JPG, PNG, WebP or HEIC (where the browser opens it); up to 10 per run.
- 3
Read
Text is recognised, then read. Each receipt shows whether it adds up.
- 4
Correct and download
Fix flagged values, choose categories, then export Excel, CSV or an expense report.
Example
Before and after
A phone photo of a curled restaurant receipt: two dishes, Subtotal 42.00, Tax 3.78, Tip 8.00 handwritten, Total 53.78 handwritten.
Merchant, date, two items, subtotal and tax read from the print; tip and total left empty and the receipt marked “Not checked” until 8.00 and 53.78 are typed in, after which it adds up.
Settings & formats
What you can control
- Photo handling
- Turned upright from the photo's orientation tag and made into one PDF page in your browser.
- Text recognition language
- Choose the language printed on the receipt; English is added as a fallback.
- Check
- Items (+ tax when added on top) (+ tip) = total, and items = subtotal when printed.
Input
Output
Good to know before you start
- It reads photos you upload; it isn't a camera app and doesn't crop or flatten pictures for you.
- Faded thermal receipts and handwritten tips or totals often can't be recognised.
- HEIC photos work only in browsers that can open them (Safari can); otherwise export them as JPG.
- Up to 10 receipts per run; each photo is read as one receipt.
If something goes wrong
If something goes wrong
- “No text could be recognised”
- The photo is too dark, blurred or small. Retake it closer, flatter and in better light.
- The date is wrong by months
- Dates like 03/04 can be read either way. Correct the date with the date picker; your edit is used in every export.
Receipt scanner: questions people ask
Do I need an app?
No. Take photos with your phone's camera and upload them here in the browser, on the phone or later on a computer.
Can I scan several receipts in one photo?
No. Each photo is read as one receipt, so take one photo per receipt.
Is it accurate?
No reader is perfect on photos, which is why each receipt is added up and flagged when it doesn't match. The flag tells you which value to look at.
What does it cost?
Reading uses from 1 AI credit per page read (the Free plan includes 20 AI credits a month, Pro 500); scans and photos also use OCR pages (20 a month on Free). Exports are free and can be repeated.
What happens to my photos?
Page text is sent to our AI provider, OpenAI, under API terms that exclude it from model training. The files are saved to your PDFBetter documents, where you can delete them.