eDiscovery Daily Blog

How to Avoid Getting “Burned” by Redactions: eDiscovery Best Practices

Having addressed this issue with a client recently, it seems appropriate to revisit this topic…

On the surface, it may seem easy enough to redact a document during eDiscovery review to obscure confidential or privileged information.  All you need to do is draw a black box over the affected text, right?  Not necessarily.  There’s a lot more to consider in order to ensure that you don’t inadvertently produce information that was intended to be redacted.  Here are a few things to consider to avoid getting “burned by redaction failures.

Failure to “Burn” the Redaction into the Image

If the redaction isn’t “burned” into the image so that it cannot be removed, the redacted data can still be viewed.  Especially when your images are Adobe Acrobat PDF files, the most common mistake is to redact by obscuring the text by drawing a black box over the text or images you want redacted.  A simple “cut and paste” can remove the black box, revealing the redacted text.  Acrobat provides a redaction tool (for those editing the PDF there) to properly apply a redaction – it’s best to save the file to a new name after the redaction has been applied.

If you’re using a review application to manage the review, the application should ensure a “burned in” redaction for anything exported or printed, regardless of whether it lets you look at the redacted data within the application itself.  For example, CloudNine provides a tool to enable the reviewer to draw a gray box over the text to be redacted so that text can still be viewed within the application.  However, if the file is exported or printed, that box gets “burned” in as a black box to completely obscure the redacted text.

Failing to Update Corresponding Text Files to Remove Redacted Text

Even if the image is handled properly, you can still disclose redacted text if you don’t make sure that the corresponding text file, whether extracted from the native file or generated via Optical Character Recognition (OCR), isn’t updated to remove the redacted text.  If you don’t update the corresponding text files, you’re allowing redacted text to slip through the production “back door”.  This happens more often than you might think.

Producing Un-Redacted Native Files

If you’re producing native files, you’ve hopefully discussed with opposing counsel how to handle native files that require redaction.  Typically, the approach is to convert those to an image format and redact the image.  Sometimes the parties agree to “redact” the native files themselves and produce those.  If so, as is the case with Adobe PDF files, there’s a right way and wrong way to redact native files.  Changing the text to white or the background to match the text color is not the same as redacting the text.  All you have to do is to revert back to the original formatting or simply highlight the affected area to see the redacted text.  Instead, you’ll want to agree on a procedure where the text is deleted or replaced with an equal amount of meaningless content (e.g., all “X”s) to preserve text flow and pagination (make sure track changes is off before redacting).  You may even want to agree to copy the entire content of a redacted document to a new file (to remove residual document composition information that might remain).

Failing to Redact Metadata

You may redact content on the document that you produce separately as metadata, via a load or data file.  Failing to check the produced metadata for redacted documents could enable redacted data to slip through.  So, don’t forget to check and remove any sensitive data here, as well.

Quality Control (QC) Check before Producing

Generally, when producing documents with redactions, you should have a checklist that ensures that image redactions are “burned” in, that redacted native files (if produced natively) are properly redacted, and that corresponding text files and metadata have been checked to ensure that redacted data has been removed from those as well.  Otherwise, you could be “burned” by inadvertent production of redacted materials.

P.S., to see what I “redacted” up above, highlight it with your cursor.  :o)

So, what do you think?  How do you handle redactions within your productions?  Do you have a process to QC check redactions before producing?  Please share any comments you might have or if you’d like to know more about a particular topic.

Disclaimer: The views represented herein are exclusively the views of the author, and do not necessarily represent the views held by CloudNine. eDiscovery Daily is made available by CloudNine solely for educational purposes to provide general information about general eDiscovery principles and not to provide specific legal advice applicable to any particular circumstance. eDiscovery Daily should not be used as a substitute for competent legal advice from a lawyer you have retained and who has agreed to represent you.