Case 3:20-cv-06754-WHA Document 813 Filed 06/09/23 Page 1 of 18
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`UNITED STATES DISTRICT COURT
`
`
`NORTHERN DISTRICT OF CALIFORNIA
`
`SONOS, INC.,
`
`
`
`Plaintiff,
`
`v.
`
`GOOGLE LLC,
`
`Defendant.
`
`
`
`
`
`
`
`
`
`No. C 20-06754 WHA
`No. C 21-07559 WHA
`
`(Consolidated)
`
`MEMORANDUM OPINION RE
`SONOS’S DAMAGES THEORY
`
`
`
`INTRODUCTION
`
`This memorandum opinion explains an oral ruling that struck a major part of plaintiff’s
`
`damages theory before jury deliberations began.
`
`Sonos, Inc. sued Google LLC for patent infringement. Google sued Sonos for a
`
`declaratory judgment. The related actions were consolidated for trial, which concluded two
`
`weeks ago.
`
`The final pretrial order had deferred ruling on two of Google’s motions in limine,
`
`cautioning that there were “serious questions about Sonos’s damages theory and associated
`
`opinion,” but allowing Sonos to put on the contested evidence “with the understanding that the
`
`undersigned may strike it from the record, tell the jury to disregard it, and grant one of these
`
`motions in limine under Rule 50, if appropriate, having benefitted from hearing the evidence
`
`and cross-examination” (Dkt. No. 660 at 2). That is ultimately what happened.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`Northern District of California
`
`United States District Court
`
`

`

`Case 3:20-cv-06754-WHA Document 813 Filed 06/09/23 Page 2 of 18
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`To the extent stated herein, Google’s first motion in limine was GRANTED, and Google’s
`
`second motion in limine was DENIED AS MOOT.
`
`STATEMENT
`
`Sonos offered Mr. James Malackowski as an expert to testify regarding its claim for
`
`damages for Google’s infringement of the two remaining patents-in-suit, U.S. Patent Nos.
`
`10,848,885 and 10,496,966. These patents cover technology for customizing and saving
`
`overlapping groups of smart speakers or other “zone players” according to a common theme,
`
`and then later invoking such groups, called “zone scenes,” on demand. Whereas the ’885
`
`patent claims the technology from the perspective of a zone player (e.g., a smart speaker), the
`
`’966 patent claims the technology from the perspective of a computing device that controls at
`
`least three zone players (e.g., a smart phone). As such, the accused products and requested
`
`damages differed. With respect to the ’885 patent, Sonos accused Google media players (e.g.,
`
`a Google Nest Mini speaker). With respect to the ’966 patent, Sonos accused all smartphones
`
`and other computing devices that have or had the Google Home application installed (e.g., an
`
`iPhone with Google Home).1
`
`To calculate damages, Mr. Malackowski assumed that Sonos and Google would have
`
`used a subscription price of a third-party application offered on the Google Play Store in
`
`hypothetical patent licensing negotiations, apportioning this price to arrive at a hypothetical
`
`royalty. Specifically, he relied upon the opinion of Sonos’s technical expert, Dr. Kevin
`
`Almeroth, that a free third-party scripting application could provide technology that was
`
`comparable to the claimed invention (Malackowski Rpt. 80; Tr. 1120:21–24, 1125:23–1126:1).
`
`Because a user of this application eventually had to pay for an add-on subscription in order for
`
`the application to provide the ostensibly comparable technology, Mr. Malackowski proceeded
`
`to use an optional monthly subscription price as a starting point for calculating a royalty,
`
`
`1 At trial, contrary to Sonos’s position, the undersigned ruled that the mere installation of the
`Google Home application on a computing device did not itself infringe, and that Google was
`incapable of infringing the ’966 patent unless the accused products were networked with at least
`three zone players that might be added to overlapping zone scenes using Google Home (Dkt.
`No. 762 at 15; Tr. 1403:5–11).
`
`2
`
`Northern District of California
`
`United States District Court
`
`

`

`Case 3:20-cv-06754-WHA Document 813 Filed 06/09/23 Page 3 of 18
`
`
`
`apportioning downward to account for incomparable features and Georgia-Pacific factors
`
`(Malackowski Rpt. 84–90, 94–124; Malackowski Reply Rpt. 24; Tr. 1125:17–1126:8, 1128:3–
`
`8, 1134:13–1137:4). From a $1.99 monthly fee charged to some premium users of a third-
`
`party application, Mr. Malackowski derived a $90 million damages award, mostly attributable
`
`to infringement of the ’966 patent.2
`
`This third-party application is called If This Then That, or IFTTT. According to IFTTT,
`
`“IFTTT can do anything!” (Dkt. No. 607-3 at 1). With less hyperbole, it claims to offer users
`
`“the best way to integrate apps, devices, and services,” providing small software applications
`
`or “applets” that use a combination of “triggers” (if’s) and “actions” (then’s) to create
`
`automations (ibid.; Dkt. No. 607-2 at 1). By way of example, IFTTT applets can automate
`
`sending a notification if the International Space Station passes over one’s house, sending an
`
`email if the forecast suggests it will rain the following day, or sending tracked hours to a
`
`calendar application if one is at work (Dkt. No. 607-3 at 2). An IFTTT user can use published
`
`IFTTT applets, like the applets just described, or create their own to control an array of
`
`integrated products, including those of Sonos and Google (Dkt. No. 607-4 at 1; Tr. 828:17–18).
`
`From its inception through September 2020, IFTTT offered all of its services for free,
`
`with no limitations in terms of applet count or complexity. In September 2020, however, it
`
`introduced paid subscription plans for its advanced functionalities — almost one year after the
`
`hypothetical negotiation for the ’966 patent, and a couple months before the hypothetical
`
`negotiation for the ’885 patent (Malackowski Rpt. 84; Tr. 1129:25–1130:2, 1204:24–25). The
`
`parties stipulated that the hypothetical negotiations would have taken place when the patents
`
`issued in November 2019 and November 2020 for the ’966 patent and the ’885 patent,
`
`respectively (Dkt. No. 615 at 7).
`
`
`2 Breaking it down, he derived $12,246,294 in damages for infringement of the ’885 patent, and
`$77,546,923 in damages for infringement of the ’966 patent (Tr. 1136:22–1137:4). In his opening
`report, Mr. Malackowski derived $144,373,860 in damages for infringement of the ’966 patent
`(Malackowski Rpt. 130). That number was never introduced into evidence, however, as Mr.
`Malackowski implemented quantitative adjustments in his reply report that decreased it to
`$77,546,923 (Malackowski Reply Rpt. 24).
`
`3
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`Northern District of California
`
`United States District Court
`
`

`

`Case 3:20-cv-06754-WHA Document 813 Filed 06/09/23 Page 4 of 18
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`Google filed two motions in limine to exclude evidence related to Sonos’s damages
`
`theory (Dkt. Nos. 607, 610). According to Google, the theory was unreliable in light of its use
`
`of IFTTT as its foundation. In the first motion, Google sought to exclude the opening and
`
`reply reports and testimony of Mr. Malackowski, as well as the related report language and
`
`testimony of Dr. Almeroth. In the second motion, Google sought to exclude only select report
`
`language and testimony of Mr. Malackowski. Sonos opposed both motions (Dkt. Nos. 607-11,
`
`610-6). It also filed a separate trial brief in support of its use of IFTTT (Dkt. No. 735).3
`
`The final pretrial order deferred ruling on the motions, and both witnesses testified before
`
`the jury (Dkt. No. 660 at 2; Tr. 669:10–12, 1077:18–20). After the direct examinations and
`
`cross-examinations, a ruling from the bench struck Sonos’s damages theory based on IFTTT as
`
`unreliable (Tr. 1402:20–1403:2). The final charge instructed the jury that it could not factor
`
`any information regarding IFTTT into its calculation of damages, including the damages
`
`figures that Mr. Malackowski derived. But it allowed the jury to consider his testimony
`
`unrelated to IFTTT and to use other evidence to calculate damages, such as admitted license
`
`agreements (Dkt. No. 762 at 21–22). The jury ultimately calculated a damages award of
`
`$35,507,183.40 for infringement of the ’885 patent, having found no infringement of the ’966
`
`patent (Dkt. No. 774). For the record on appeal, this memorandum opinion provides the
`
`reasons for the ruling that struck all things IFTTT.
`
`ANALYSIS
`
`“[E]stimating a reasonable royalty is not an exact science.” Summit 6, LLC v. Samsung
`
`Elecs. Co., 802 F.3d 1283, 1296 (Fed. Cir. 2015). “[W]hile all approximations involve some
`
`degree of uncertainty, the admissibility inquiry centers on whether the methodology employed
`
`is reliable.” Ibid. (citing Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 589–95 (1993)).
`
`“Rule 702 grants the district judge the discretionary authority, reviewable for its abuse, to
`
`determine reliability in light of the particular facts and circumstances of the particular case.”
`
`Kumho Tire Co. v. Carmichael, 526 U.S. 137, 158 (1999).
`
`
`3 Unless otherwise indicated, all report and deposition excerpts referenced in this memorandum
`opinion were attached as exhibits to the briefing on the motions in limine.
`4
`
`Northern District of California
`
`United States District Court
`
`

`

`Case 3:20-cv-06754-WHA Document 813 Filed 06/09/23 Page 5 of 18
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`“While questions regarding which facts are most relevant for calculating a reasonable
`
`royalty are properly left to the jury, a critical prerequisite is that the underlying methodology
`
`be sound.” Virnetx, Inc. v. Cisco Sys., Inc., 767 F.3d 1308, 1328 (Fed. Cir. 2014). “[A]
`
`reasonable or scientifically valid methodology is nonetheless unreliable where the data used is
`
`not sufficiently tied to the facts of the case. Likewise, ideal input data cannot save a
`
`methodology that is plagued by logical deficiencies or is otherwise unreasonable.” Apple Inc.
`
`v. Wi-LAN Inc., 25 F.4th 960, 971 (Fed. Cir. 2022) (internal quotations and citations omitted).
`
`The hypothetical negotiation “attempts to ascertain the royalty upon which the parties
`
`would have agreed had they successfully negotiated an agreement just before infringement
`
`began.” Lucent Techs., Inc. v. Gateway, Inc., 580 F.3d 1301, 1324 (Fed. Cir. 2009). It is a
`
`well-established methodology for estimating a royalty. The hypothetical negotiation based on
`
`IFTTT, however, is not. Here, this methodology was unreliable for the following reasons.
`
`First, IFTTT could not be used as a “benchmark” product in the reasonable royalty
`
`analysis. That a technical expert could jerry-rig this generic scripting application to
`
`approximate some claim limitations in no way shows that this application had a feature that
`
`was technologically comparable to the claimed invention.
`
`Second, even assuming arguendo that IFTTT could be used as a benchmark product, a
`
`price eventually set for an optional add-on subscription plan could not be a starting point for
`
`the reasonable royalty analysis. The price had no relationship to the claimed invention.
`
`Third, even assuming arguendo that the price could be a starting point, it was not
`
`apportioned to the incremental value of the claimed invention. The primary apportionments
`
`did not factor out the application’s incomparable features and/or were not tied to the facts of
`
`the case.
`
`1.
`
`
`IFTTT COULD NOT QUALIFY AS A BENCHMARK PRODUCT.
`
`One way to calculate a reasonable royalty is to “value the infringed features based upon
`
`comparable features in the marketplace.” Apple Inc. v. Motorola, Inc., 757 F.3d 1286, 1315
`
`(Fed. Cir. 2014), overruled on other grounds, Williamson v. Citrix Online, LLC, 792 F.3d 1339
`
`(Fed. Cir. 2015). Both sides recognize that the Federal Circuit has allowed the use of
`
`5
`
`Northern District of California
`
`United States District Court
`
`

`

`Case 3:20-cv-06754-WHA Document 813 Filed 06/09/23 Page 6 of 18
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`“benchmark” products with technologically comparable features to inform a reasonable royalty
`
`analysis (Sonos IFTTT Trial Br. 1; Google JMOL #1 Br. 21 (each citing id. at 1318)). See also
`
`i4i Ltd. P’ship v. Microsoft Corp., 598 F.3d 831, 853 (Fed. Cir. 2010).4
`
`According to Sonos, IFTTT was one such benchmark product with a technologically
`
`comparable feature, and this application could therefore be used to calculate a reasonable
`
`royalty (Sonos IFTTT Trial Br. 6–7). Dr. Almeroth provided the technical foundation. He
`
`opined that an IFTTT user could build applets that provide functionality that is technologically
`
`comparable to the claimed zone-scene technology (Almeroth Rpt. ¶ 798; Tr. 821:23–822:3,
`
`829:8–22). As a proof of concept, he set up a first IFTTT applet with actions that allowed him
`
`to play music on a first set of Sonos speakers (e.g., “Garden”) and a second IFTTT applet with
`
`actions that allowed him to play music on a second set of Sonos speakers (e.g., “Evening”),
`
`wherein the two sets “overlapped” (i.e., shared at least one speaker), thereby mimicking the
`
`claims-in-suit (Almeroth Rpt. ¶¶ 798, 802–07; Tr. 826:11–18, 827:14–19; see also TX 442).
`
`According to Dr. Almeroth, his IFTTT applets were technologically comparable to the claimed
`
`invention because they customized and saved overlapping groups (e.g., “Garden” and
`
`“Evening” shared one speaker), and because they caused playback to start on at least two
`
`speakers at the same time when a group was invoked (e.g., “Garden” and “Evening” each had
`
`more than one speaker) (Almeroth Rpt. ¶ 814; Tr. 827:9–12, 22–25).
`
`These IFTTT applets could not actually replicate the claimed zone-scene technology,
`
`however. As Dr. Almeroth himself acknowledged, his applets were not “configured for the
`
`
`4 More precisely, both sides recognize this now. Neither mentioned benchmark products with
`technologically comparable features in the briefing on the motions in limine. That briefing
`selectively quoted cases analyzing benchmark licenses, with alterations to make them fit products.
`For example, Google’s first motion stated: “While a patentee may rely on comparable technology
`to support a proposed royalty, it has the ‘burden to prove that the [technology is] sufficiently
`comparable’ to the technology and value of the asserted patent.” (Google MIL #1 Br. 3 (quoting
`Lucent, 580 F.3d at 1329)). Where Google inserted “technology is,” Lucent included “licenses
`were” (see also Google MIL #1 Opp. 5 (quoting Virnetx, 767 F.3d at 1330)). Of course, one can
`“use the royalty rate from sufficiently comparable licenses” to calculate a royalty, but the Federal
`Circuit has distinguished this from “valu[ing] the infringed features based upon comparable
`features in the marketplace.” See Apple, 757 F.3d at 1315. The parties incorporated the law on
`benchmark products with technologically comparable features in (unsolicited) later-filed briefs
`(see Sonos IFTTT Trial Br. 1–3, 6–7; Google JMOL #1 Br. 21).
`
`
`6
`
`Northern District of California
`
`United States District Court
`
`

`

`Case 3:20-cv-06754-WHA Document 813 Filed 06/09/23 Page 7 of 18
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`synchronous playback of media,” an important aspect of the invention described in the
`
`specification and covered by the claims (Almeroth Rpt. ¶ 815; Tr. 829:2–7; see, e.g., ’885 and
`
`’966 patents col. 3:9–12; ’885 patent col. 12:19; ’966 patent col. 12:12). He opined that the
`
`speakers grouped, saved, and invoked using an IFTTT applet would have had an echo caused
`
`by clock drift or differences in playback start time (Almeroth Rpt. ¶ 815; Tr. 922:14–923:16).
`
`This would certainly be annoying to a listener who could hear sound coming from more than
`
`one speaker. Nevertheless, Dr. Almeroth concluded that the applets’ stab at replication of
`
`overlapping zone scenes was sufficient to show technological comparability (Almeroth Rpt.
`
`¶ 816; Tr. 822:6–18, 829:3–7).
`
`*
`
`*
`
`*
`
`The thrust of Sonos’s argument was as follows: because IFTTT was capable of
`
`customizing, saving, and invoking overlapping groups of speakers in a way that was
`
`comparable to the claimed technology, IFTTT was a benchmark product with a technologically
`
`comparable feature. But mere capability does not confer comparability.
`
`Let’s start with the caselaw. In all of the instances in which the Federal Circuit has
`
`allowed for the value of benchmark products with technologically comparable features to
`
`inform the reasonable royalty analysis, those products did not need to be separately configured
`
`in order to become benchmark products with technologically comparable features. Rather,
`
`they were benchmark products with technologically comparable features right out of the box,
`
`no assembly required. That is necessarily so. The idea behind using a benchmark product with
`
`a technologically comparable feature to calculate a reasonable royalty is that one can isolate
`
`the value of the technologically comparable feature by, inter alia, subtracting the value of other
`
`features. If there is no technologically comparable feature at the outset, subtracting the value
`
`of other features would not isolate the value of the technologically comparable feature, as
`
`required for calculating a reasonable royalty based on that value.
`
`By way of example, in Apple, the Federal Circuit endorsed the use of the “Magic
`
`Trackpad” as a benchmark product for calculating a reasonable royalty for infringement of a
`
`patent that disclosed the use of finger contacts to control a touchscreen computer. 757 F.3d
`
`7
`
`Northern District of California
`
`United States District Court
`
`

`

`Case 3:20-cv-06754-WHA Document 813 Filed 06/09/23 Page 8 of 18
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`at 1316. Like the asserted claims, the trackpad translated finger contacts into computer
`
`commands, including some of the same finger contacts and computer commands that were
`
`asserted. It did not need to be separately configured to do so. Meanwhile, in i4i, the Federal
`
`Circuit endorsed the use of XMetaL as a benchmark product for calculating a reasonable
`
`royalty for infringement of a patent that disclosed an improved method for editing documents
`
`containing markup languages like XML. 598 F.3d at 853–55. Like the asserted claims, this
`
`software processed and edited documents containing markup languages like XML. Again, it
`
`did not need to be separately configured to do so.
`
`IFTTT was no such benchmark product. Unlike the asserted claims, it merely allowed
`
`users to create applets comprising chains of conditional “if” and “then” statements. Right out
`
`of the box, IFTTT did not customize, save, and invoke overlapping groups of speakers. (Nor
`
`did any published IFTTT applets customize, save, and invoke overlapping groups of speakers,
`
`for that matter.) In order to customize, save, and invoke overlapping groups of speakers, an
`
`IFTTT user would have had to create two applets comprising chains of conditional “if” and
`
`“then” statements that together customized, saved, and invoked overlapping groups of
`
`speakers. Whereas the Magic Trackpad was designed and marketed to translate finger contacts
`
`into computer commands, and XMetaL was designed and marketed to process and edit
`
`documents containing markup languages like XML, IFTTT was not designed and marketed to
`
`customize, save, and invoke overlapping groups of speakers. IFTTT was designed and
`
`marketed to “do anything!” with triggers and actions (Dkt. No. 607-3 at 1).
`
`Indeed, there is no evidence on this record that IFTTT has ever been used to customize,
`
`save, and invoke overlapping groups of speakers by anyone other than Dr. Almeroth and the
`
`Sonos team. Mr. Malackowski confirmed this on the stand and in his depositions
`
`(Tr. 1147:13–1148:7, 1153:12–15; Malackowski Jan. 2022 Dep. 219:14–17; Malackowski
`
`Aug. 2022 Dep. 132:12–13). What’s more, Dr. Almeroth’s proof of concept demonstrated that
`
`IFTTT can only customize, save, and invoke overlapping groups of speakers in a crude way,
`
`and Mr. Malackowski testified that he did not expect consumers would actually use IFTTT as a
`
`substitute for this purpose (Tr. 1215:10–17; Malackowski Jan. 2022 Dep. 219:21–25;
`
`8
`
`Northern District of California
`
`United States District Court
`
`

`

`Case 3:20-cv-06754-WHA Document 813 Filed 06/09/23 Page 9 of 18
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`Malackowski Aug. 2022 Dep. 132:13–17). Customizing, saving, and invoking overlapping
`
`groups of speakers was simply not a “feature” of IFTTT.5
`
`To further illustrate, imagine a box of loose, conventional electronic parts, such as
`
`resistors, capacitors, transistors, diodes, earphones, knobs, meters, and wires. These parts
`
`could be configured to build radio transmitters and receivers, among hundreds of other things.
`
`(Perhaps a few readers will recall making radios and other such things using the all-in-one
`
`electronic project kits that were popular some decades ago.) But no one would say that a box
`
`of loose, conventional electronic parts could be a benchmark product for calculating the value
`
`of radio technology. Radio technology is not a “feature” of a box of loose, conventional
`
`electronic parts. One could not derive the value of radio technology by factoring out the value
`
`of other features, like the damages expert in Apple derived the value of the Magic Trackpad’s
`
`translation of finger contacts into computer commands by factoring out the value of its wireless
`
`mouse functionality, for instance. Apple, 757 F.3d at 1316.
`
`IFTTT is another box of parts, only digital ones, not analog ones. Yes, it could be
`
`configured to customize, save, and invoke overlapping groups of speakers, among “millions”
`
`of other things (Tr. 920:16–21, 1141:15–22). But no, IFTTT could not qualify as a benchmark
`
`product for calculating the value of customizing, saving, and invoking overlapping groups of
`
`speakers. Customizing, saving, and invoking overlapping groups of speakers was not an as-is
`
`“feature.” One could not isolate the value of customizing, saving, and invoking overlapping
`
`groups of speakers by subtracting the value of other features.
`
`It bears repeating that IFTTT was designed and marketed to “do anything!” (Dkt.
`
`No. 607-3 at 1). Allowing this generalist third-party scripting application to be a benchmark
`
`
`5 At trial, Mr. Malackowski testified that “[w]e can see on forums that people talk about using
`IFTTT for grouping speakers” and that “on the Sonos forum there are customers[’] suggestions in
`that regard” (Tr. 1147:25–1148:1, 1216:2–3). There was nothing in his reports to support any of
`this, however. And, even if there was, the mere fact that some individuals suggested and talked
`about using IFTTT to group speakers is not evidence that these individuals actually used IFTTT to
`customize, save, and invoke overlapping groups of speakers. For what it is worth, the undersigned
`sustained Google’s objection to the admission of the only (late-produced) forum post Sonos
`sought to introduce on this point, in which an individual said he used Lutron to group speakers,
`not IFTTT (Tr. 365:16–369:10).
`
`
`9
`
`Northern District of California
`
`United States District Court
`
`

`

`Case 3:20-cv-06754-WHA Document 813 Filed 06/09/23 Page 10 of 18
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`product on account of a mere capability to be fashioned in a way that loosely approximates the
`
`claimed zone-scene technology simply “proves too much.” Sonos would open the floodgates
`
`to using IFTTT as a benchmark product for calculating a reasonable royalty in almost every
`
`patent case involving software-based technology, because almost every software-based
`
`technology could be loosely approximated using IFTTT. This is the first case in which this
`
`unusual methodology has been proposed. It should be the last.
`
`*
`
`*
`
`*
`
`“When relying on licenses to prove a reasonable royalty, alleging a loose or vague
`
`comparability between different technologies or licenses does not suffice.” LaserDynamics,
`
`Inc. v. Quanta Computer, Inc., 694 F.3d 51, 79 (Fed. Cir. 2012) (emphasis added). Indeed, the
`
`Federal Circuit has expressly warned against “us[ing] licenses with no relationship to the
`
`claimed invention to drive the royalty rate up.” Ibid. (quoting ResQNet.com, Inc. v. Lansa,
`
`Inc., 594 F.3d 860, 870 (Fed. Cir. 2010)). Otherwise, a patent owner “would be free to inflate
`
`the reasonable royalty analysis with conveniently selected licenses without an economic or
`
`other link to the technology in question.” Ibid. (quoting ResQNet.com, 594 F.3d at 872).
`
`Surely the same warning should hold when relying on products to prove a reasonable royalty,
`
`and when using products with no relationship to the claimed invention to drive the royalty rate
`
`up. Otherwise, a patent owner would be free to inflate the reasonable royalty analysis with
`
`conveniently selected products without an economic or other link to the technology in
`
`question. This is what Sonos did here. IFTTT “simply ha[d] no place in this case.”
`
`ResQNet.com, 594 F.3d at 871.
`
`True, that a benchmark product is an imperfect benchmark product, or that there exists a
`
`better benchmark product, goes to evidentiary weight, not admissibility. Apple, 757 F.3d
`
`at 1319. That IFTTT could not qualify as a benchmark product at all, however, goes to
`
`admissibility, not evidentiary weight. District judges are admonished to be “gatekeepers.” See
`
`Daubert, 509 U.S. at 596–97; Kumho Tire, 526 U.S. at 147. This gate should remain firmly
`
`closed.
`
`
`
`10
`
`Northern District of California
`
`United States District Court
`
`

`

`Case 3:20-cv-06754-WHA Document 813 Filed 06/09/23 Page 11 of 18
`
`
`
`2.
`
`
`IFTTT’S PRICE COULD NOT BE A STARTING POINT.
`
`Sonos’s damages theory based on IFTTT also failed on the economics. “[T]he trial court
`
`must carefully tie proof of damages to the claimed invention’s footprint in the market place.”
`
`ResQNet.com, 594 F.3d at 869. It was unable to do so here.
`
`In brief, Mr. Malackowski used a $1.99 minimum monthly price eventually set for
`
`IFTTT’s “Pro” subscription to establish a price that consumers ostensibly would have been
`
`willing to pay for IFTTT’s technologically comparable “feature.” With this price as a starting
`
`point, he then “apportioned down” to establish a royalty that Google ostensibly would have
`
`been willing to pay for the claimed zone-scene technology at the time of the hypothetical
`
`negotiations. But the fact that a Pro subscription has been offered for sale on the Google Play
`
`Store for a minimum monthly price of $1.99 does not mean that this price was a viable starting
`
`point for the reasonable royalty analysis. Mr. Malackowski’s starting point rested on a series
`
`of unsupported assumptions.
`
`To begin, Mr. Malackowski asked us to assume that a price eventually set for an IFTTT
`
`Pro subscription could be used to assess the value of the claimed technology at the time of the
`
`hypothetical negotiations. Yet IFTTT was entirely free at the time of the hypothetical
`
`negotiation for the ’966 patent, when Sonos contends infringement began (Malackowski
`
`Rpt. 84–88; Tr. 1127:16–1128:2, 1131:1–9). There was no Pro subscription at that time.
`
`IFTTT did not roll out its add-on subscription plans until September 2020 — ten months after
`
`the hypothetical negotiation for the ’966 patent in November 2019, and two months before the
`
`hypothetical negotiation for the ’885 patent in November 2020 (Malackowski Rpt. 84;
`
`Tr. 1129:25–1130:2, 1204:24–25). According to Mr. Malackowski, Sonos and Google would
`
`have understood that IFTTT could not keep all of its offerings free forever, and the parties
`
`would have therefore negotiated a reasonable royalty at both hypothetical negotiations drawing
`
`upon on the minimum monthly price later set for a Pro subscription (Malackowski Rpt. 87;
`
`Tr. 1130:10–1131:2). But many applications do remain completely free forever. This is not
`
`something that Mr. Malackowski was qualified to speculate on, and there was nothing in the
`
`record to support his ipse dixit.
`
`11
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`Northern District of California
`
`United States District Court
`
`

`

`Case 3:20-cv-06754-WHA Document 813 Filed 06/09/23 Page 12 of 18
`
`
`
`Next, Mr. Malackowski asked us to assume that a $1.99 minimum monthly price
`
`eventually set for an IFTTT Pro subscription could be used to establish the price that
`
`consumers would have been willing to pay for the claimed technology.6 That is because a Pro
`
`subscription allowed for the use of multiple applets, and Dr. Almeroth demonstrated that
`
`consumers would need multiple applets to approximate the claimed technology (Malackowski
`
`Rpt. 87–88; Tr. 1125:23–1126:14, 1127:18–1128:8). Recognizing that a Pro subscription has
`
`allowed for the use of twenty applets, and that Dr. Almeroth demonstrated consumers would
`
`need at least two applets to approximate the claimed technology, Mr. Malackowski ultimately
`
`“apportioned down” this $1.99 starting price by 90% (Malackowski Rpt. 88; Tr. 1131:12–16).
`
`Yet “[b]eginning from a fundamentally flawed premise and adjusting it based on legitimate
`
`considerations specific to the facts of the case nevertheless results in a fundamentally flawed
`
`conclusion.” Uniloc USA, Inc. v. Microsoft Corp., 632 F.3d 1292, 1317 (Fed. Cir. 2011).
`
`Because IFTTT could be used to do an enormous number of things with twenty applets and
`
`two applets alike, the price of a Pro subscription had no relationship to the price that
`
`consumers would have been willing to pay for the technologically comparable “feature” in the
`
`first place. This price had no relationship to the price that consumers would have been willing
`
`to pay for one (potential) configuration of applets that could (crudely) customize, save, and
`
`invoke overlapping groups of speakers.
`
`Truth be told, the $1.99 minimum monthly price was not even the price set by IFTTT for
`
`the use of twenty applets. According to Mr. Malackowski’s report, the Pro subscription that
`
`IFTTT offered for a minimum monthly price of $1.99 allowed users to create unlimited applets
`
`(Malackowski Rpt. 84–85). It was not until IFTTT released an updated Pro subscription in
`
`November 2021 — well after the hypothetical negotiations — that this subscription plan
`
`
`6 For context, the Pro subscription was offered for a “pay-what-you-want,” minimum monthly fee
`of $1.99 from September 2020 through October 2020 (Malackowski Rpt. 85). At the time of the
`hypothetical negotiation for the ’885 patent in November 2020 through November 2021, the Pro
`subscription was priced at $3.99 per month. Between November 2021 and November 2022, it was
`apparently priced even higher. As of November 2022, the Pro subscription has been priced at
`$2.50 per month. According to Mr. Malackowski, to be conservative, he based his calculation on
`the lowest fee IFTTT offered for the Pro subscription (Malackowski Rpt. 87).
`12
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`Northern District of California
`
`United States District Court
`
`

`

`Case 3:20-cv-06754-WHA Document 813 Filed 06/09/23 Page 13 of 18
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`28
`
`became limited to twenty apple