OCR cost shifting rejected

March 2nd, 2009 | By Steve Puiszis

Proctor & Gamble Co. v. S.C. Johnson & Son, Inc., 2009 WL 440543 (E.D. Tex., Feb. 19, 2009)

Given today’s economic climate, shifting the costs of electronic discovery is a strategy that should be considered whenever possible. In Proctor & Gamble, the district court rejected the defendant’s attempt to shift the costs of applying optical character recognition (“OCR”) to documents it had agreed to produce in a tagged image file format (“TIFF”). The application of OCR to the TIFF images would make the imaged documents electronically searchable.

While the district court recognized that the application of OCR to the imaged documents was “not absolutely necessary to [the] litigation,” it rejected the defendant’s attempt to shift the cost of OCR processing because the defendant failed to provide any support for its cost estimate, and no valid legal argument to support its claim that the cost burden should be shifted. The defendant advised the court that it did not intend to use the OCR process, and argued that it should not be forced to bear the added OCR expense solely for the plaintiff’s convenience. That argument failed to carry the day in Proctor & Gamble.

Any cost-shifting strategy requires careful advanced planning, and a thorough knowledge of the client’s information systems. While the district court’s ruling in Proctor & Gamble appears relatively straightforward, there are a number of issues that are relevant to any cost-shifting strategy that must be carefully analyzed before a party should even begin to evaluate cost-shifting factors outlined in Zubulake, which was the focus of the district court’s opinion in Proctor & Gamble.

Evaluate the accessibility of the data.

A number of district courts have concluded that cost-shifting should not be considered absent a showing of inaccessibility. See, e.g., Peskoff v. Farber, 240 F.R.D. 26, 31 (D.D.C. 2007) (“cost-shifting does not even become a possibility unless there is first a showing of inaccessibility”); Quinby v. WestLB AG, 245 F.R.D. 94, 104 (S.D.N.Y. 2006)(“Cost-shifting is appropriate only where electronic discovery imposes and undue burden or expense”); Zubulake v.UBS Warburg, LLC, 217 F.R.D. 309, 323 (S.D.N.Y. 2003)(“A court should consider cost-shifting only when the electronic data is relatively inaccessible, such as in backup tapes). The burden of proof is placed on the party asserting inaccessibility.

While information stored on backup tapes, on legacy systems, and fragmented data on hard drives may generally be considered inaccessible sources, there is no clear dividing line between accessible and inaccessible data sources. Under Rule 26(b)(2)(B), the inaccessibility of a data source is determined by how burdensome and costly the preservation and production of that data is, which will vary from client to client, and case to case. The information about the cost and burden of producing data from various sources can only come from the client. Thus, the client’s IT staff will play a key role in any attempt to establish that certain forms or sources of data are inaccessible, which is one of the first issues to address in any cost-shifting analysis.

Evaluate the proportionality of the discovery being sought.

The fact that the data to be produced is from an accessible source should not be the end of your cost-shifting analysis. “The ready accessibility of information, whether electronically stored or not, does not in itself require the production of that information.” Peskoff v. Farber, 244 F.3d 54, 59 (D.D.C. 2007). Rule 26(b)(2)(C)’s proportionality factors are a relevant consideration for any discovery being sought, even accessible electronic data. Under Rule 26(b)(2)(C)’s proportionality principles, one factor that a court can consider in limiting discovery that would otherwise be allowed is whether:

the burden or expense of the proposed discovery outweighs its likely benefit, considering the needs of the case, the amount in controversy, the parties’ resources, the importance of the issues at stake in the action, and the importance of the discovery in resolving the issues.

When a party can demonstrate that the burden of producing the requested data is out of proportion to the amount in controversy or is not justified, on balance, by its potential relevancy to the issues at stake in the litigation, then a court may limit the requested discovery or condition its production upon a payment of some or all of its cost by the requesting party. Remember however, that courts require evidence of the burden involved, and as the Proctor & Gamble decision teaches, courts will not shift the cost of ediscovery based on a mere assertion that the requested discovery is unduly burdensome or disproportionate.

Evaluate the discovery your opponent requested.

It seems fundamental, but don’t overlook the fact that courts generally won’t compel discovery that has not been sought. See D’Onofrio v. SFX Sports Group, Inc., 247 F.R.D. 43, 48 (D.D.C. 2008). Carefully review the discovery your opponent has requested. If the requesting party has not sought production of electronic data in its native state with accompanying metadata fields, or in the case of TIFF or PDF images, has not requested OCR be added to the images, there is no need to necessarily produce the data with those features.

Evaluate whether the data is created in a unique or unusual format.

Data that is created and stored in unique or unusual formats can complicate how that data is produced, and your cost-shifting analysis. When electronic data is produced in its native state, to simply view the data, the party receiving it usually needs the same software application used to create the data. That is why the federal ediscovery rules provide that data may be produced in the format in which it is ordinarily maintained or in a reasonably useable form. Unless the party possessing data in a unique or unusual format can demonstrate that the cost of translating the requested data into a reasonably useable form is out of proportion with the amount at issue in the case, or that the relevancy of the data is marginal, that party will likely be forced to incur the cost of translating the data into another reasonably useable form.

Evaluate whether the data is electronically searchable in its “native state.”

The 2006 Advisory Committee Note to Rule 34(b) specifies: “If the responding party ordinarily maintains the information it is producing in a way that makes it searchable by electronic means, the information should not be produced in a form that removes or significantly degrades this feature.” Generally speaking, the electronic data generated by many of today’s applications is electronically searchable at some level within the application. However, that point should always be confirmed. How easily native data can be searched depends on the particular application, and the volume of information involved.

However, producing data in its native state has several perceived disadvantages. Application and system metadata, as well as any embedded comments or track changes accompany documents or data when produced in their native state. Moreover, electronic documents cannot be Bates stamped, and privileged or confidential information cannot be redacted from documents produced in their native state. Documents can also be altered after production has occurred in their native state.

These factors have led many parties to produce electronic data in an imaged (PDF or TIFF) format. However, without the addition of OCR, the imaged documents are not electronically searchable. The application of OCR to documents once they have been imaged permits them to be searched on a common platform. When different forms of data are produced natively, they generally have to be searched within their respective applications. Thus, the inclusion of OCR to imaged documents adds to the convenience of the search process as the defendant claimed in Proctor & Gamble. Nonetheless, if a particular data set was electronically searchable in its native state, compliance with the Advisory Committee Note to Rule 34(b) may require that OCR be added when that data is produced in an imaged format.

Evaluate Zubulake’s cost-shifting factors.

Once the various issues noted above have been carefully considered, the cost-shifting factors outlined in Zubulake should be analyzed. Those factors, in order of their relative importance are:

• The extent to which the request is tailored to discover relevant information;
• The availability of such information from other sources;
• The total cost of production, compared to the amount in controversy;
• The total cost of production, compared to the resources available to each party;
• The relative ability of each party to control costs, and their incentive to do so;
• The importance of the issues at stake in the litigation; and
• The relative benefits to the parties of obtaining the information.

The defendant in Proctor & Gamble merely argued that plaintiff’s discovery requests were overbroad without any further explanation. That argument fell woefully short of meeting any of Zubulake’s cost-shifting factors in the court’s view.

Other Relevant Considerations.

Courts are unlikely to entertain any cost-shifting arguments where the producing party failed to adequately preserve the data from an accessible source. One district court, in responding to such an argument explained: “It reminds me of Leo Kosten’s definition of chutzpah: ‘that quality enshrined in a man who, after having killed his mother and father, throws himself on the mercy of the court because he is an orphan.” Disability Rights Counsel of Greater Washington v. Washington Metro. Transit Authority, 242 F.R.D. 139, 148 (D.D.C. 2007).

Finally, while this point may also seem fundamental, any cost-shifting argument should be brought before production occurs. Post production cost-shifting arguments are untimely, and will be rejected. See, e.g., Cason-Medina v. Detroit Medical Center, 2008 WL 2714239 (E.D. Mich. July 7, 2008).

As can be seen from this brief discussion of ediscovery cost-shifting, the factors to consider are numerous and varied. However, given the expense of electronic discovery, the analysis required is well worth the effort.

Chess photo courtesy Flickr user dbking under this Creative Commons license.

Leave a Comment »

---