Make a folder of scanned documents searchable

Most households and small offices have a folder like this: property papers, old agreements, receipts, certificates, notices, all scanned at some point and saved as PDFs or photos with names like scan0042.pdf. Finding the one that mentions a particular survey number or company means opening them one by one.

You can fix that without special software. Convert each document to text once, keep the text next to the original, and let your computer's built-in search do the rest.

The idea

For every scan0042.pdf, you keep a scan0042.md beside it. The .md file is plain text: the words and numbers from the document, readable by any editor. Search tools can look inside text files, so searching your folder for "Survey No. 113" finds the .md file, and the original is right next to it.

The originals stay exactly as they were. Nothing is uploaded anywhere.

Step 1: Give the files sensible names

Before converting, rename the documents you can identify, using a pattern that sorts well:

2019-03-14 Sale deed - flat 4B.pdf
2023-07-01 Electricity bill - July.pdf
2024-11-20 Property tax receipt.pdf

Starting with the date (year-month-day) keeps the folder in date order. Ten minutes spent naming the important ones pays off for years.

Step 2: Convert in batches

Drop several documents onto the converter at once. Choose the OCR language first, and add English if your documents mix languages, as most Indian official papers do.

On a free account you can convert several files at a time and download them together as a ZIP. Each converted file is named after its original, so 2019-03-14 Sale deed - flat 4B.pdf becomes 2019-03-14 Sale deed - flat 4B.md. Unzip them into the same folder as the originals.

Scanned pages use your daily OCR allowance, so a large archive may take a few sessions on a free account, or one sitting on Pro.

Step 3: Let your computer search inside them

On Windows, File Explorer's search looks at file names by default and only looks inside file types it has been told to index. To add Markdown files:

1. Open Indexing Options from the Start menu.
2. Click Advanced, then the File Types tab.
3. Find md in the list (or type md in the box at the bottom and click Add), select it, and choose Index Properties and File Contents.
4. Make sure the folder with your documents is included in the indexed locations (Indexing Options → Modify).

Once indexing finishes, searching the folder in File Explorer finds words inside the .md files.

On a Mac, Spotlight and Finder search already index the contents of plain-text files in most locations. Search the folder from Finder's search box, and choose the folder rather than "This Mac" to narrow results.

In any editor, tools such as Visual Studio Code or Notepad++ can search the contents of every file in a folder at once, which is often the fastest option of all.

Step 4: Add a few words of your own

OCR captures what is printed, not what the document means to you. At the top of a .md file, add a line or two of your own:

Notes: Sale deed for flat 4B, bought from R. Mehta. Registration fee receipt is in 2019-03-20.

Those words are searchable too. It's the easiest way to connect related papers and to find a document by what it is rather than what it says.

Keeping it up to date

When a new document arrives, scan it, name it, convert it and drop the .md beside it. If you use cloud storage such as Dropbox, Google Drive or OneDrive for the folder, the text files sync with the originals and can often be searched from your phone too.

Two cautions

• The text is as sensitive as the documents. A searchable archive of property papers and IDs is convenient for you and for anyone else who gets access. Keep it in a protected folder and include it in your backups.
• OCR text is for finding, not for quoting. Use the .md files to locate the right document, then rely on the original for exact names, numbers and dates. See how to proofread OCR text if you do need the text itself to be exact.