OCR and PDF search explained
Why you can search a scan in an app but not its exported PDF
An app can recognize words and keep them in its own search index without writing those words into the PDF it exports. That is why a phrase may find a page inside one scanner but return no result in another PDF viewer. FreeScanner is a clear example: it keeps OCR text in an on-device index for in-app retrieval, while its PDF export contains the scanned page images without an embedded text layer.
Short answer
Why you can search a scan in an app but not its exported PDF
App search and PDF text search use different data. FreeScanner runs on-device OCR so you can find recognized words and return to the matching page in its library; its exported PDF is built from the scanned page images and does not contain that OCR text. Search a phrase inside the app, export the PDF, then search the same phrase in a separate viewer. If only the app finds it, use an OCR tool that explicitly adds a text layer when you need a portable searchable PDF.
Decision brief
What to remember
- An app may keep OCR results in a private index that is separate from the exported document.
- A PDF is independently searchable only when its file contains text, usually as a layer aligned with page images.
- FreeScanner's on-device OCR search works in its library; its exported PDFs contain page images without that text layer.
- If you need the same search to work in other viewers, apply OCR to the PDF and test the resulting copy there.
At a glance
Workflow comparison
| Question | App OCR index | PDF text layer |
|---|---|---|
| Where the words live | In the scanner app's document index | Inside the PDF file, aligned with page images |
| Where search works | In that app's library; it may open the matching page | In compatible PDF viewers and document search tools |
| FreeScanner today | On-device OCR search can find recognized text and return to a page | Exported PDF contains scanned page images and no embedded OCR text |
| How to check | Search a distinctive phrase in the app | Export, then search the same phrase in a separate viewer |
OCR can be stored in two different places
A scan begins as a page image. OCR recognizes words in that image, but recognition alone does not decide where the words are saved. Some apps keep recognized text in a separate library index so their own search can find documents or pages. A searchable PDF stores text inside the PDF file, commonly aligned with the original page image, so other compatible viewers can search it too.
The two outputs solve related but different jobs. An app index can be useful for finding a scan without changing the exported image. A PDF text layer makes search travel with the file. Neither output verifies the recognized words; check dates, amounts, names, or other important details against the visible page.
FreeScanner keeps its OCR index separate from PDF export
FreeScanner recognizes text on the device and stores the result in its document index. The app can use that index to find a scan and open a relevant page. Its PDF export is generated from the scanned page images; the OCR index is not embedded in the file. So it is possible for an in-app search to find a receipt or form while a separate PDF viewer cannot find the same words in the exported copy.
This describes two different search surfaces, not a broken export. The PDF still preserves the page image, and the OCR index remains useful in FreeScanner. If a portable text layer is part of the requirement, choose a workflow that explicitly creates searchable PDFs and verify the exported file independently.
Test both search surfaces with the same phrase
Use a distinctive phrase that is easy to verify in the page image. First wait for OCR to finish, search the phrase in the scanner app, and open the matching result. Then export the PDF and search for the same phrase in a separate viewer that does not run its own OCR automatically. If the app finds it but the viewer does not, the app index is working while the PDF has no searchable text layer.
If the exported file must be searchable on its own, run it through an OCR workflow that explicitly adds recognized text to the PDF. Save a new copy, open that copy in a different viewer, and repeat the search. Also select or copy a short phrase if the viewer allows it. Compare the result with the page image: OCR can substitute characters, omit faint text, or misread small print.
- Search the same phrase in the app and in an independent PDF viewer.
- Use a viewer without automatic OCR when checking whether text is embedded.
- If you add an OCR layer, save a new copy and verify that copy separately.
- Check recognized names and numbers against the visible page.
Use the right workflow for the file you need
If you mainly need to retrieve a document inside a scanner library, an app's local OCR index may be enough. If colleagues, a filing system, or another PDF viewer must search the file, require an embedded text layer and check that it survives export. If you need expense totals or categories, look for a separate extraction workflow; OCR text search does not produce verified bookkeeping data.
For receipt capture, FreeScanner's existing guide covers preparing the paper, preserving important details, naming files, and exporting a clean PDF. This page focuses only on why OCR search inside an app can differ from text search in the PDF file.
Primary sources
What this page checked
Competitor and market facts are linked to official sources. Product availability, pricing, and feature boundaries can change; verify a current detail before making a purchasing, migration, legal, or compliance decision.
- Adobe Acrobat Help: Recognize text in scanned PDFs Adobe explains that OCR adds a searchable text layer and recommends reviewing recognition accuracy.
- FreeScanner: How to Search Scanned Documents with OCR Official guide describes recognized-text search as on-device document retrieval.
- FreeScanner: How to Scan Receipts to PDF Official capture, receipt review, naming, and PDF export guide; it does not cover PDF text-layer search.