# OCR for Clio: Making Scans Searchable Before Upload

- Author: Schema Law (https://www.schemalaw.com/)
- Published: 2026-10-10T01:33:41Z
- Updated: 2026-10-11T19:22:20Z
- Category: OCR in everyday software
- Canonical: https://www.schemalaw.com/blog/ocr-for-clio

> OCR for Clio is a step you run before upload. What Clio's help centre says about document search, and how to make a scan findable in the matter.

OCR for Clio is best done before upload. The Clio help pages opened for this post say global search reads the full-text content of documents, and none of them describes a text recognition step for scans. So run OCR on the scanned PDF first, then upload the searchable copy to the matter in Clio.

Two things to know before going further. Schema Law has no integration with Clio and no connection to any practice-management system. It is one of many tools that can turn a scan into a searchable PDF, which you then upload to Clio yourself. And everything below about Clio comes from Clio's own help centre articles, read on October 9, 2026 and linked where they are used. Help articles get revised, so treat those links as the thing to re-read, not this post.

## Does Clio search the words inside a document?

Yes, through global search, with one condition. Clio's article on [searching and downloading documents in Clio Manage](https://help.clio.com/hc/en-us/articles/14983859054619-Search-and-Download-Documents-in-Clio-Manage) says global search looks at document names and "the full-text content of the documents". The same article says that if a firm opts out of AI features, document content is left out of results.

That condition is easy to miss, so here it is in Clio's words:

> If your firm opts out of AI features in Clio Manage, global search matches documents by name and metadata. Document content is not included in search results.

Clio's page on [AI features in Clio Manage](https://help.clio.com/hc/en-us/articles/41990965598491-AI-Features-in-Clio-Manage) repeats the point and says where the switch lives: a firm administrator turns Account Level AI Access on or off under Settings, then Security & Compliance, then AI Features. If your firm opted out for confidentiality reasons, a text layer will not make a document findable by its contents inside Clio. It still makes the file searchable once somebody opens it.

There are also two search boxes, and they do different jobs. The search field on the Documents tab is described in that search article as a place to "start typing part of the document name". Content matches come from the global search bar in the header, which Clio's [global search article](https://help.clio.com/hc/en-us/articles/9290347515291-Global-Search) says opens with Ctrl+K on Windows or Command+K on a Mac. A paralegal who types a claim number into the Documents tab and gets nothing has not tested content search at all.

### What the pages do not say

None of the Clio articles opened for this post says what happens to a PDF that holds only page images. They do not list which file types have their contents searched. They do not mention OCR: a search of Clio's help centre for that term returned no articles on the day this post was written. A second Clio support site asked for a sign-in before showing its articles, so nothing from it is used here.

Silence is not a "no". It means you cannot learn the answer from the public documentation, and the test at the end of this post is how you find out for your own account.

## What Clio's help centre says about uploads, scanning and the desktop app

Clio documents a 5 GB limit per uploaded file, a preview panel that shows each document's format and size, a desktop app with a file drive, and two ways to scan straight into a matter. None of those pages says that a scan arrives with text. The table gives each point and the article it comes from.

| Topic | What Clio's article says | Clio article |
| --- | --- | --- |
| Upload size | "The maximum upload size for a document is 5 GB." The upload form asks for a received date and lets you change the file name and pick a category. | [Create and Upload Document Files and Folders](https://help.clio.com/hc/en-us/articles/9290308200091-Create-and-Upload-Document-Files-and-Folders) |
| Plan | The Documents page "is available to select plans". | [Get Started with Documents](https://help.clio.com/hc/en-us/articles/55085715266587-Get-Started-with-Documents) |
| Preview | Clicking a file name opens a preview panel with Details, Versions, Comments and History tabs. Details includes "Format and size". | [Preview and Review Documents](https://help.clio.com/hc/en-us/articles/55223630859419-Preview-and-Review-Documents) |
| Versions | You can upload a new version of a file from its version history. | [Combine or Replace Documents](https://help.clio.com/hc/en-us/articles/43337504015387-Combine-or-Replace-Documents) |
| Clio Drive | Files in Clio Manage appear in File Explorer or Finder through the Clio desktop app. Individual files are limited to 5 GB. | [Clio Drive File Management](https://help.clio.com/hc/en-us/articles/9031909985435-Clio-Drive-File-Management) |
| Search in Clio Drive | Uses Finder or File Explorer search, which "will search the document headings of your synced Clio Drive folders". | [Clio Drive Search and File Share](https://help.clio.com/hc/en-us/articles/9032107704475-Clio-Drive-Search-and-File-Share) |
| Phone scanning | The mobile app's scanner photographs pages, crops backgrounds and saves several pages as one PDF. | [Mobile App: Documents](https://help.clio.com/hc/en-us/articles/9289697637403-Mobile-App-Documents) |
| Desktop scanning | Listed ScanSnap models can scan to a matter through Clio Launcher. | [ScanSnap and Clio](https://help.clio.com/hc/en-us/articles/9870321516571-ScanSnap-and-Clio) |

Read the two scanning rows with care. The mobile article describes taking images and combining them into a PDF. The ScanSnap article describes sending a scan to a matter. Neither says the resulting PDF has a text layer, and neither says it does not. If your office scans this way, open one of those files and try to select a word before you assume anything.

The Clio Drive row matters for a different reason. Staff who work from File Explorer all day may search there and conclude a document is missing. By Clio's description that search covers document headings, so a phrase from page nine belongs in the global search bar in the browser.

One thing this post could not check is Clio's app directory. The pages on clio.com refused automated requests on two attempts, so no third-party scanning or OCR app is described here. Look at the directory yourself, and read what each listing says it does with your files.

## Why can a scan sit in a Clio matter and never come up in search?

Because a search over document text can only match text that exists. A scanned PDF with no text layer is a set of page pictures. If nothing in the chain reads those pictures, the only words attached to the file are its name and whatever details someone typed at upload.

The post on why you cannot copy text from a PDF explains [what an image-only PDF is](https://www.schemalaw.com/blog/why-cant-i-copy-text-from-a-pdf#what-is-an-image-only-pdf) and how to spot one in a few seconds. For a Clio matter, the practical effect shows up months later. Somebody searches for an adjuster's surname or a policy number, the letter that contains it does not appear, and the firm concludes it never received the letter.

Four causes produce the same empty result, and they need different fixes:

- **The file has no text.** The scan was uploaded as it came off the copier.
- **Content search is off for the firm.** The firm opted out of AI features, so Clio matches names and metadata only.
- **The wrong box was used.** The Documents tab field and the Clio Drive search are described as name searches.
- **The text is there but wrong.** A scanner added a poor text layer, so the word in the file is not the word on the page.

Only the first and last are about OCR. Rule out the middle two before you convert anything.

## Converting a scan first, then uploading it to the Clio matter

Convert on your own computer, check the result, name the file, and only then upload. The copy that lands in the matter is the one people will open and search for years, so it should be the searchable one from the first day.

1. **Collect the scan.** Take it from the copier, the email attachment or the phone. Keep one document per file. A day's post saved as a single PDF is hard to name and hard to find.
2. **Run OCR with a tool your firm has approved.** Desktop PDF editors, scanner software and browser tools can all add a text layer. The guide to [how to make a scanned PDF searchable](https://www.schemalaw.com/blog/how-to-make-a-scanned-pdf-searchable) explains what that layer is.
3. **Open the output before it goes anywhere.** Search it for a word on the first page and a word on the last. Read the names, dates and amounts on page one against the picture.
4. **Name it.** Put the date first, then the sender or party, then what the document is, for example `2026-09-30_Northfield-Mutual_Denial-Letter.pdf`. Clio's upload form lets you change the name at that point, but a file that is already named cannot be uploaded under the wrong one.
5. **Upload to the matter.** In Clio Manage, go to the matter's Documents subtab, choose New, then Upload files. Fill in the received date and the category. Clio's preview panel later shows both in the document's details.
6. **Deal with the original scan as firm policy says.** Some firms keep the scan exactly as received next to the searchable copy, with a suffix such as `_as-received`. Others keep only one file. This post does not tell you which is right for your records. Decide once and write it down.

Naming patterns for court papers, and the order of scanning, converting and filing in general, belong to the post on [case management OCR and getting scans into a matter file](https://www.schemalaw.com/blog/case-management-ocr). Use its pattern if you file pleadings.

If your policy is to replace a scan already in Clio, the article on [combining or replacing documents](https://help.clio.com/hc/en-us/articles/43337504015387-Combine-or-Replace-Documents) describes adding a new version from the file's version history. The pages opened for this post do not say which version's text global search reads, so test that before relying on it. Clio's article on [file management in Clio Drive](https://help.clio.com/hc/en-us/articles/9031909985435-Clio-Drive-File-Management) also says a deleted document can be recovered in Clio Manage for up to 30 days, which is worth knowing before anyone clears out originals.

### Using Schema Law for the conversion step

Schema Law is a browser tool for step 2. You sign in with a Google account, upload a PDF, PNG or JPEG of up to 20 MB, and export the result in any of eight formats on every plan. Each account belongs to one person. There is no link to Clio: you download the export and upload it to the matter by hand.

Clio accepts files up to 5 GB, so a scan can fit in Clio and still be too large for a 20 MB upload elsewhere. The post on [converting a large scanned PDF](https://www.schemalaw.com/blog/convert-large-scanned-pdf-to-word-online) covers splitting or shrinking one. Page allowances per document are on the [pricing page](https://www.schemalaw.com/pricing).

## Which export belongs in a Clio matter: PDF, Word or Excel?

The searchable PDF is the record copy, because it still shows the page as it was received. Word is for drafting from the text. Excel is for tables you need to sort or total. Upload the PDF to the matter every time, and add the others only when someone will work from them.

| Export | Use it for | In the matter |
| --- | --- | --- |
| Searchable PDF | The record of what arrived: stamps, signatures and layout stay visible | Always, as the main copy |
| Word | Drafting a response, reusing clauses, quoting long passages | When needed, named `_working` so nobody takes it for the original |
| Excel | Statements, schedules, invoices and other tables | When someone will check or total the figures |
| Plain text | Pasting a paragraph into an email or a note | Rarely worth filing |

One detail about Schema Law's searchable PDF affects how you use it. The export keeps your original pages and adds hidden text to each one, in reading order from the top. The text is not positioned under each word. A search inside the PDF therefore takes you to the right page, but the highlight may not sit on the word itself. For finding a document and the page within it, that is enough. If you need highlights that land exactly on each word, test the file in your PDF viewer before you standardize on it.

The Word and Excel copies are readings of a picture, not the document. Check them against the PDF before anyone relies on a figure or a quoted sentence. The post on [why OCR output is full of errors](https://www.schemalaw.com/blog/why-is-ocr-output-full-of-errors) explains what to look for, and the guide to [scanned PDF to Excel](https://www.schemalaw.com/blog/scanned-pdf-to-excel) covers tables. The full list of what Schema Law exports is under [formats on the home page](https://www.schemalaw.com/#formats).

## A short checklist for a scan on its way into Clio

Six checks, in the order you meet them. Print the list and keep it by the scanner until the routine is habit.

- One document per file, pages in order, nothing upside down.
- A word on the last page can be selected and found with Ctrl+F in the PDF.
- Names, dates and amounts on page one match the page image.
- The file name carries the date, the sender and the document type.
- Received date and category are filled in on Clio's upload form.
- The original scan is kept or discarded as firm policy says, and labelled if kept.

A confidential scan leaves your computer when you send it to any online OCR service. Check that service's terms first. Schema Law's commitments are listed under [security on the home page](https://www.schemalaw.com/#security).

## How do you confirm Clio's global search finds the uploaded scan?

Upload two copies of one page to a test matter, one image-only and one searchable, and look for a word that appears on the page but in neither file name. Use the global search bar, not the Documents tab field. Which copy comes back tells you what your Clio account does with text.

1. Choose a typed page with an unusual word on it, such as a surname or a part number.
2. Make the searchable copy. Confirm you can select the word in it, and that you cannot select anything in the plain scan.
3. Name them `clio-test-scan.pdf` and `clio-test-text.pdf`. Neither name contains the word.
4. Upload both to a matter created for testing.
5. Open global search with Ctrl+K or Command+K, type the word, and open the full results. Look under documents.
6. If nothing comes back, try again in an hour and the next morning. The articles opened for this post give no indexing time.
7. Search for `clio-test` to confirm both files are there and visible to you.

Read the outcome this way:

| Result | What it suggests | What to do |
| --- | --- | --- |
| Only the searchable copy is found | Clio searches text already in the file | OCR every scan before upload |
| Both are found | Something in your account reads scans | Still check how well, using a poor fax page |
| Neither is found, both appear by name | Content search may be off | Ask your administrator about the AI features setting and your plan |

Write the result into your office procedures, with the date you ran it, and repeat the test after a plan change or a change to the AI setting. Firms on other systems face the same question with different answers: see the posts on [NetDocuments OCR](https://www.schemalaw.com/blog/netdocuments-ocr) and [SharePoint OCR](https://www.schemalaw.com/blog/sharepoint-ocr) for what those vendors document.

## Try it on a real document

Upload a scanned PDF and export it as Word, a searchable PDF or plain text on the free plan. Your documents are never used to train an AI model. [Start free](https://www.schemalaw.com/api/auth/google)