Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 1 of 20
`
`
`
`IN THE UNITED STATES DISTRICT COURT
`WESTERN DISTRICT OF TEXAS
`WACO DIVISION
`
`
`
`
`
`Civil Action No. 6:21-cv-00898-ADA-DTG
`
`
`JURY TRIAL DEMANDED
`
`
`
`v.
`
`Defendants.
`
`ALMONDNET, INC.,
`
`
`Plaintiff,
`
`
`AMAZON.COM, INC., AMAZON.COM
`SERVICES LLC, and AMAZON WEB
`SERVICES, INC.,
`
`
`
`
`
`
`
`
`
`
`
`AMAZON’S MOTION TO EXCLUDE THE UNRELIABLE TESTIMONY OF
`ALMONDNET’S DAMAGES EXPERT JIM W. BERGMAN
`
`
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 2 of 20
`
`TABLE OF CONTENTS
`
`
`
`Page
`
`I.
`
`FACTUAL BACKGROUND ..............................................................................................2
`
`A.
`
`B.
`
`C.
`
`D.
`
`The patents-in-suit claim particular business methods relating to serving
`off-site advertisements. ............................................................................................2
`
`Amazon’s sophisticated ad platform uses real-time bidding, machine
`learning, and Amazon’s extensive customer data to help advertisers and
`publishers efficiently deliver targeted advertisements. ............................................3
`
`Mr. Bergman’s damages model captures value from unclaimed features. ..............5
`
`Mr. Bergman estimates an arbitrary bargaining split. ..............................................8
`
`II.
`
`MR. BERGMAN’S UNRELIABLE TESTIMONY SHOULD BE EXCLUDED ..............9
`
`A.
`
`B.
`
`Mr. Bergman’s failure to apportion value to the claimed inventions
`renders his testimony unreliable. ...........................................................................10
`
`Mr. Bergman’s arbitrary bargaining split renders his opinions unreliable. ...........12
`
`III.
`
`CONCLUSION ..................................................................................................................14
`
`
`
`
`
`i
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 3 of 20
`
`TABLE OF AUTHORITIES
`
`
`
`Page(s)
`
`CASES
`
`Bayer HealthCare LLC v. Baxalta Inc.,
`No. 16-cv-1122-RGA, 2019 WL 330149 (D. Del. Jan. 25, 2019).....................................12, 13
`
`BMC Software, Inc. v. ServiceNow, Inc.,
`No. 2:14-cv-903-JRG, Dkt. No. 328 (E.D. Tex. Feb. 1, 2016)................................................13
`
`CSIRO v. Cisco Sys.,
`809 F.3d 1295 (Fed. Cir. 2015)................................................................................................10
`
`Daubert v. Merrell Dow Pharms., Inc.,
`509 U.S. 579 (1993) .......................................................................................................2, 13, 14
`
`Ericsson, Inc. v. D-Link Sys.,
`773 F.3d 1201 (Fed. Cir. 2014)..........................................................................................10, 11
`
`Finjan, Inc. v. Blue Coat Sys., Inc.,
`879 F.3d 1299 (Fed. Cir. 2018)................................................................................................11
`
`Kumho Tire Co. v. Carmichael,
`526 U.S. 137 (1999) ...................................................................................................................9
`
`LaserDynamics, Inc. v. Quanta Comput., Inc.,
`694 F.3d 51 (Fed. Cir. 2012)........................................................................................12, 13, 14
`
`Power Integrations, Inc. v. Fairchild Semiconductor Int’l, Inc.,
`711 F.3d 1348 (Fed. Cir. 2013)..................................................................................................9
`
`Uniloc USA, Inc. v. Microsoft Corp.,
`632 F.3d 1292 (Fed. Cir. 2011)....................................................................................12, 13, 14
`
`VirnetX, Inc. v. Cisco Sys., Inc.,
`767 F.3d 1308 (Fed. Cir. 2014)................................................................................................10
`
`OTHER AUTHORITIES
`
`Federal Rule of Evidence 702 ....................................................................................................9, 14
`
`
`
`ii
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 4 of 20
`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 4 of 20
`
`TABLE OF EXHIBITS
`
`Exhibit No.|Description
`
`2780
`
`2794
`
`3557
`
`3612
`
`3669
`
`
`
`Expert Report of Dr. Eric Koskinen Regarding Infringement, served June 16,
`Ex.A|2023
`Ex. B
`Rebuttal Expert Report of Dr. Henry Houh, served July 28, 223
`Ex. C
`Expert Report of Jim W. Bergman, served June 16, 2023
`Ex.D
`Excerpted transcript from the deposition of Jim W. Bergman, taken August 9,
`‘
`2023
`Ex. E
`Excerpted transcript from the deposition of Eric Koskinen, taken August 15,
`‘
`2023
`Ex. F
`:
`Ex.G
`,
`Ex.H
`*
`Ex I
`.
`Ex. J
`‘
`Ex.K
`‘
`Ex. L
`.
`Ex.M
`
`
`
`, produced at ALMONDNET-AMAZON-0002775-
`
`, produced at ALMONDNET-AMAZON-0002790-
`
`, produced at ALMONDNET-AMAZON-0003547-
`
`, produced at ALMONDNET-AMAZON-0003608-
`
`, produced at ALMONDNET-AMAZON-0003665-
`
`, produced at ALMONDNET-AMAZON-0046010-
`
`46015
`Plaintiffs’ Fourth Supplemental Objections and Responses to Defendants’
`First Set of Interrogatories
`(Nos. 1 and 9), served April 11, 2023
`Excerpted transcript from the deposition of Roy Shkedi, taken March 21,
`2023
`
`ill
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 5 of 20
`
`
`
`The Court should exclude the unreliable testimony of AlmondNet’s damages expert, Jim
`
`Bergman, because his analysis includes two fatal errors. First, Mr. Bergman does not perform an
`
`appropriate apportionment analysis to isolate the incremental value of the claimed inventions—
`
`i.e., the specific value Amazon purportedly derives from the asserted claims. Indeed, though Mr.
`
`Bergman’s task is ultimately to value the asserted claims, he admits he did not even consider the
`
`claims, or what they actually require, in performing his analysis. He also neither asked for nor
`
`received an opinion from any of AlmondNet’s technical experts regarding which of the hundreds
`
`of features of Amazon’s accused advertising system fall within the scope of the asserted claims
`
`under AlmondNet’s infringement theory. Instead, he purports to value a vague and high-level
`
`characterization of the purported “benefits” of the claims—i.e., the functionalities the patents
`
`purportedly enable. As a result, by Mr. Bergman’s own admission, his damages estimate captures
`
`value from numerous unpatented features—such as real-time bidding, machine-learning
`
`algorithms, and Amazon’s data regarding its customer’s behavior (“Amazon first-party data”)—
`
`that AlmondNet’s technical expert concedes are neither disclosed in nor enabled by the patents.
`
`Mr. Bergman’s damages estimate thus cannot reflect the incremental value provided by the asserted
`
`claims, as Federal Circuit law requires, and it must be excluded.
`
`Second, Mr. Bergman estimates an arbitrary “bargaining split,” under which AlmondNet
`
`would receive between 35% and 50% of the accused revenue. The only “support” he provides for
`
`this bargaining split is (1)
`
`; and (2) Mr. Bergman’s experience,
`
`
`
`
`
`.
`
`Mr. Bergman’s opinion is unsupported, arbitrary, and unreliable: the only quantitative evidence
`
`1
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 6 of 20
`
`
`
`he cites is irrelevant on its face, and his “experience”
`
`
`
`cannot be meaningfully tested or challenged. Mr. Bergman should not be permitted to present
`
`these opinions at trial; the Court should exercise its gatekeeping responsibilities under Daubert
`
`and exclude his testimony in its entirety.
`
`I.
`
`FACTUAL BACKGROUND
`A.
`
`The patents-in-suit claim particular business methods relating to serving off-
`site advertisements.
`
`The three patents-in-suit generally relate to business arrangements used in targeted
`
`advertising on the Internet. (U.S. Patent Nos. 7,822,639 (“the ’639 patent”); 8,244,586 (“the ’586
`
`patent”); 8,671,139 (“the ’139 patent”).) The’639 patent and ’586 patent generally provide a
`
`business solution to the problem of saturation—when a first website cannot accommodate
`
`additional advertisements and overflow ads for a particular visitor are instead displayed on a
`
`second website. The asserted claims of the ’639 and ’586 patents1 generally require a process
`
`wherein a user that visits a first website is tagged (i.e., using a cookie), and a second website with
`
`capacity displays a targeted ad upon recognizing the tagged user. (See ’639 patent, claim 24; ’586
`
`patent, claim 1). The first and second websites share the advertiser revenue from the display of
`
`the targeted ad on the second website. (Id.) The ’139 patent generally discloses an “automatic
`
`system” for calculating profit for an ad placement and placing ads only when it is profitable. The
`
`asserted claims of the ’139 patent2 require a process wherein an attribute about an electronic visitor
`
`to a website is recorded, and that attribute is used to send targeted ads to the visitor when they
`
`access the second website provided a condition—e.g., the profitability of displaying the ad—is
`
`met. (See ’139 patent, claim 37.) Thus, each of the patents-in-suit generally requires collecting
`
`
`1 AlmondNet accuses Amazon of infringing claims 24, 29, 31, 32 and 44 of U.S. Patent No.
`7,822,639 and claims 1, 3, 4, 11–13, 16, 17, and 20 of U.S. Patent No. 8,244,586..
`2 AlmondNet accuses Amazon of infringing claims 37, 42, 43, 45, and 48 of the ’139 patent.
`
`2
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 7 of 20
`
`
`
`information about and tagging a visitor to a first website and recognizing and presenting an ad to
`
`the user on a second website. But the patents do not claim or describe any new solution for tagging
`
`or identifying users, targeting users with advertisements, or delivering advertisements for display
`
`on websites.
`
`B.
`
`Amazon’s sophisticated ad platform uses real-time bidding, machine
`learning, and Amazon’s extensive customer data to help advertisers and
`publishers efficiently deliver targeted advertisements.
`
`AlmondNet accuses certain systems of the Amazon Ads Platform that assist advertisers
`
`who want to display targeted advertisements on Internet sites operated by third party publishers
`
`through real-time bidding. (Ex. A3 (Koskinen Rep.) ¶¶ 43-48, 300.)
`
`
`
`
`
`
`
`
`
`
`
`
`
`Real-time bidding (RTB) is an automated, programmatic way for advertisers to bid for
`
`advertisement opportunities through a real-time auction that runs when a visitor arrives at a
`
`website. (Ex. B (Houh Reb. Rep.) ¶ 65). In a simplified real-world example, a publisher’s website
`
`like cnn.com may have space to display a banner advertisement. (Id. ¶ 66.) When the user visits
`
`cnn.com, the opportunity to fill the ad spot with an ad to that user triggers a bid request—an
`
`opportunity for advertisers to submit bids to a real-time auction. (Id.) The winner of the auction
`
`then has their ad shown to the user. (Id.) The entire process occurs within around 100
`
`
`3 Each Exhibit referenced herein is attached to the Declaration of Ravi R. Ranganath submitted
`herewith.
`
`3
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 8 of 20
`
`
`
`milliseconds. (Id.)
`
`As real-time bidding has grown in sophistication and complexity, large internet companies
`
`like Google, Meta, Amazon, and others have leveraged their own extensive, high-quality data and
`
`expertise in machine learning and artificial intelligence to optimize real-time bids. (Id. ¶ 81; Ex.
`
`C (Bergman Rep.) ¶ 118.)
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Amazon’s bid optimization systems maximize
`
`advertisers’ budgets while improving the performance of the advertisements based on an
`
`advertiser’s campaign goal (e.g., increased reach, clicks, views, subsequent purchases, and other
`
`actions as a result of the ad). (Id.; see generally id. ¶¶ 148-181
`
`
`
`4
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 9 of 20
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`C. Mr. Bergman’s damages model captures value from unclaimed features.
`
`Mr. Bergman does not account for the sophisticated, unclaimed features employed by
`
`Amazon’s advertising platform in his damages analysis.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Notably, Mr. Bergman made no attempt to determine if those features were attributable in
`
`5
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 10 of 20
`
`
`
`any way to the asserted claims.4
`
`
`
`
`
`
`
`
`
`
`
`Nor did Mr. Bergman ask AlmondNet’s technical expert, Dr. Koskinen, for an opinion with
`
`respect to
`
`
`
`In fact, if Mr. Bergman had asked Dr. Koskinen, he would have learned
`
`
`4 Mr. Bergman did concede that
` (Ex. D (Bergman Dep.) at 78:21–79:6 (emphasis added).)
`
`6
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 11 of 20
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Mr. Bergman also made no attempt to quantify the value Amazon derives from real-time
`
`bidding, machine learning, or Amazon’s first-party data.
`
`
`
`
`
`
`
`
`
` Thus, Mr.
`
`Bergman does not, and cannot, remove the value of these unclaimed features from his royalty
`
`calculation. (See Ex. D (Bergman Dep) at 85:8–13
`
`
`
`
`
`
`
`Although he does not excise the value of unpatented features from his damages calculation,
`
`Mr. Bergman does make two adjustments to Amazon’s advertising revenue which he labels
`
`“apportionment” to arrive at what he claims is the “incremental benefit” from the asserted claims.
`
`First, he attempts to reduce
`
`
`
`(Ex. C (Bergman Rep.) ¶¶ 238, 253–54.) Next, Mr. Bergman claims to
`
`
`
`7
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 12 of 20
`
`
`
` But neither of these purported “apportionment”
`
`steps seeks to distinguish value from claimed versus unclaimed features. His royalty base—
`
`—is what Mr. Bergman asserts is
`
`
`
`
`
` (Id. ¶ 268.) In other words, Mr. Bergman’s final damages
`
`calculation captures revenue that advertisers pay to Amazon—in the form of first-party targeting
`
`fees—to utilize Amazon’s vast user dataset and sophisticated machine-learning algorithms that
`
`optimize bids to maximize performance and value.
`
`D. Mr. Bergman estimates an arbitrary bargaining split.
`
`Once Mr. Bergman determines what he claims is the incremental benefit from the asserted
`
`claims, he opines that AlmondNet would receive between 35% and 50% of that value. (Ex. C
`
`(Bergman Rep.) ¶¶ 419, 433–34; Ex. D (Bergman Dep.) at 162:16–25.)
`
`Mr. Bergman provides two bases for this estimated bargaining split. First, Mr. Bergman
`
`relies on his experience
`
` (Ex. C ¶ 420.) Despite purporting to rely on these
`
`. (Ex. D (Bergman Dep.) at 162:5–
`
`
`
`
`
`15.) Second, Mr. Bergman relies on a handful of
`
` entered into
`
`by non-party
`
`, (Ex. D (Bergman Dep) at 165:9–16), and various third parties, (Ex. C
`
`(Bergman Rep.) ¶¶ 428–29); Exs. F-K (
`
` ).)
`
`
`
`
`
`
`
`8
`
`
`
`
`
`
`
`
`
`
`
`
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 13 of 20
`
`
`
` Nor does Mr. Bergman rely on an opinion of a technical expert to determine that the technology
`
`at issue
`
` is comparable to the accused Amazon products. (See Ex. C
`
`(Bergman Rep.) ¶¶ 426–33.) They are obviously not comparable:
`
` (Id. ¶ 126; Ex. M (Shkedi Mar. 21, 2023 30(b)(6) Dep.) at 132:21–133:1
`
`
`
`
`
`
`
`
`
`); Exs. F-K.) Moreover, Mr. Bergman
`
`concedes that
`
` would not have been part of the hypothetical negotiation between
`
`AlmondNet and Amazon that would have occurred in or around August 2012—because
`
`
`
`did not own, and has never owned, any of the asserted patents. (Ex. D at 174:12–16; Ex. C
`
`(Bergman Rep.) ¶ 436.)
`
`In other words, the only quantitative evidence supporting Mr. Bergman’s incremental
`
`benefit split comes from
`
`agreements entered into between non-parties for noncomparable
`
`technology dated
`
` after the hypothetical negotiation.
`
`II. MR. BERGMAN’S UNRELIABLE TESTIMONY SHOULD BE EXCLUDED
`
`Under Federal Rule of Evidence 702, expert testimony must “be the product of reliable
`
`principles and methods applied to sufficient facts or data.” Power Integrations, Inc. v. Fairchild
`
`Semiconductor Int’l, Inc., 711 F.3d 1348, 1373 (Fed. Cir. 2013). This standard places a special
`
`obligation upon the court to act as a “gatekeeper” and ensure that expert testimony “is not only
`
`relevant, but reliable.” Kumho Tire Co. v. Carmichael, 526 U.S. 137, 147 (1999) (quoting Daubert
`
`v. Merrell Dow Pharms., Inc., 509 U.S. 579, 589 (1993)). Mr. Bergman’s failure to apportion his
`
`royalty analysis to the incremental value of the claimed inventions and his arbitrary bargaining
`
`9
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 14 of 20
`
`
`
`split each independently render his damages opinions both irrelevant and unreliable, and warrant
`
`excluding the entirety of his testimony.
`
`A. Mr. Bergman’s failure to apportion value to the claimed inventions renders
`his testimony unreliable.
`
`“[A] patentee ‘must in every case give evidence tending to separate or apportion the
`
`defendant’s profits and the patentee’s damages between the patented feature and the unpatented
`
`features . . . [unless] the entire value of the whole machine, as a marketable article, is properly and
`
`legally attributable to the patented feature.” VirnetX, Inc. v. Cisco Sys., Inc., 767 F.3d 1308, 1326
`
`(Fed. Cir. 2014) (quoting Garretson v. Clark, 111 U.S. 120, 121 (1884)). Proof of damages must
`
`be carefully tied to “the claimed invention’s footprint in the market place.” Ericsson, Inc. v. D-
`
`Link Sys., 773 F.3d 1201, 1226 (Fed. Cir. 2014) (“The essential requirement is that the ultimate
`
`reasonable royalty award must be based on the incremental value that the patented invention adds
`
`to the end product.”); CSIRO v. Cisco Sys., 809 F.3d 1295 (Fed. Cir. 2015) (“[D]amages awarded
`
`for patent infringement must reflect the value attributable to the infringing features of the product,
`
`and no more”); VirnetX, 767 F.3d at 1326 (“No matter what the form of the royalty, a patentee
`
`must take care to seek only those damages attributable to the infringing features.”).
`
`Despite conceding
`
`Bergman fails to do so. He acknowledges
`
`
`
` Mr.
`
`
`
`, (id. at 70:15–17; 77:20–78:6; 82:2–9; 95:17–96:14; 105:18–25), and that he never
`
`, (id. at 88:2–12; 104:24–105:9; 115:14–21.)
`
`
`
`Lacking the bare minimum information required to apportion between unclaimed and claimed
`
`features, Mr. Bergman takes no steps to do so, or even assign value to the numerous unclaimed
`
`10
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 15 of 20
`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 15 of 20
`
`features—such as real-time bidding, machine learning, and Amazon’s first-party data—in the
`
`accused system.(Jd. at 83:3-10; 91:20-92:2: 100:1-6.) Mr. Bergmanadmits|i
`
`i. at 7526-11; 77:11-18; 78:18-20; 81:19-25; 82:11-15; 88:21-89:3; 91:8-13;
`
`105:11—17; 106:12—107:9; accord Ex. E (Koskinen Dep.) at 66:3-67:12.)
`
`Mr. Bergman’s only explanation for his failure to consider the value contributed by these
`
`citcl es5s
`I> (22:22Dep) at 1172-
`
`10.) In other words, Mr. Bergman appearsto be estimating “hold-up” damages. But this analysis
`
`violates black-letter Federal Circuit law, which provides “whether viewed as valuable, important,
`
`or even essential, the patented feature must be separated.” Finjan, Inc. v. Blue Coat Sys., Inc., 879
`
`F.3d 1299, 1311 (Fed. Cir. 2018) (emphases added).
`
`In other words, even if Mr. Bergman is
`
`I5 2220 separate the valve fom
`
`the claimed technology from the unclaimed technology and other value provided by Amazon—
`
`i.e., Amazon’s first-party data and machine learning algorithms which optimize advertising bids
`
`used in the real-time-bidding process. His failure to do so renders his opinion unreliable and
`
`inadmissible. Ericsson, 773 F.3d at 1226 (“[T]he ultimate reasonable royalty award must be based
`
`on the incrementalvaluethat the patented invention adds to the end product.”).®
`
`> As explained in Amazon’s concurrently filed Motion for Summary Judgment of
`Unpatentability Under § 101, AlmondNet’s incredibly broad reading of these claims renders them
`patent-ineligible.
`® Even Mr. Bergman’s assumption
`
`1s unsupported,
`by a technical expert opinion
`
`or otherwise.
`
`11
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 16 of 20
`
`
`
`The Court should also reject Mr. Bergman’s post-hoc explanation that he apportioned
`
`between claimed and unclaimed features through his bargaining split, for at least two reasons. (Ex.
`
`D (Bergman Dep.) at 92:10–21.) First, he estimates and applies his bargaining split only after he
`
`claims to identify the incremental benefit from the claimed inventions. (Ex. C (Bergman Rep.) ¶
`
`268 (identifying the “incremental benefit”; ¶¶ 417–418 (bargaining split)).) Thus, the bargaining
`
`split cannot itself be a form of apportionment. Second, as explained below, Mr. Bergman’s
`
`bargaining split is also arbitrary and unreliable, and should also be excluded for the reasons below.
`
`B. Mr. Bergman’s arbitrary bargaining split renders his opinions unreliable.
`
`Courts routinely exclude expert testimony based on arbitrary and unsupported profit splits.
`
`Uniloc USA, Inc. v. Microsoft Corp., 632 F.3d 1292, 1317-18 (Fed. Cir. 2011) (finding a 25/75
`
`profit split “arbitrary” and “fundamentally flawed”); LaserDynamics, Inc. v. Quanta Comput., Inc.,
`
`694 F.3d 51, 69 (Fed. Cir. 2012) (finding a 33/67 profit split arbitrary and “alone justify[ing]
`
`excluding” expert’s testimony); Bayer HealthCare LLC v. Baxalta Inc., No. 16-cv-1122-RGA,
`
`2019 WL 330149, *6 (D. Del. Jan. 25, 2019) (similar). This Court should do the same.
`
`Here, Mr. Bergman’s opinion that Amazon and AlmondNet would agree that AlmondNet
`
`would take
`
`of the “incremental benefit” from the patented technology is unsupported
`
`and arbitrary. The only record evidence he relies on, and in fact the basis for his
`
`numbers, are
`
` between non-party
`
` and other non-parties
`
`(Exs. F-K.)
`
`
`
`
`
`
`
`
`
` (Id.; Ex. C (Bergman Rep.) ¶¶ 126, 426–33; Ex. D (Bergman Dep) at 165:9–16; id. at
`
`166:12–15; Ex. M (Shkedi Mar. 21 2023 30(b)(6) Dep.) at 132:21–133:1.) Nor does Mr. Bergman
`
`rely on any opinion that the technology at issue
`
` are comparable in any
`
`way to the accused Amazon technology. (See Ex. C (Bergman Rep.) ¶¶ 426–33 (
`
`
`
`12
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 17 of 20
`
`
`
`).) Moreover,
`
`not only is not a party to this case and did not own any of the
`
`patents-in-suit as of the date of the hypothetical negotiation,
`
`.
`
`(Ex. C (Bergman Rep.) ¶¶ 426–33; Ex. D (Bergman Dep) at 166:12–15; Ex. L at 7-8.) Thus, no
`
`reliable or relevant record evidence supports Mr. Bergman’s bargaining split.
`
`Mr. Bergman alternatively purports to rely on his own personal experience
`
`
`
`. (Ex. C (Bergman Rep.) ¶
`
`420; Ex. D (Bergman Dep.) at 162:5–15.) But these sorts of opinions are “nothing more than ipse
`
`dixit statements” because they “are not based on reliable principles and methods that can be
`
`accurately reproduced, examined, and challenged.” BMC Software, Inc. v. ServiceNow, Inc., No.
`
`2:14-cv-903-JRG, Dkt. No. 328 at 3–4 (E.D. Tex. Feb. 1, 2016).
`
`Stripped of this unreliable evidence, Mr. Bergman’s bargaining split is as unsupported and
`
`arbitrary as the “25 percent rule of thumb” the Federal Circuit has repeatedly found “fundamentally
`
`flawed” and inadmissible under Daubert and the Federal Rules of Evidence. The “25 percent rule
`
`of thumb” is a generic approach used approximate the reasonable royalty rate a manufacturer of a
`
`patented product would be willing to offer to pay to a patentee during a hypothetical negotiation.
`
`Uniloc, 632 F.3d at 1312. In Uniloc v. Microsoft, plaintiff’s expert using the “25 percent rule of
`
`thumb,” a generic principle suggesting that a licensee would pay 25% of its profits as a royalty,
`
`concluded that “25% of the value of the product would go to the patent owner and the other 75%
`
`would remain with Microsoft, resulting in a baseline royalty rate of $2.50 per license issued.” Id.
`
`at 1311. Finding that there must be a “basis in fact to associate the royalty rates . . . to the particular
`
`hypothetical negotiation at issue in the case,” the Federal Circuit rejected the use of the “25 percent
`
`rule of thumb” as “unrelated to the facts of the case,” “arbitrary,” and “fundamentally flawed,”
`
`and confirmed inadmissibility of the expert opinions under Daubert. Id. at 1312, 1318. Since
`
`13
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 18 of 20
`
`
`
`Uniloc, the Federal Circuit and numerous district courts have excluded expert opinions relying on
`
`similarly arbitrary and unsupported bargaining splits, like the one Mr. Bergman employs in this
`
`case. For example, in LaserDynamics, the patent owner’s expert applied a one-third profit split
`
`but offered no analysis to support it. LaserDynamics, 694 F.3d at 69. The Federal Circuit found
`
`this profit split “echo[ed] the kind of arbitrariness of the ‘25% Rule,’” and would “alone justify
`
`excluding” the expert’s opinion. Id.
`
`Under Uniloc and its progeny, Mr. Bergman’s bargaining split opinions are arbitrary,
`
`unsupported, and inadmissible under Daubert. Mr. Bergman failed to tie his
`
`bargaining
`
`split range to any reliable evidence of record. The Court should therefore exclude his testimony
`
`under Fed. R. Evid. 702 for this reason alone.
`
`III. CONCLUSION
`
`As Mr. Bergman failed to apportion value derived from the accused products to the claimed
`
`inventions, and as he relies on an arbitrary and unsupported bargaining split, his opinions should
`
`be struck in their entirety.
`
`
`
`
`
`14
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 19 of 20
`
`
`
`Dated: August 23, 2023
`
`
`
`
`Of Counsel:
`
`Deron R. Dacus (TX Bar #00790553)
`ddacus@dacusfirm.com
`THE DACUS FIRM, P.C.
`821 ESE Loop 323, Suite 430
`Tyler, TX 75701
`Tel: (903) 705-1117
`Fax: (903) 581-2543
`
`
` Respectfully submitted,
`
`
`
`By: /s/ Ravi R. Ranganath
`J. David Hadden (CA Bar No. 176148)
`Email: dhadden@fenwick.com
`Saina S. Shamilov (CA Bar No. 215636)
`Email: sshamilov@fenwick.com
`Ravi R. Ranganath (CA Bar No. 272981)
`Email: rranganath@fenwick.com
`Johnson K. Kuncheria (TX Bar No. 24070092)
`Email: jkuncheria@fenwick.com
`Johnathan L. Chai (CA Bar No. 339315)
`Email: jchai@fenwick.com
`FENWICK & WEST LLP
`801 California Street
`Mountain View, CA 94041
`Tel: (650) 988-8500
`Fax: (650) 938-5200
`
`Eric B. Young, (CA Bar No. 318754)
`Email: eyoung@fenwick.com
`Dargaye Churnet (Admitted Pro Hac Vice)
`Email: dchurnet@fenwick.com
`Christopher L. Larson (Admitted Pro Hac Vice)
`Email: clarson@fenwick.com
`FENWICK & WEST LLP
`555 California Street, 12th Floor
`San Francisco, CA 94104
`Tel: (415) 875-2300
`
`Jeffrey A. Ware (WA Bar No. 43779)
`Email: jware@fenwick.com
`FENWICK & WEST LLP
`401 Union Street, 5th Floor
`Seattle, WA 98101
`Tel: (206) 389-4510
`
`Jessica Lin (NY Bar No. 5035860)
`Email: jessica.lin@fenwick.com
`Eric Menist (NY Bar No. 5721568)
`Email: emenist@fenwick.com
`FENWICK & WEST LLP
`902 Broadway, 18th Floor
`New York, NY 10010
`Tel: (212) 921-2001
`
` Counsel for Defendants
`AMAZON.COM, INC., AMAZON.COM SERVICES
`LLC, and AMAZON WEB SERVICES, INC.
`
`15
`
`

`

`Case 6:21-cv-00898-ADA Document 141 Filed 08/30/23 Page 20 of 20
`
`CERTIFICATE OF SERVICE
`
`
`
`I hereby certify that all counsel of record who are deemed to have consented to electronic
`
`service are being served with a true and correct copy of this document on August 23, 2023, via
`
`the Court’s CM/ECF system per Local Rule CV 5(a)(3). Additionally, this document and the
`
`attachments thereto were served via email on all counsel of record.
`
` /s/ Ravi R. Ranganath
` Ravi R. Ranganath
`
`
`
`
`
`16
`
`