Best Searchable PDF Software: What to Compare Before Buying

The best searchable PDF software for your office is the one that keeps the scanned page looking exactly as it did, puts the hidden text under the right words, accepts files as long as the ones you receive, and adds the extras your work requires. Compare candidates on those points with your own scans before buying.

No product is named here, because the ranking changes with the job. A program that suits a records room can be the wrong purchase for someone who files in court every week. What follows is a set of six comparisons, each with a test you can run on a short sample, and a way to match the results to the work you do.

Match the software to the file you will cite. A deposition transcript needs page labels that match the court reporter's numbers. Pleadings you will file need a text layer the court's portal can accept. Discovery you will review needs search that lands on the right word. Store each searchable copy in document management beside the image you received, and keep the two names distinct.

What should the best searchable PDF software leave untouched in a scan?

The best searchable PDF software leaves the page picture alone. Signatures, stamps, margin notes, exhibit stickers and the grain of the scan should look the same after conversion as before. The only addition should be an invisible layer of text. Anything else the program changes is something you now have to explain.

A searchable PDF is a scan with recognized text placed behind the image. The post on how to make a scanned PDF searchable explains how that layer is built. For a buyer, the question is narrower: what did this program do to the picture while it was adding the text?

Changes to look for

Programs differ in what they do by default. Some recompress the image to make the file smaller. Some straighten pages, crop borders or convert color to gray. Some rebuild the page out of fonts so that it looks typed. Each of these can be useful and each can be unwanted, depending on whether the file is a working copy or a copy of record.

Run this comparison on five pages that include a signature, a stamp and some handwriting:

1. Open the original and the converted file side by side at 200% zoom.
2. Compare the signature and the stamp. Look for softened edges, blocky patches or lost color.
3. Compare the page edges. Check whether a margin note or a page number near the border was cropped.
4. Check the page count and page order in both files.
5. Look in the program's settings for each change you found, and note whether it can be switched off.

A program that changes the image only when you ask it to scores well. A program that changes it silently, with no setting, scores badly for legal work however good its recognition is. The post on whether OCR changes the original PDF covers how to show that a searchable copy matches its source.

Does it write over your original?

Check where the output goes. Software that saves the searchable version over the scan you opened leaves you with no original. Prefer a program that writes a new file, or make a habit of working on a copy. Either way, find this out during the trial and not on a live matter.

Comparing searchable PDF software on file size, page count and batch limits

Compare capacity on four numbers: the largest file the software accepts, the most pages it takes in one document, the pages it allows in a month, and how many files it handles in one run. Measure your own scans first, then read each candidate's published limits against those measurements.

Start with what you have. Pull the ten largest scanned PDFs your office received in the last quarter and write down each one's size in megabytes and its page count. Add up a typical month of pages. Those three figures are your requirement.

Installed software and browser software state these limits in different places. An installed program may have no page allowance at all but is bound by the memory and speed of the computer. A browser service publishes an allowance and an upload ceiling. Put both kinds on the same table so you compare like with like.

Schema Law's own numbers show how to read a plan this way. Uploads are PDF, PNG or JPEG files of up to 20 MB. The Free plan lists up to 3 pages per document, Solo lists 1,000 pages a month and up to 250 pages per document, and Team lists 10,000 pages a month and up to 1,000 pages per document. Every plan takes several files at a time. The pricing page has the full list. If your largest file from last quarter is a 600-page records set, that single fact decides which row applies.

Size of the output

A searchable PDF can come out smaller or larger than the scan, depending on what the program does to the image. Record the size before and after for each candidate. If a court portal or an email system caps attachment size, a program that doubles the file is a problem. The post on searchable PDF file size covers shrinking a file without losing its text.

Does searchable PDF software put the text where the words are?

Searchable PDF software has placed its text correctly when selecting a word highlights that word and nothing else, and when a search hit lands on the word you can see. A program can recognize every character and still fail here, which makes highlighting and copying unreliable even though search returns results.

Recognition and placement are separate qualities. Recognition is whether "Harlow" was read as "Harlow". Placement is whether the invisible "Harlow" sits under the printed one. Poor placement shows up as highlights that float above the line, selections that grab half of the next column, or a search that jumps to the right page but marks an empty patch of margin.

A selection and highlight test

Use one page of dense text, one page with two columns or a table, and one page that was scanned slightly crooked.

1. Drag across a single line. The highlight should cover that line from first word to last, at the height of the letters.
2. Double-click one word in the middle of a paragraph. Only that word should select.
3. Search for a word that appears once on the page. The marked area should sit on the printed word.
4. Select a whole paragraph and paste it into a blank document. The words should arrive in reading order, with no lines from a neighboring column mixed in.
5. Select one cell of the table. The selection should stay inside the cell.
6. Repeat steps 1 and 3 on the crooked page. Check whether the highlight follows the slant of the text or sits level across it.

Score each step pass or fail for each program. Six steps on three pages give eighteen results per candidate, which is enough to see a pattern.

Why placement matters in legal work

Three everyday tasks depend on it. Highlighting a passage for a partner needs the highlight to sit on the passage. Copying a quotation needs the words in order. And a redaction tool that finds words by searching relies on the text layer, so a misplaced layer is a reason to check each redaction against the page image. That last risk has its own owner in the post on whether to OCR before or after redaction.

This short test checks placement only. A full trial that also scores recognition, tables and correction time is set out in the post on how to test OCR software.

Bookmarks, PDF/A and page labels: features only some searchable PDF software has

Bookmarks, PDF/A output and page labels are separate features from recognition. A program can make a perfectly searchable file and offer none of them. Decide which of the three your work requires, then check each candidate's documentation for that feature by name before you assume it is there.

Questions to put to each candidate

• Bookmarks. Can the program add them at all? If the scan already had bookmarks, are they still there after conversion?
• PDF/A. Is it an output option, and which version does the program write? Does the documentation say the output can be validated?
• Page labels. Can you set them, and does the program keep labels that were already in the file?

Whether you need any of these is set by someone else: the court, the archive or the recipient. Read the rule or the specification first. The post on searchable PDFs for court filing covers where to find a court's requirements.

One program or two

You do not need a single program that does everything. One workable division is to make the text layer in one tool and add bookmarks or convert to PDF/A in a PDF editor the office already owns. If you split the work that way, test the hand-off: open the searchable file in the second program, add a bookmark, save, and confirm the text is still searchable afterwards.

Schema Law's format list describes its searchable PDF export in one line: "Your original file with a hidden text layer, so search finds the page". The list says nothing about bookmarks, PDF/A or page labels, so plan to add those in a PDF editor if your work calls for them.

What does free searchable PDF software leave out?

Free searchable PDF software leaves out one or more of four things: pages, features, support or a commercial licence. Which one depends on the kind of free. Find the limit in writing before you build a routine on a free tool, and check that the limit is one your work can live with.

Free options fall into three groups, and each asks a different question.

Free tiers of paid services. The limit is volume. Read how many pages or documents the tier allows and how long files are kept. Schema Law's Free plan, for example, lists 1 document a day and up to 3 pages per document, with searchable PDF among the export formats every plan includes. That is a fair way to see what the output looks like on a real page, and no way to process a deposition.

Features inside software you already have. A scanner's driver, a copier's panel or an existing PDF subscription may make searchable files at no extra charge. The limit is control: fewer settings, and whatever recognition the maker included. The post on making a PDF searchable without Acrobat walks through those routes.

Free programs you install. The limit is support and upkeep. You vet the installer, update it and solve your own problems. Licence terms matter too, because a program free for personal use is not free for a firm. The vetting steps are in the post on free OCR to Word software downloads.

The same kinds of free, judged for Word output, are compared in the post on the best free OCR PDF to Word converter.

A checklist for any free option

• What is the page or document limit, and over what period?
• Is the output marked in any way, such as a watermark or an added page?
• Does the licence or the terms of service permit business use?
• For a web tool, what do the terms say about keeping and using uploaded files?
• Which of the extras you need are missing: batch runs, bookmarks, PDF/A?
• If it fails on a deadline, what is the fallback?

Do not skip the confidentiality item. A free web converter with no account and no stated retention period is a poor place for a client's medical records, whatever the quality of its output. The post on whether free online OCR is safe for legal documents sets out what to read before you upload.

Matching searchable PDF software to filing, review or archive work

Match the software to the work by ranking the comparisons differently for each purpose. Filing puts page fidelity, output size and court-required extras first. Review puts capacity and text placement first. Archive work puts an untouched image and a standard file format first. Score candidates against the ranking that fits your work.

A worked example

Okafor Lindqvist LLP is a fictional four-lawyer firm. Its work splits three ways. One lawyer files appellate appendices of 300 to 500 pages. Two handle commercial disputes and receive productions of several thousand pages as hundreds of small files. The office manager is scanning twenty years of closed files.

The firm writes down three requirements instead of one:

• Appendices. Per-document capacity of at least 500 pages, output small enough for the appellate court's portal, bookmarks and page labels. The lawyer checks the court's rules for the exact figures.
• Productions. Batch upload, a monthly allowance in the thousands of pages, and text placement good enough to highlight and copy.
• Closed files. An unaltered page image, a monthly allowance that covers the backlog over a year, and PDF/A if the firm's retention policy calls for it.

The partners then test two candidates on fifteen pages from each kind of work. One candidate passes the placement test and handles the batch but has no bookmark feature. The firm decides that is acceptable, because its existing PDF editor adds bookmarks, and it confirms that the text survives the round trip. The purchase is made on the production requirement, which is the one the existing tools could not meet.

Turning the comparisons into a decision

1. Write down which of the three purposes you have, and in what proportion.
2. Take the measurements: largest file, longest document, monthly pages.
3. Strike any candidate whose published limits fall below those measurements.
4. Run the image comparison and the selection test on what remains.
5. Check the extras against the rule or specification that requires them.
6. Buy for the requirement you cannot meet with tools already in the office.