Searching is not working right as it used to do

👋 Hi ​@Abe, welcome to the Box Community. Thank you for reaching out.

There are several key reasons why Box Search may fail to locate PDFs containing a specific alphanumeric string like HVWS70:
 

1. Document indexing limits & non-searchable PDFs

• OCR / scanned image PDFs: Box standard search indexing does not perform Optical Character Recognition (OCR) on image-based or scanned documents. If your PDF contains scanned images without an underlying selectable text layer, Box cannot index the body text.
• Text indexing cap (10,000 Bytes): Box search indexes up to the first 10,000 bytes (~10,000 English characters / ~10 KB) per document. If HVWS70 appears further down past this 10 KB threshold in the PDF body, it will not be indexed.
• Indexing delay: When new files are uploaded or modified, indexing typically takes about 10 minutes (and can take longer under heavy system load). Newly updated files will not appear immediately in search results.
• Current version only: Box Search indexes only the most current version of a file. Content located in older, non-current versions cannot be queried.
   

2. Search syntax & matching limitations

• No infix / suffix matching: While Box supports prefix matching on file titles (e.g., searching Bo matches Box), it does not support prefix, infix (middle of a word), or suffix matching in document body content. If HVWS70 is embedded inside a longer alphanumeric string or word without spaces/punctuation, search will not match it.
• Double quotes for exact terms: Box treats standard multi-word or standalone terms with fuzzy matching/stemming. Searching with quotation marks (e.g., "HVWS70") forces an exact word/phrase sequence search.
• Punctuation & splitting: Punctuation characters split words in the search index into multiple resultant tokens as well as concatenated forms.
   

To help resolve, here are the recommended troubleshooting steps:

1. Use targeted search syntax:
   • Search with quotation marks: "HVWS70"
   • Target body content specifically: contents:HVWS70 or contents:"HVWS70"
   • If searching file titles: name:HVWS70
2. Apply filters:
   • In the Search Filters panel, explicitly set the Type filter to PDF and set the In folder filter to the specific folder where the documents reside.
3. Verify text layer:
   • Open one of the PDFs in Box Preview or a PDF viewer and try highlighting/selecting the text HVWS70. If you cannot select or copy the text, the document is an image/scanned PDF and will require OCR processing before its content can be indexed by Box.

📌 To learn more, please visit: Search for Files, Folders, and Content

If you need further assistance, you can submit a ticket to our Product Support. Click here to access the request form if you’re unable to sign in or not seeing the option to open a case. Thank you!