How to Find a File by What's Inside It on Windows (Without Knowing the Filename)

You wrote a document. You remember exactly what it said. You can't remember what you named it. If that's why you're here, you already know the frustrating part: opening files one by one, hoping to recognize the right one, isn't a real solution. There are four real ways to find a file on Windows by its content. Here's how each works, and which one to actually use.

Method 1: File Explorer with content: prefix

Windows File Explorer has a built-in content search syntax most users have never seen:

content:"quarterly revenue"

Type that into the search box of any folder window. File Explorer will look for files containing the phrase "quarterly revenue".

Pros: Built-in. No installation. Works without any setup if you only need it once.

Cons: Slow — often takes minutes on large folders. Unreliable: Office documents are inconsistently supported. Most people try this once, give up, and never find the file.

Use it for: A one-time search in a small folder of plain text files.

Method 2: Enable Windows Indexing for file contents

You can configure Windows Search to index file contents (not just filenames) so subsequent searches are faster:

1. Open Indexing Options (search "indexing" in the Start menu).
2. Click Advanced → File Types tab.
3. Find each format you care about (.docx, .xlsx, .pdf...) and select Index Properties and File Contents.
4. Click OK. Windows will reindex — this can take hours.

Pros: Once configured, searches are noticeably faster. Integrates with File Explorer.

Cons: Initial reindex consumes CPU and disk. The index breaks regularly. Office format support is still inconsistent. Most people forget this setting exists, and the rare ones who enable it find the results disappointing.

Use it for: Office workers who only deal with .docx and .pdf and have time to maintain the index.

Method 3: PowerShell Select-String (for plain text)

If you're comfortable with the command line, PowerShell has a built-in grep equivalent:

Get-ChildItem -Path C:\notes -Recurse -Include *.txt,*.md |
  Select-String -Pattern "quarterly revenue"

This recursively searches every .txt and .md file under C:\notes for the phrase.

Pros: Fast for plain text. Built into Windows. Scriptable.

Cons: Only works for plain text formats. Cannot read Word, Excel, PDF — those are binary or compressed XML and Select-String sees gibberish. No GUI.

Use it for: Searching code, markdown notes, log files, configuration files.

Method 4: A dedicated content search tool

The most reliable option is a tool built specifically to read inside documents. Three free options:

• LocalSynapse — Free, open-source, AI-powered. Searches inside 13+ formats including .docx, .xlsx, .pptx, .pdf, .hwp, .eml, .msg. 100% offline. Indexes once, searches in ~0.3 seconds. Also runs on macOS.
• DocFetcher — Open-source, Java-based. Basic keyword search across Word, Excel, PDF. Older project, but solid for plain keyword matching.
• AnyTXT Searcher — Freeware, Windows only. Fast indexing, supports most office formats. UI is dated but the search itself works.

Pros (all three): Actually search inside Word, Excel, PowerPoint, PDF reliably. Much faster than Windows native search after the initial index.

Cons: Initial install + first index takes time. Adds 50-600 MB of disk usage for the indexer. Slight learning curve for the search syntax.

Use it for: Anyone who searches by content more than once a week.

Which one should you use?

If you'll do this once and never again: Method 1 (File Explorer content: prefix). It's slow but it's free of any setup.

If you'll do this regularly (weekly or more): Method 4 — install a dedicated content search tool. The 5-minute setup pays back the first time you find a file in 0.3 seconds instead of 10 minutes of folder browsing.

If you only deal with plain text and you're a PowerShell user: Method 3.

What about Method 2 (Windows Indexing)? Skip it unless you've already tried it and it works for you. The maintenance cost rarely pays off.

Why this is harder than it should be

Modern Office documents (.docx, .xlsx, .pptx) are not plain text — they're ZIP archives containing XML files. To search inside one, a tool has to unzip the file, parse the XML, extract the text, and only then run the search. Windows Search and File Explorer technically know how to do this, but the implementation has been quietly broken for years and nobody at Microsoft seems to be fixing it.

Dedicated tools like LocalSynapse build their own parsers for each format and maintain them. That's why they work where Windows native search doesn't.

The shortest path

If you're tired of opening files one by one and you want this problem to actually go away:

1. Download LocalSynapse from localsynapse.com (free, no account needed).
2. Let it scan your Documents folder once (takes a few minutes).
3. Search by what you remember — phrase, topic, anything inside the file.

The next time you can't remember a filename, you'll find the file in under a second instead of giving up.