Opinion
2020-1950 2020-2028
09-27-2021
TRUSTID, INC., Appellant v. NEXT CALLER, INC., Cross-Appellant
Byron Leroy Pickard, Sterne Kessler Goldstein & Fox, PLLC, Washington, DC, argued for appellant. Also represented by Richard M. Bemben, Michael D. Specht. Sarah Chapin Columbia, McDermott, Will & Emery LLP, Boston, MA, argued for cross-appellant. Also represented by Ian Barnett Brooks, Nicole M. Jantzi, Paul Michael Schoenhard, Washington, DC.
This disposition is nonprecedential.
Appeals from the United States Patent and Trademark Office, Patent Trial and Appeal Board in No. IPR2019-00039.
Byron Leroy Pickard, Sterne Kessler Goldstein & Fox, PLLC, Washington, DC, argued for appellant. Also represented by Richard M. Bemben, Michael D. Specht.
Sarah Chapin Columbia, McDermott, Will & Emery LLP, Boston, MA, argued for cross-appellant. Also represented by Ian Barnett Brooks, Nicole M. Jantzi, Paul Michael Schoenhard, Washington, DC.
Before Reyna, Schall, and Stoll, Circuit Judges.
SCHALL, CIRCUIT JUDGE.
TRUSTID, Inc. ("TRUSTID"), the owner of U.S. Patent No. 9, 001, 985 ("the '985 patent"), appeals a final written decision of the Patent Trial and Appeal Board ("Board") determining that certain claims of the '985 patent were shown to be unpatentable. Next Caller Inc. v. TRUSTID, Inc., No. IPR2019-00039 (P.T.A.B. Feb. 24, 2020), Paper No. 67, Corrected Non-Confidential Joint Appendix ("J.A.") 1-92 ("Final Written Decision"). Next Caller, Inc. ("Next Caller") cross-appeals the Board's determination that other claims of the '985 patent were not shown to be unpatentable. We affirm-in-part, vacate-in-part, and remand. In particular, we affirm the Board's decision finding claims 1-7, 12-14, 16-18, and 22 of the '985 patent unpatentable. However, because the Board did not adequately explain the reasoning for its non-obviousness determination as to claims 8-11, 19, and 20 of the '985 patent, we vacate the Board's decision with respect to those claims and remand for further proceedings.
Background
I.
The '985 patent is directed to preventing call spoofing by discovering and reporting the trustworthiness and credibility of calling party number information associated with an incoming call. See '985 patent Abstract, col. 11. 23-col. 21. 9. Claim 1 of the '985 patent recites:
1. A method of determining a source origin confidence metric of a calling party number or billing number associated with an incoming call to a called party telephonic device from a calling party telephonic device, comprising:
receiving by an electronic system associated with the called party telephonic device the calling party
number or billing number, wherein the electronic system receives the calling party number or billing number from the called party telephonic device;
after receiving the calling party number or billing number and before the incoming call is answered, gathering by the electronic system associated with the called party telephonic device operational status information associated with the calling party number or billing number, and
determining by the electronic system associated with the called party telephonic device the source origin confidence metric for the calling party number or billing number.Id. at col. 15 ll. 2-19.
Claim 4 depends from claim 1 and further recites "determining by the electronic system associated with the called party telephonic device whether the format of the calling party number or billing number is valid." Id. at col. 15 ll. 29-32. Claim 8 also depends from claim 1 and recites "adjusting . . . the source origin confidence metric based on personal risk factors of an entity associated with the calling party number or billing number." Id. at col. 15 ll. 46-50.
II.
Next Caller petitioned for inter partes review ("IPR") of all of the claims of the '985 patent after TRUSTID brought suit against it for infringement. The petition presented four grounds of invalidity, three of which are at issue in this appeal. Ground 2 challenged claims 1-7, 12-18, 21, and 22 as obvious over the combination of U.S. Patent Publication No. 2007/0201625 ("Martin") in view of U.S. Patent Publication No. 2007/0081648 ("Abramson"). Grounds 3 and 4 challenged claims 8-11, 19, and 20 as obvious over a combination of Martin, and then Martin with Abramson, both in view of U.S. Patent No. 7, 912, 192 ("Kealy").
The Board instituted IPR and in due course issued the Final Written Decision. In its decision, the Board determined that claims 1-7, 12-14, 16-18, and 22 were un-patentable because they would have been obvious in view of the combination of Martin and Abramson. J.A. 90. In its analysis for the "operational status information" claim element of independent claims 1 and 13, the Board discussed Abramson's teaching of "gathering and using in its checks information that is the same as that gathered and used in the '985 [p]atent." Id. at 55, 65. When addressing the "source origin confidence metric" element of independent claims 1 and 13, the Board relied on an embodiment set forth in Abramson ¶ 74 that was not explicitly set forth in Next Caller's petition. Compare id. at 41-42, with id. at 192-95. Finally, the Board also found that Martin taught the limitations of dependent claim 4, citing Martin's teaching of "using the call source identification information transmitted by the telephone system (typically between the first and second ring of a call) to make decisions as to how to process the call and process[ing] valid calls." Id. at 70 (citing Martin ¶ 6).
The Board held claims 15 and 21 had not been shown to be unpatentable. Those claims are not at issue on appeal.
As noted, the Board determined that Next Caller had not shown claims 8-11, 19, and 20 to be unpatentable. The Board rejected Next Caller's argument that Kealy's teaching of reducing a trust rating based on complaints renders obvious the claim limitation "adjusting . . . the source origin confidence metric based on personal risk factors" of claim 8. Id. at 87. The Board stated that Next Caller "[did] not explain how Kealy's complaints are 'personal risk factors'" and did "not show[ ] why or how a person having ordinary skill in the art would have modified the Martin and Abramson combination with Kealy to perform the further step of 'adjusting . . . the source origin confidence metric based on personal risk factors.'" Id. at 87-88.
TRUSTID sought rehearing of the Board's determination with respect to claims 1-7, 12-14, 16-18, and 22, arguing that the Board erroneously relied on Abramson as teaching the claimed "operational status information" when Next Caller had relied on Martin for this claim element. J.A. 952-55. The Board denied rehearing, concluding that the petition "identifies disclosures of both Martin and Abramson relied upon for the elements of the claims, including 'operational status information.'" J.A. 98-99.
TRUSTID and Next Caller timely appealed and cross-appealed, respectively. We have jurisdiction under 28 U.S.C. § 1295(a)(4)(A).
Discussion
I.
We review the Board's legal determinations, including obviousness, de novo, and its underlying factual findings for substantial evidence. Belden Inc. v. Berk-Tek LLC, 805 F.3d 1064, 1073 (Fed. Cir. 2015). A finding is supported by substantial evidence if a reasonable mind might accept the evidence as adequate to support the finding. Consol. Edison Co. v. NLRB, 305 U.S. 197, 229 (1938).
We review the Board's decision for compliance with the Administrative Procedure Act ("APA") de novo. We must "hold unlawful and set aside agency action, findings, and conclusions found to be . . . arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law," "without observance of procedure required by law," or "unsupported by substantial evidence." 5 U.S.C. § 706. Whether a ground the Board relied upon was new, requiring a new opportunity to respond, is a question of law that we review de novo. In re NuVasive, Inc., 841 F.3d 966, 970 (Fed. Cir. 2016).
II.
We begin with TRUSTID's appeal. TRUSTID makes three arguments, which we address in turn.
A.
TRUSTID's first argument is that the Board violated the APA in the Final Written Decision by relying on Abramson to teach the claimed "operational status information" element of independent claims 1 and 13, when Next Caller's petition, the Board's institution decision, and a decision of the Board denying rehearing of the institution decision had relied solely on Martin for this claim element. TRUSTID Br. 9-14, 25-29. In making this argument, TRUSTID relies on EmeraChem Holdings, LLC v. Volkswagen Group of America, Inc., 859 F.3d 1341 (Fed. Cir. 2017), where our court held the Board's final written decision violated the APA because it relied on a different reference than that relied upon in the petition or the institution decision. TRUSTID separately argues that there is no evidence in the record for the Board to find that Martin teaches or makes obvious the claimed "operational status information." TRUSTID Br. 28-29.
In response, Next Caller argues that the Board found that Martin alone teaches "operational status information," that that finding is supported by substantial evidence, and that the Board's finding that Abramson also taught that element was "nothing more than a permissible identification of an alternative basis." Next Caller Br. 22, 34-37. We agree.
In the Final Written Decision, in its analysis for the "operational status information" element of claim 1, the Board began by noting the argument Next Caller made in the petition, including Next Caller's citations to Martin and to a declaration of Next Caller's expert, Mr. James Geier ("Geier Decl."):
Petitioner points to Martin's teaching of central monitoring station 100 processing status bits 210-330 if the aforementioned match is found between the received caller ID information 75 and information 200. Pet. 31-34 (citing, e.g., [Martin] ¶¶ 6, 27, Figs. 1, 2, 5; [Geier Decl.] ¶¶ 92-97). Petitioner also points to Martin's teaching of exemplary status bits 210-330. Pet. 13-14, 32 n.10 (citing, e.g., [Martin] ¶¶ 21-23, 28; [Geier Decl.] ¶¶ 50-51, 98).
J.A. 53. The Board then stated:
We agree with Petitioner and credit and give weight to the testimony of Mr. Geier because Petitioner's contentions and Mr. Geier's testimony are consistent with the evidence of record. For instance, Martin teaches if the received caller ID information 75 matches caller ID information 200 stored in memory 140, processing circuits 130 read status bits 210-330. [Martin] ¶¶ 6, 27, Fig. 5.J.A. 54. We agree with Next Caller that this constitutes a finding by the Board that Martin teaches the claimed "operational status information."
We also agree with Next Caller that this finding is supported by substantial evidence. Martin explains that "the processing circuits 120 read status bits 210-330 associated with the caller ID information 200." Martin ¶ 27; see also id. at ¶¶ 6, 21-23, 28, Figs. 1, 2, 5. Further, Mr. Geier explained that Martin's "status information comprises various forms of information as illustrated in Figure 2," which shows status bits 210-330. J.A. 1304-05 ¶ 97. And according to Mr. Geier, a person of ordinary skill in the art "would have understood that Martin's status information is composed of 'operational status information.'" J.A. 1305 ¶ 98; see also J.A. 1282 ¶¶ 50-51; J.A. 1302 ¶ 92; J.A. 1304 ¶¶ 95-97.
It is true, however, that the Board then pointed to Abramson's teaching of "gathering and using in its checks information that is the same as that gathered and used in the '985 patent." J.A. 55. Its reliance on Abramson as a supplemental teaching of this claim element for claim 1, in view of its prior finding that Martin teaches the element, was, at most, harmless error. See Yeda Rsch. v. Mylan Pharmas. Inc., 906 F.3d 1031, 1042 (Fed. Cir. 2018) (concluding that the Board's reliance on a reference, to the extent reliance on that reference was improper, was a harmless error because substantial evidence otherwise supported the Board's conclusion).
We reach the same conclusion for independent claim 13. Although the Board's discussion of the "operational status information" element for that claim pertained primarily to Abramson, the Board noted that Next Caller "relie[d] on the same teachings discussed with respect to claim 1." J.A. 65. The Board then referred back to its claim 1 analysis (e.g., "see supra III.E.4.c."), which concluded that Martin taught this element, before additionally finding that Abramson taught the claimed "operational status element." Id. at 65-66. As with claim 1, the Board's reliance on Abramson as a supplemental teaching of this claim element for claim 13 was, at most, harmless error.
B.
TRUSTID's second argument is that the Board violated the APA when it relied on a different embodiment of Abramson to teach the claimed "source origin confidence metric" than Next Caller had relied upon in its petition. Specifically, TRUSTID contends that it did not have notice and an opportunity to respond to the theory that Abram-son's processing of tasks 501 through 503, as illustrated in Abramson's Figure 5 and described in ¶ 74, generates a number within a range of zero to three that corresponds to the claimed "source origin confidence metric." TRUSTID Br. 30-31; TRUSTID's Resp. and Reply Br. 7-9.
Next Caller responds that TRUSTID had the opportunity to respond, and did respond, to Next Caller's position, which was consistent with the Board's findings on this claim element. Next Caller Br. 38-41.
"The notice and opportunity to be heard provisions of the APA have been applied 'to mean that an agency may not change theories in midstream without giving respondents reasonable notice of the change' and 'the opportunity to present argument under the new theory.'" Genzyme Therapeutic Prods. L.P. v. Biomarin Pharm. Inc., 825 F.3d 1360, 1366 (Fed. Cir. 2016) (quoting Belden, 805 F.3d at 1080). We agree with Next Caller that there was no APA violation here because TRUSTID received adequate notice and an opportunity to be heard.
In the petition, Next Caller pointed to Abramson's teaching of performing three tests (tasks 501, 502, and 503) that pertain to characteristics of the calling party's terminal. J.A. 192-94 & nn. 6, 7 (citing, inter alia, Abramson ¶¶ 66-72, 75-82, and Fig. 5; J.A. 1299 (Geier Decl.) ¶¶ 76-83). If all three tests are passed, then call privileges are granted; if any test is not passed, the call is terminated. Id. at 194-95 n.7. Next Caller's petition argued that this teaches a "source origin confidence metric" if the term is broad enough to cover a binary result. In the alternative, Next Caller stated:
A [person of skill in the art] would have found [sic] obvious to modify Abramson to assess each characteristic and assign a probabilistic score-if, e.g., only two of three tasks resulted in a "yes." Under such circumstances, Abramson could advantageously gain greater flexibility in evaluating callers-assigning weight to certain characteristics and determining validity based on a surpassed threshold. These modifications would have been straightforward to a [person of skill in the art], as they would require minor changes to the
programming of Martin's and Abramson's processors, and would have yielded predictable results. [Geier Decl.] ¶¶ 84-86.J.A. 194-95 n.7. In its response to the petition, TRUSTID disputed that Abramson taught or rendered obvious the claimed "metric." J.A. 384-85, 389-92. In reply, Next Caller explained that "Abramson at minimum renders obvious the claimed metric," cited ¶¶ 65-73 of Abramson, and rebutted TRUSTID's arguments with reference to Abram-son's ¶ 74. J.A. 509. Next Caller stated:
Abramson suggests performing a calculation using known information to decide whether to grant privileges. . . . [A person of ordinary skill in the art] would have understood one way to address the situation where privileges could still be granted despite not having all "yeses" would have been to weigh/score the results and decide whether to grant access based on the score. Such a score could be determined in the same manner as described above.Id. at 509-10 (footnote omitted); see also J.A. 506-07 (discussing Martin and explaining that "[o]ne obvious manner of making that determination would have been to weigh each of the status bits/bytes and then output a metric (e.g., number in a range) based on assigned weights of the status bits/bytes that could be used to determine whether the call is valid." (footnote omitted)).
In its sur-reply, TRUSTID cited ¶ 74 of Abramson to show that "because Abramson already discloses a mechanism for handling the situation where one of the tasks returns an unexpected result," a person of ordinary skill in the art would not have modified Abramson to generate a metric based on considering two of three tasks. J.A. 627. TRUSTID pointed to cross-examination testimony of Mr. Geier, which it contended showed that Mr. Geier's opinions regarding weighing status bits/bytes and output-ting a score/metric were "general." J.A. 626.
In the Final Written Decision, the Board construed "source origin confidence metric" to exclude binary measurements. J.A. 29-30, 40. Citing to ¶ 74 of Abramson, the Board stated:
Also consistent with Petitioner's contentions and Mr. Geier's testimony, Abramson further describes an alternative embodiment in which "exchange 205 might still grant privileges, or a limited set of privileges, even if at least one of the results is unexpected, indeterminate, or unknown-for example, the received telephone type is unexpected but the received signaling protocol is as expected."J.A. 41 (quoting Abramson ¶ 74) . The Board continued:
Abramson discloses three separate tasks, i.e., tasks 501, 502, and 503, each [of] which may result in "yes" or "no." [Abramson] ¶¶ 66-74, Fig. 5. Processing of tasks 501 through 503 illustrated in the flowchart depicted in Figure 5 results in a number that is a measurement of characteristics that agreed with what was expected. Id. That number is within a range of "0" through "3." Id.J.A. 41-42. The Board credited Mr. Geier's testimony that a person of ordinary skill in the art "would have found [sic] obvious to modify Abramson to access each characteristic and assign a probabilistic score-if, e.g., only two of three tasks resulted in a 'yes.'" J.A. 42 (emphasis omitted) (quoting J.A. 1300 (Geier Decl.) ¶ 86). While TRUSTID's expert, Dr. Leonard J. Forys, disagreed with Mr. Geier, the Board "credit[ed] and [gave] significant weight to the testimony of Mr. Geier over that of Dr. Forys" because it found Mr. Geier's testimony to be "consistent with the evidence of record," specifically, Abramson ¶ 74 itself. J.A. 43.
We disagree with TRUSTID that it did not have notice and an opportunity to respond to the theory the Board relied upon. Although presented in different terms, the Board's finding that Abramson's tasks 501 through 503 could be processed to result in a number that is a measurement of characteristics and that is within a range of "0" through "3" is not an altogether different theory from that presented in the petition or in Next Caller's reply. See Ar-threx, Inc. v. Smith & Nephew, Inc., 935 F.3d 1319, 1326-28 (Fed. Cir. 2019) ("[T]he mere fact that the Board did not use the exact language of the petition in the final written decision does not mean it changed theories in a manner inconsistent with the APA and our case law."). As noted, the petition indicated that a person of ordinary skill in the art "would have found [sic] obvious to modify Abramson to assess each characteristic and assign a probabilistic score- if, e.g., only two of three tasks resulted in a 'yes.'" J.A. 194- 95 n.7. In reaching its conclusion, moreover, the Board cited the same portions of Abramson cited in the petition and in Next Caller's reply, to which TRUSTID was given the opportunity to respond, and did in fact respond. We therefore reject TRUSTID's argument that the Board's Final Written Decision violated its procedural rights with respect to the "source origin confidence metric" claim element of independent claims 1 and 13.
C.
Third, TRUSTID argues that the Board's finding that Martin teaches the limitations of dependent claim 4 is not supported by substantial evidence. Claim 4 recites "determining by the electronic system associated with the called party telephonic device whether the format of the calling party number or billing number is valid." '985 patent col. 15 ll. 29-32.
TRUSTID contends that Martin compares information about the incoming call, such as its caller ID information, to a list of numbers stored in its memory, and processes a call only when there is a match. According to TRUSTID, however, "[c]hecking to see whether two numbers match . . . is not determining whether a telephone number is properly formatted." TRUSTID Resp. and Reply Br. 11; see also TRUSTID Br. 23-24, 31-33. TRUSTID contends that Mr. Geier conceded that if there is no match "you wouldn't know something about formatting." TRUSTID Br. 33 (quoting J.A. 4649).
Next Caller responds that the Board's findings with respect to claim 4 are supported by substantial evidence, including Martin ¶¶ 6 and 27, and the testimony of Mr. Geier. Next Caller Br. 41-43. Next Caller points to Mr. Geier's declaration testimony that a person of ordinary skill in the art "would have understood that if there was no match, one reason for that 'no match' is that the format of the call source identification is not valid. Similarly, a match would indicate that the format is valid." Id. at 42 (quoting J.A. 1313 (Geier Decl.) ¶ 123). Alternately, Next Caller asserts, Mr. Geier explained that a person of ordinary skill would have found the limitation obvious. Id.
We agree with Next Caller. Martin discloses checking the validity of an incoming call, and processing only valid calls:
The central monitoring station will process the call by checking if the call source identification information matches call source identification information stored in memory, and it will check status data that may be associated with the call source identification information, also stored in memory, to determine if the call should be disconnected, transferred, or connected to the receiver. Since the receiver only processes valid (legitimate) local alarm system reports, the lines are less likely to be tied up by invalid alarm calls and the efficiency of the alarm answering process is increased.
. . . [T]he processing circuits 130 look for a match between the received caller ID information 75 and the caller ID information 200 stored in memory 140. If a match is not found, the processing circuits 130 cause the switching circuits 120 to connect the call to the invalid alarm call station 195. . . . If there is a match, the processing circuits 120 read the status bits 210-330 associated with the caller ID information 200. If the processing circuits 120 determine the call is a valid alarm call, the processing circuits 130 cause the switching circuits 120 to connect the call to the receiver 150 and transmit the status information 145.
Martin ¶¶ 6, 27. Mr. Geier explained that a person of skill in the art "would have understood that if there was no match, one reason for that 'no match' is that the format of the call source identification is not valid." J.A. 1313 (Geier Decl.) ¶ 123. That a "no match" determination does not necessarily provide information about whether the format is valid does not mean Martin does not teach performing the claimed "determining" when it does find a match. As Mr. Geier explained, "a match would indicate that the format is valid." Id.
As we have explained, "a prior art product that sometimes, but not always, embodies a claimed method nonetheless teaches that aspect of the invention." Hewlett-Packard Co. v. Mustek Sys., Inc., 340 F.3d 1314, 1326 (Fed. Cir. 2003).
Accordingly, we agree with Next Caller that the Board's finding that Martin teaches claim 4 is supported by substantial evidence.
III.
We turn next to Next Caller's cross-appeal. Next Caller argues that the Board erred when it determined that claims 8-11, 19, and 20 of the '985 patent had not been shown to be unpatentable. Next Caller takes issue with the Board's rationale with respect to those claims, urging that it was insufficient under In re NuVasive, Inc., 842 F.3d 1376 (Fed. Cir. 2016), and that the Board disregarded relevant findings regarding Kealy that had been made by the patent examiner during examination of the application that issued as the '985 patent. Next Caller Br. 24-31. Next Caller contends that the Board erred when it failed to defer to the examiner's previous findings and because it did not address the petition's evidence or articulate any rationale for rejecting it. Id.
TRUSTID responds that Next Caller waived or forfeited its arguments regarding Kealy by failing to request that the Board defer to the examiner's findings. TRUSTID Resp. and Reply Br. 23-26. TRUSTID also argues that "there is no legal authority requiring the Board to defer to original prosecution in an IPR." Id. at 26. Further, TRUSTID contends that the Board's reasoning was sufficiently articulated and supported by substantial evidence. Id. at 29-33.
During examination of the application that ultimately issued as the '985 patent, the examiner issued an Office Action in which the application claims that correspond with issued claims 8-10, 19, and 20 were rejected as obvious over a combination of references including Kealy. The Office Action stated:
[U.S. Patent No. 5, 963, 625 to Kawecki et al. ("Kawecki")] teaches all subject matters as claimed above, except for the features of adjusting the source origin confidence metric including one or more fraud score, risk score, etc.; storing and retrieving them from an external database. However [Kealy] teaches trust ratings embedded in certificates of calling device [sic], and a method of managing trust ratings based on fraud and risk factors
such as accumulated complaints, etc. When a quantity of accumulated complaints exceeded a given threshold or no complaints in a given time, a trust rate of a given calling device is adjusted such as reduced or increased (col. 14, line 24 through col. 15, line 16).
Therefore, it would have been obvious to one of ordinary skill in the art at the time the invention was made to incorporate the use of the features of adjusting the source origin confidence metric including one or more fraud score, risk score, etc.; storing and retrieving them from an external database, as taught by Kealy, into [sic] view of Kawecki in order to certify the calling device for calls from unwanted "junk" phone calls.J.A. 1120-21.
Next Caller provided a copy of the office action with its petition. J.A. 174. Next Caller also quoted the examiner's statement of what Kealy teaches, including the examiner's citation to column 14, line 24, through column 15, line 16, of Kealy. Id. In its claim charts and its accompanying note, Next Caller pointed to Kealy's teaching of "reducing a trust rating of a calling party based on the number of complaints received," J.A. 210 n. 22, and cited Kealy itself and Mr. Geier's explanation of Kealy. See J.A. 209-12 & n. 22 (citing Kealy at Abstract, col. 3 ll. 28-36, col. 14 ll. 13-20, and col. 14 l. 58-col. 15 l. 6; Geier Decl. ¶¶ 147-51).
In addition, in the petition, Next Caller asserted that Kealy's teachings of trust ratings would have been beneficial "to further bolster Martin's [and Abramson's] incoming call processing system[s]" and "particularly beneficial to Martin's assessment of VoIP calls," and such a motivation would have had the benefit of "reduc[ing] costs associated" with such calls. Id. at 187, 188. The petition stated:
A [person of ordinary skill in the art] would have found it obvious and straightforward to use Kealy's advantageous teachings of trust ratings to further assess the legitimacy of an alarm call in Martin's system. A [person of ordinary skill in the art] would have understood how to make the modification to Martin's program, and such a modification would have been routine and predictable. Indeed, Martin's status information is readily adaptable and could maintain trust ratings as part of its "additional information." [Martin ¶ 23], Fig. 2[, ] [Geier Decl.] ¶ 67.J.A. 188; see also id. at 188-89 (citing Geier Decl. ¶ 69).
As noted, in the Final Written Decision, the Board rejected Next Caller's argument that Kealy's teaching of reducing the trust rating based on complaints rendered obvious the claim limitation of adjusting the source origin confidence metric based on personal risk factors, stating that Next Caller "[did] not explain how Kealy's complaints are 'personal risk factors'" and did "not show[ ] why or how a person having ordinary skill in the art would have modified the Martin and Abramson combination with Kealy to perform the further step of 'adjusting . . . the source origin confidence metric based on personal risk factors.'" J.A. 87-88. The Board made this statement after first reciting certain of Next Caller's arguments, and then block quoting column 14, line 58, through column 15, line 6 of Kealy.
We agree with TRUSTID that there is no legal basis for the proposition the Board must defer to the examiner's previous findings in deciding the merits of an IPR. Indeed, "the Supreme Court has characterized the 'congressional objective' of the IPR process as 'giving the Patent Office significant power to revisit and revise earlier patent grants." Skky, Inc. v. MindGeek, s.a.r.l., 859 F.3d 1014, 1021 (Fed. Cir. 2017) (quoting Cuozzo Speed Techs., LLC v. Lee, 136 S.Ct. 2131, 2139-40 (2016)). That said, we agree with Next Caller that the Board's analysis with respect to Kealy's teachings and motivation to combine were insufficient. "[T]he Board is obligated to provide an administrative record showing the evidence on which the findings are based, accompanied by the agency's reasoning in reaching its conclusions." Alacritech, Inc. v. Intel Corp., 966 F.3d 1367, 1370 (Fed. Cir. 2020) (quoting TQ Delta, LLC v. Cisco Sys., Inc., 942 F.3d 1352, 1358 (Fed. Cir. 2019)). "We do not require 'perfect explanations,' and 'we will uphold a decision of less than ideal clarity if the agency's path may reasonably be discerned.'" Id. at 1370-71 (quoting NuVasive, 842 F.3d at 1382-83). "We do, however, require that the Board's own explanation be sufficient 'for us to see that the agency has done its job.'" Id. (quoting NuVasive, 842 F.3d at 1383). Here, the Board merely partially reiterated and summarily rejected Next Caller's arguments without explanation. This is not sufficient under the APA and our precedent. NuVasive, 842 F.3d at 1383 (explaining that the Board cannot "summarize and reject arguments without explaining why [it] accepts the prevailing argument."). Thus, while there is no legal basis to require that the Board have deferred to the examiner's reasoning, the Board did need to have provided its own reasoning.
Conclusion
For the reasons set forth above, we affirm the Board's decision finding claims 1-7, 12-14, 16-18, 21, and 22 of the '985 patent unpatentable. We vacate the Board's non-obviousness determination as to claims 8-11, 19, and 20 of the '985 patent and remand for further proceedings consistent with this opinion.
AFFIRMED-IN-PART, VACATED-IN-PART, AND REMANDED.
Costs
No costs.