Originally published in BNA's Patent, Trademark & Copyright Journal, November 2011.
Unlike trademark infringement cases where surveys have been commonly used for years, surveys in the damages phase of patent infringement cases are rarely seen. The playing field, however, has now changed.
Recent decisions by the U.S. Court of Appeals for the Federal Circuit reemphasized the principle that damages evidence in patent infringement cases must be based on a ''firm scientific or technical grounding'' and must relate to the invention's ''footprint in the marketplace.'' Lucent Technologies Inc. v. Gateway Inc. 580 F.3d. 1301, 92 USPQ2d 1555 (Fed. Cir. 2009) (78 PTCJ 583, 9/18/09); ResQNet.com Inc. v. Lansa Inc. 594 F.3d 860, 869, 93 USPQ2d 1553 (Fed. Cir. 2010) (79 PTCJ 422, 2/12/10); and Uniloc USA Inc. v. Microsoft Corp., 632 F.3d 1292, 1315, 98 USPQ2d 1203(Fed. Cir. 2011)(81 PTCJ 275, 1/7/11). These three rulings brought on a demand for the real-world data that can readily be found through survey evidence.
With this new demand, however, come challenges. How does one make sure the survey most accurately tests the true value of the technology at issue in the patent infringement case? How does one make sure the survey does not run afoul of the damages model in the patent infringement case? These challenges and others can be met by keeping in mind several key points when enlisting an expert to conduct a survey in a patent case. Tie the Survey to the Patent-in-Suit. One of the more important points to remember when using a survey in a patent infringement case is to tie the survey to the patent-in-suit. If the survey is not tied to the patent-insuit, then the survey evidence (and corresponding testimony) may be open to a successful Daubert challenge that alleges the evidence and testimony confuses the jury and compensates for infringement that ''punishes beyond the reach of the statute.'' ResQNet.com, 594 F.3d at 869.
To tie a survey to the patent-in-suit, the survey needs to focus on or separately account for the value attributable to the claimed technology. This focus can be achieved by considering the differences between the claimed invention and the prior art or non-infringing alternatives. Any advantages of the invention mentioned in the specification should also be carefully considered. Lastly, admissions on the record about the scope of the invention may also help focus the survey design on the claimed technology.
In a recent decision by the U.S. District Court for the Eastern District of Texas, none of these considerations were incorporated into the two surveys conducted by the plaintiff. Fractus S.A. v. Samsung Electronics Co., No. 6:09-cv-00203-LED-JDL, Doc. No. 896 (E.D. Tex. April 29, 2011).
In Fractus, the claimed technology was directed to a particular type of internal antenna in cell phones that was alleged in the specification to have certain advantages, specifically multiband capabilities and reduced size. Id. at p. 2. Indeed, using internal antennas in cell phones was already known in the prior art and the plaintiff admitted it did not invent the concept of an internal antenna. Id.
Rather than direct survey questions to the particular type of internal antenna claimed in the patents-in-suit or testing consumer demand for the advantages of the claimed technology, the surveys conducted by the plaintiff in the Fractus case attempted to broadly value customer preference of cell phones with internal antennas versus external antennas. Id. As a result, the court found the surveys were ''not tied to the alleged advantageous technical characteristics of the patents-in-suit'' and did ''not measure the value of Plaintiff's technology.'' Id.
Accordingly, both surveys were excluded as they confused the issues for the jury and punished beyond the reach of the law. Id. at pp. 2-3.
Decide Which Features to Test in a Multi-Feature Device. When an asserted patent is directed only to a single feature of a multi-feature device, surveys in patent infringement cases are increasingly being used to show the relative contribution of the accused feature to the product as a whole for purposes of capturing damages based on the entire value of the product. This type of survey centers around showing the customer preference or demand for the accused feature of a device (e.g., internal antenna for a cell phone) as compared to other features in the device (e.g., keyboard of a cell phone).
However, picking the most relevant, non-accused features to test against the accused feature can be cumbersome, especially when literally hundreds of features make up a device. To narrow the number of features to choose from, consider using market reports and market analyses to select which features to test against the accused feature.
Most of the features mentioned in these documents play a role in a consumer's purchase decision and/or are generally valued by consumers. Also consider talking to or deposing employees of the parties and nonparties, especially employees in the marketing and product design departments that are familiar with the accused product, its market, consumer preferences, and bases for certain design choices.
Moreover, it is important to consider testing all the accused features when multiple features of the device are accused. In another recent Eastern District of Texas case, Mirror Worlds LLC v. Apple Inc., the surveys failed to show sufficient evidence that certain accused features drove consumer demand because those features were excluded from the survey. 2011 WL 1304488, at *18 (E.D. Tex. April 4, 2011) (81 PTCJ 787, 4/15/11). The survey tested one accused feature that generally related to one aspect of customer demand for the accused software, but did not capture any demand of the accused features related to the accused hardware. Id. The hardware had many more features ''with varying attributes that are unaccounted for.'' Id.
Accordingly, the court rejected wholesale the plaintiff's damages theory based upon the entire market value rule, and because the plaintiff did not sufficiently support its alternate damages model, the court vacated the jury's damages award.
Consider the Relevant Time Periods. When designing a survey for a patent infringement matter, consider the relevant time periods (e.g., the hypothetical negotiation date, the date of any relevant market shift, the time the lawsuit was filed) in the case. For example, gearing the survey toward preferences existing at the hypothetical negotiation date may produce results that further support a party's claim of what a willing licensee sitting at the negotiation table would have paid at the time of first infringement.
As another example, keeping in mind relevant shifts in the market that occur over a product's life span may also be important. If a market shift shows that at a certain point in time, customers preferred the technology by 90 percent, a survey that analyzes this point in time may or may not be the best evidence to use. Other sources of relevant information including market reports or in-house surveys dated at the time of the shift and prior to the litigation, may be better, more accurate evidence.
Finally, an analysis of the time period around the filing of the lawsuit may also be taken into account. This time frame is fairly recent and, if designed properly, the survey may allow survey participants to more accurately recall the reasons underlying their purchasing decisions.
Test the Correct Universe. If the end user of the accused products is a retailer rather than an end-user consumer, then the survey may be better designed if it tests retailers rather than end-user consumers.
In addition, if the accused technology is not discernible to an end-user consumer, the test universe may be knowledgeable retailers or possibly design engineers. Prepare Testifying Experts. Testifying experts (both damages and survey experts) should be ready to adequately defend the survey. If a damages expert is relying on the survey in his or her damages model, he or she should understand the survey, as well as its faults and its strengths, and in the end should be able to defend the survey.
Because surveys offer real-world data, the demand for surveys in patent cases will likely only grow in today's climate where courts are requiring more thorough and careful evidence supporting a damages model. Challenges to these surveys will also be rampant.
Keeping abreast of future opinions relating to surveys in patent cases and following the above-mentioned key points, will increase the chances that a survey will withstand these challenges, will avoid being excluded, and will remain intact to support a damages model.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.