Debra Pratt v. Wisconsin Aluminum Foundry

Opinion

In the

United States Court of Appeals For the Seventh Circuit ____________________ No. 24-1901 DEBRA PRATT, Plaintiff-Appellant, v.

WISCONSIN ALUMINUM FOUNDRY, Defendant-Appellee. ____________________

Appeal from the United States District Court for the Eastern District of Wisconsin. No. 1:22-cv-00568 — William C. Griesbach, Judge. ____________________

ARGUED JANUARY 28, 2025 — DECIDED JULY 23, 2026 ____________________

Before HAMILTON, KIRSCH, and MALDONADO, Circuit Judges. MALDONADO, Circuit Judge. After reporting claims of dis- crimination and harassment at the Wisconsin Aluminum Foundry (WAF), human resources manager Debra Pratt found herself under attack. Other managers targeted her, she received a poor performance review, and after she com- plained about discrimination and retaliation, she was fired. Pratt sued, alleging sex discrimination, pay discrimination, 2 No. 24-1901

and retaliation. WAF said it fired Pratt because of the perfor- mance review as well as a lack of trust in Pratt’s department and confidentiality issues. The district court granted sum- mary judgment in favor of WAF on all of Pratt’s claims. We reverse in part. Pratt presented sufficient evidence of sex discrimination and retaliation under Title VII such that a reasonable jury could find in her favor on those claims. To be sure, WAF presented an alternative narrative that Pratt was fired for poor performance, but summary judgment is not the time to decide which narrative to believe. The time for weigh- ing competing evidence is trial, and the job is for a jury not a judge. See Wallace v. SMC Pneumatics, Inc., 103 F.3d 1394, 1396 (7th Cir. 1997) (“[C]ourts should be careful in a discrimination case as in any case not to grant summary judgment if there is an issue of material fact that is genuinely contestable, which an issue of intent often though not always will be.”); Johnson v. Advoc. Health & Hosps. Corp., 892 F.3d 887, 893 (7th Cir. 2018) (“As we have said many times, summary judgment cannot be used to resolve swearing contests between litigants.” (citation omitted)). We agree with the district court, however, that WAF is entitled to summary judgment on Pratt’s pay discrim- ination claim. Accordingly, we remand for trial on Pratt’s sex discrimination and retaliation claims. BACKGROUND The following facts are undisputed, except where other- wise noted, and viewed in the light most favorable to Pratt as the non-movant. Gaddis v. DeMattei, 30 F.4th 625, 628, 630 (7th Cir. 2022). No. 24-1901 3

I. Factual Background WAF is a family-owned aluminum foundry in Manitowoc, Wisconsin. Pratt began working for WAF in 2016 in the hu- man resources department. In 2017, she was promoted to hu- man resources manager. Her new responsibilities included developing and implementing personnel policies, maintain- ing personnel records, and responding to and investigating employee complaints. The parties dispute the scope of her responsibilities for en- vironmental health and safety, which bears on her pay dis- crimination claim. Pratt’s predecessor performed those du- ties, but WAF contends that Pratt did not when she was pro- moted. A few months after Pratt was promoted, WAF hired Emery Coonen to be its Environmental Health and Safety Manager, and he took over these responsibilities. Coonen was hired at an annual salary of $85,328. Pratt’s starting salary as HR manager was $65,016. Pratt performed well in her first year as HR manager and received a bonus and a raise at the end of 2017, bringing her annual salary to $69,567. But she experienced some early set- backs. In early 2018, she committed errors on a benefits spreadsheet and did not catch a rate change for insurance pre- miums. She also filled a union position with an external can- didate, which WAF contends violated its collective bargaining agreement. Pratt also began to ruffle some feathers within the com- pany by reporting concerns of discrimination and harassment to her supervisor, Ben Jacobs, WAF’s Senior Vice President and Chief Product Officer. She believed her predecessor, who had been in the role for over thirty years, did not do enough 4 No. 24-1901

to investigate employee complaints. So, she took it upon her- self to ensure that, under her direction, the department would take complaints seriously and report them to WAF leadership. Between 2017 and 2018, Pratt reported multiple employee complaints of harassment and discrimination to Jacobs and other managers. Employees complained to Pratt about dis- crimination based on sex, race, ethnicity, and disability. She also shared with Jacobs her personal experiences as a female manager at WAF. Eugene Boyd, WAF’s Vice President of Operations, was the subject of many of these complaints. Pratt not only re- ported other employees’ complaints about Boyd, but her own as well. For instance, in conversation with Pratt, Boyd referred to another female manager as a “bitch” and told Pratt he could “rile her up.” Boyd also treated Pratt and her employees in the HR department—composed wholly of women—as secretar- ies, asking them to do menial tasks not related to their job du- ties. One of Pratt’s reports to Jacobs involved Boyd discussing a female employee’s “ass” at a meeting. During that meeting, Boyd’s female subordinate, Lili Goehring, remarked that she fell down the stairs and bruised her buttocks. The parties dis- pute what happened next, but they agree that Pratt later told Jacobs that, after Goehring’s remarks, Boyd told Goehring to get up on the table and show him her “ass.” One meeting at- tendee told Pratt that Boyd also stuck his thumb up on the table and told Goehring to “sit on it and he would spin her.” Another attendee informed Pratt that while the jokes did not seem to bother Goehring, the innuendo made them uncom- fortable. No. 24-1901 5

After learning about this incident, Pratt prepared a written report for Jacobs summarizing her conversations with the wit- nesses who approached her. The report included her “find- ings,” in which she wrote that the “sexual harassment” be- tween Boyd and Goehring was her “biggest concern.” She noted that the interaction “appear[ed] to be of a joking na- ture,” but it was “still unprofessional and against the law.” She advised that any witnesses to the interaction would “have the ability to file a sexual harassment complaint with the com- pany as well as the EEOC” and that it “would be a difficult case to defend.” Pratt sent the report to Jacobs a few weeks later. She did not know, however, that Jacobs already had a copy of it. An- other manager at WAF told Jacobs that he found the report on a copy machine near where Pratt worked. Jacobs was con- cerned that a confidential report was apparently left out in the open, so WAF engaged its outside counsel to investigate. The investigator spoke to Pratt, who admitted preparing the re- port but was surprised that someone else had found it. Pratt was later reprimanded for breaching confidentiality. Jacobs was also concerned about the way Pratt conducted her inves- tigation before compiling the report, and she was given addi- tional training on conducting and documenting HR investiga- tions. Neither Boyd nor Goehring were disciplined for their part, although Jacobs testified they were given additional “training.” The problems with Boyd were not isolated incidents. The same female manager Boyd called a “bitch” complained to Pratt about Boyd barring her from necessary work meetings. Pratt reported another incident to Jacobs in which Boyd ob- scenely flipped off a female receptionist with his middle fin- 6 No. 24-1901

ger after she asked him to move his car from a visitor parking lot. Pratt’s reports to Jacobs also involved complaints of race discrimination, including a number complaining of Tom Culp, a white shift leader at WAF. Based on complaints she received, Pratt grew concerned that Culp treated his minority employees differently than his white employees and reported multiple incidents to Jacobs and to Boyd, Culp’s supervisor. Finally, Pratt also told Jacobs about her own experiences with discrimination and harassment at WAF. She reported that another manager called her a “cunt” and “bitch” because she asked him to sign a payroll change form. Pratt also re- ported that Boyd’s subordinates hounded an HR employee for a date, and one of them sent her an email stating “U Smoken Hot.” Pratt and the other employees in the HR de- partment felt uncomfortable with these remarks and the email, and Pratt reported as much to Jacobs. Pratt’s reporting did not make her popular within the company. Jacobs testified that managers complained to him about the “direction” the HR department was moving in and expressed their distrust of Pratt. Suspicions about Pratt were later memorialized when she was reviewed by a third-party consulting group, Utech. Utech, which WAF engaged to help manage its growth, interviewed about 50 WAF employees and managers, including Pratt, about the strengths and weak- nesses of the company and its leaders. Utech compiled the employee feedback into a September 2018 report that identified strengths and weaknesses with WAF’s culture generally. Utech noted that “[m]orale [had] gone downhill over [the] last 3 years,” (predating Pratt’s hir- No. 24-1901 7

ing) and that there was a lack of trust and accountability. Fur- ther, Utech observed that “[t]he current culture at WAF is one of gossip and drama. Issues are avoided or talked about indi- rectly, causing more division within the organization. Sides are being taken, people are protective of their areas and peo- ple are really unsure who to trust . . . This starts with leader- ship.” Utech also provided individual feedback for WAF’s man- agers, including Pratt, Boyd, and Jacobs. Under Pratt’s strengths, the report said that she worked hard to “make things better” and knew her job well. The Utech report elabo- rated that “Deb hasn’t been in her role long[,] yet is seen as trying to make improvements within the HR department and the organization, but has been met with some resistance.” The report also said that Pratt lacked the trust of other employees who believed that she had a “hidden agenda,” gossiped, and held grudges. The report further stated that many employees thought Pratt “[i]nvestigates people instead of focusing on what’s best for the company.” The Utech report also criticized each of the other WAF managers. Some of the criticism was starkly similar to Pratt’s. The report stated that employees did not trust Jacobs because, for example, he shared “information with others that was told to him in confidence.” Employees at WAF also found that he lacked assertiveness. The report stated that Kory Brockman, WAF’s Senior Vice President and Chief Financial Officer, had a “hidden agenda” and perpetuated “drama” within the com- pany. Boyd, who was at a similar level within the organiza- tion as Pratt, received feedback that he lacked “respect and professionalism” and “instigate[d] and perpetuate[d] drama.” The report stated that Tom Behnke, also at a similar 8 No. 24-1901

level, talked “badly about people” and could be “hot-headed and unprofessional.” Of the 11 managers reviewed in the Utech report, Pratt was the only woman. In November 2018, relying on the Utech report, Jacobs gave Pratt a yearly performance review. Out of a total possible score of 32, Pratt received a 10. Citing the Utech report, Jacobs wrote that Pratt had been a “very [divisive] figure at WAF.” He also stated that while she was “good on the benefits side,” she required more training on the “HR side.” Because of her low score, Pratt did not receive a performance bonus for 2018 as she had in 2017. Concerned with her score, Pratt met with Jacobs in Decem- ber 2018 to ask for additional feedback. During that meeting, Pratt shared that she felt she was being retaliated against for making reports about harassment and discrimination, partic- ularly her report about Boyd and Goehring. She felt like she had a target on her back. Pratt was unsatisfied with Jacobs’ response during that meeting and raised her concerns again in an email to Jacobs on March 7, 2019. She wanted specific examples of her work that warranted the low scores. In this email, she also ex- pressed her concerns that Jacobs told other WAF employees not to approach her for HR purposes and that, by doing so, he was perpetuating a culture of distrust and gossip. She con- cluded by explaining that she felt she had been retaliated against because she reported the incident between Boyd and Goehring and could not “turn a blind eye” to it. One week later, on March 15, 2019, Pratt was fired. No. 24-1901 9

II. Procedural History Pratt brought claims against WAF under Title VII for retal- iation, sex discrimination, hostile work environment, and pay discrimination. The district court dismissed Pratt’s hostile work environment claim at the pleading stage but allowed the remainder of her claims to proceed. Pratt does not challenge this dismissal on appeal. After discovery, the district court granted summary judg- ment in favor of WAF on all the remaining claims. Applying McDonnell Douglas Corp. v. Green, 411 U.S. 792 (1973), it con- cluded that Pratt could not state a prima facie claim for sex discrimination because she was not meeting WAF’s legitimate performance expectations. As to Pratt’s pay discrimination claim, the court concluded that she had not identified a suffi- ciently close comparator. Finally, on Pratt’s retaliation claim, the court concluded that she did not engage in a protected ac- tivity. DISCUSSION We review the district court’s grant of summary judgment de novo. Joll v. Valparaiso Cmty. Schs., 953 F.3d 923, 928 (7th Cir. 2020). Summary judgment is appropriate if there are no disputes of material fact and the movant is entitled to judg- ment as a matter of law. FED. R. CIV. P. 56(a). In viewing the facts in the light most favorable to Pratt, we must “try to focus on the most persuasive story possible on [her] behalf when asking whether a verdict in her favor would be reasonable or could result only from irrational speculation.” Joll, 953 F.3d at 928. At summary judgment, we are careful not to usurp the role of the jury. See Wince v. CBRE, Inc., 66 F.4th 1033, 1041 (7th Cir. 2023) (We “try to discern what the record shows” to 10 No. 24-1901

ensure a plaintiff is not “wrongly deprived of a trial.”). As long as there is “at least one reasonable way to tell the story in favor of” the non-movant, “a jury rather than appellate judges must choose among them.” Joll, 953 F.3d at 935 (emphasis in original). I. Title VII Sex Discrimination In granting summary judgment to WAF on Pratt’s Title VII sex discrimination claim, the district court concluded that she abandoned this claim because her arguments opposing sum- mary judgment focused only on her retaliation claim. Argu- ments may be waived if they are “perfunctory and underde- veloped” or if they are so “unsupported by pertinent author- ity” that the court must conduct research and form arguments on behalf of the party. United States v. Holm, 326 F.3d 872, 877 (7th Cir. 2003) (citing United States v. Berkowitz, 927 F.2d 1376, 1384 (7th Cir. 1991)). Because waived claims cannot be consid- ered on appeal, we do not find waiver lightly. If there is even “a (barely) adequate basis on which to follow [the] argu- ment,” we will not find waiver. Id. Applying this principle, we disagree that Pratt abandoned her discrimination claim. Pratt’s brief in opposition to sum- mary judgment argued that WAF was not entitled to judg- ment on her discriminatory discharge claim and was suffi- ciently fleshed out over multiple paragraphs with citations to both record evidence and case law. She argued that her evi- dence was sufficient to meet each prima facie element of a dis- crimination claim. While some of these arguments bled into her retaliation arguments, that is no basis to find waiver of an entire claim. We therefore turn to the merits. No. 24-1901 11

Title VII makes it unlawful for an employer to discrimi- nate against an employee “because of” the employee’s sex. 42 U.S.C. § 2000e-2(a)(1). Pratt sets out her arguments in the framework of McDonnell Douglas, but the question before us is simple: “whether the evidence would permit a reasonable factfinder to conclude that the plaintiff’s race, ethnicity, sex, religion, or other proscribed factor caused the discharge . . . .” Ortiz v. Werner Enters., Inc., 834 F.3d 760, 765 (7th Cir. 2016); see also Tyburski v. City of Chicago, 964 F.3d 590, 598 (7th Cir. 2020) (“A plaintiff may put forth and a court may analyze ev- idence using the McDonnell Douglas framework, but neither must do so.”); Skiba v. Ill. Cent. R.R., 884 F.3d 708, 720 (7th Cir. 2018) (“However the plaintiff chooses to proceed, at the sum- mary judgment stage the court must consider all admissible evidence . . . .” (citation omitted)). Here, as in many Title VII cases, the “sole question that matters” is causation: “whether a statutorily proscribed factor caused” a discharge. Joll, 953 F.3d at 929 (citation omitted). Put another way, the issue is whether the employer would have fired Pratt if she had been a man and everything else had re- mained the same. Gehring v. Case Corp., 43 F.3d 340, 344 (7th Cir. 1994). To assess causation in employment discrimination cases, we “ask whether the totality of the evidence shows discrimi- nation, eschewing any framework or formula.” Igasaki v. Ill. Dep’t of Fin. & Pro. Regul., 988 F.3d 948, 958 (7th Cir. 2021) (cit- ing Ortiz, 834 F.3d at 765). “[A]ll evidence belongs in a single pile,” Ortiz, 834 F.3d at 766, because “[d]irect as well as cir- cumstantial evidence may support an inference of causation, and thus intent” in Title VII cases, Joll, 953 F.3d at 929 (citing Ortiz, 834 F.3d at 764); see also Ames v. Ohio Dep’t of Youth 12 No. 24-1901

Servs., 605 U.S. 303, 325 (2025) (Thomas, J. concurring) (quot- ing Desert Palace, Inc. v. Costa, 539 U.S. 90, 99 (2003)) (“That ‘[c]onventional rul[e] of civil litigation’—that a plaintiff can proceed with direct or circumstantial evidence—applies with full force to Title VII cases.”). “The reason for treating circumstantial and direct evi- dence alike is both clear and deep-rooted: ‘Circumstantial ev- idence is not only sufficient, but may also be more certain, sat- isfying and persuasive than direct evidence.’” Desert Palace, 539 U.S. at 100 (quoting Rogers v. Mo. Pac. R.R. Co., 352 U.S. 500, 508, n. 17 (1957)). The import of circumstantial evidence is especially acute in the employment discrimination context given that it is the rare discrimination case that might “permit easy inferences, such as the fabled employer who admits to firing an employee because of race.” Ortiz, 834 F.3d at 765. Be- cause “[f]ew discrimination cases are so straightforward—in- deed they are often factually complex and require sifting through ambiguous pieces of evidence,” id., we have made clear that “[a] plaintiff may use circumstantial evidence to prove discrimination through a chain of inferences.” Downing v. Abbott Lab’ys, 48 F.4th 793, 804 (7th Cir. 2022). Broadly, three types of circumstantial evidence can sup- port an inference of intentional discrimination: “[1] ambigu- ous or suggestive comments or conduct; [2] better treatment of people similarly situated but for the protected characteris- tic; and [3] dishonest employer justifications for disparate treatment.” Joll, 953 F.3d at 929. Pratt has produced substan- tial evidence falling within each of these categories. Viewing this evidence holistically, we conclude that there is at least one reasonable way to tell the story in favor of her sex discrimina- tion claim. No. 24-1901 13

A. Ambiguous or Suggestive Comments or Conduct The record reflects the culture at WAF as one in which “suggestive comments” flowed freely. WAF does not dispute that men in managerial roles regularly made inappropriate comments and jokes to and at the expense of women employ- ees. Pratt’s own experience reflects this. She and her HR em- ployees felt that they were treated like secretaries and not asked to perform their actual job duties because of their sex. Pratt also received pushback for reporting workplace harass- ment and discrimination and was called a “bitch” and a “cunt” for trying to do her job. Such offensive commentary is far from “suggestive.” It is definitively sexist. See Passananti v. Cook Cnty., 689 F.3d 655, 665 (7th Cir. 2012) (noting that “bitch” and “cunt” are “sexually degrading, gender-specific epithets” (citation omitted)). Pratt reported this conduct to Jacobs repeatedly and he took little to no action to address it. Pratt’s declaration that Jacobs “did not” address discrimination at WAF is particu- larly telling because Pratt, of all WAF employees, was best sit- uated to know what actions Jacobs was or was not taking in response to reports. After all, she maintained personnel rec- ords and investigated complaints. At a certain point, a reason- able jury could view Jacobs’ inaction as acquiescence, even tacit approval of the conduct Pratt experienced firsthand. 1

1 Our dissenting colleague suggests that Pratt’s claim fails because Ja-

cobs did not make any of the alleged sexist comments, and “stray remarks made by nondecisionmakers are not evidence that the decision had a dis- criminatory motive.” Post at 33 (quoting Crabtree v. Nat’l Steel Corp., 261 F.3d 715, 723 (7th Cir. 2001)). But Crabtree found only that the district court did not abuse its discretion by excluding, at trial, a single suggestive re- mark, made two years after the alleged adverse employment action. See 14 No. 24-1901

Culture is often set by those at the top. See Burlington Northern & Santa Fe Ry. Co. v. White, 548 U.S. 53, 69 (2006) (“The real social impact of workplace behavior often depends on a constellation of surrounding circumstances.” (citation omitted)). Considering Jacobs’ senior executive role at WAF, a jury might reasonably deduce that he was influenced by this sexist environment, and in fact, had lent it his imprimatur by failing to intervene and address Pratt’s complaints. See Hasan v. Foley & Lardner LLP, 552 F.3d 520, 528 (7th Cir. 2008), as cor- rected (Jan. 21, 2009) (courts are “not require[d]” to “ignore comments made by someone who is not directly responsible for an employee’s supervision” especially where such com- ments “incited” discriminatory comments from other em- ployees and “may have instigated” an employee’s termina- tion). Further, Jacobs had knowledge of his subordinates’ bias and of their participation in the Utech report, but appar- ently—and in spite of Pratt’s complaints to him on the sub- ject—did not question whether the report could have been tainted by a broader culture of sexism at WAF. This is a theory of sex discrimination that we have blessed in the past. See Pe- rez v. Thorntons, Inc., 731 F.3d 699, 709 (7th Cir. 2013) (panel majority rejected dissent’s argument that non-deci- sionmaker’s sexist comments were “stray remarks” where those remarks did not “st[an]d alone,” and the decisionmaker “was informed of [the non-decisionmaker’s] bias,” but “chose

261 F.3d at 723. Crabtree is silent on whether a panoply of sexist remarks and conduct by managers leading up to, and possibly contributing to a termination, can be considered on de novo review at summary judgment. See id.; see also Hunt v. City of Markham, 219 F.3d 649, 652–53 (7th Cir. 2000) (making clear that courts must take care not to over-read “stray remarks” cases). No. 24-1901 15

to credit” the non-decisionmaker’s account of the employee’s alleged infractions). As a result, a jury could find that because Jacobs relied on input he understood to be discriminatory and acceded to the discriminatory atmosphere, his own decision to discharge Pratt was made because of her sex. Context matters, and where derogatory comments do not stand alone, and are im- plicated within a larger web of investigations and complaints to decisionmakers, “we allow the jury to hear such evidence and weigh it for what it is worth.” Id. at 710; cf. Dey v. Colt Constr. & Dev. Co., 28 F.3d 1446, 1459 (7th Cir. 1994) (collecting cases) (“Summary judgment generally is improper where the plaintiff can show that an employee with discriminatory ani- mus provided factual information or other input that may have affected the adverse employment action.”). In highly similar contexts, several of our sister circuits long have recognized the importance of assessing workplace culture in discrimination claims by considering discrimina- tory comments and actions by non-decisionmakers. See, e.g. Merritt v. Old Dominion Freight Line, Inc., 601 F.3d 289, 301 (4th Cir. 2010) (reversing summary judgment for employer be- cause, while the views/actions of other employees cannot be attributed to an ultimate decisionmaker, “at some point the corporate environment in which [the decisionmaker] worked places” the employment decision in a “less neutral context”); Risch v. Royal Oak Police Dep’t, 581 F.3d 383, 393 (6th Cir. 2009) (reversing summary judgment for employer because “dis- criminatory remarks, even by a nondecisionmaker, can serve as probative evidence of pretext” and that type of circumstan- tial evidence may establish the evidence of a “discriminatory atmosphere”); Brewer v. Quaker State Oil Refin. Corp., 72 F.3d 16 No. 24-1901

326, 333 (3d Cir. 1995) (reversing summary judgment for em- ployer because, although “stray remarks by non-deci- sionmakers or by decisionmakers unrelated to the decision process are rarely given great weight,” such statements “may provide some relevant evidence of discrimination” at trial). Because we treat the evidence holistically, we cannot ignore the environment Pratt worked in and must consider those facts in the light most favorable to her. See Ortiz, 834 F.3d at 766. On its own, this discriminatory culture evidence may not be enough to conclude that Pratt’s discrimination claim sur- vives summary judgment. But viewed with the rest of the rec- ord evidence, the alleged sexist “remarks are part of the evi- dence of pretext that the jury should have the opportunity to weigh at trial.” Perez, 731 F.3d at 710. B. Dishonest Justifications for Disparate Treatment Pratt also presented evidence of “dishonest employer jus- tifications for disparate treatment.” Joll, 953 F.3d at 929. WAF claims it had three reasons for terminating Pratt in March 2019: (1) the negative feedback she received in the Utech re- port, particularly that employees did not trust her; (2) con- cerns about her ability to keep sensitive HR information con- fidential; and (3) errors she made in early 2018 related to a change in WAF’s benefits plan for its employees and hiring an external candidate for a union position. But none of these reasons compel us to rule for WAF as a matter of law. Some occurred a year before she was fired, and other employees who had similarly negative feedback in the Utech report received bonuses, not notices of termination. A reasonable jury could infer that these stated reasons were dis- No. 24-1901 17

honest. Evidence of Pratt’s performance issues does not jus- tify automatically keeping her claims from a jury. “Federal employment discrimination laws do not limit their protection to perfect employees.” Paterakos v. City of Chicago, 147 F.4th 787, 797 (7th Cir. 2025). Starting with the Utech report, Pratt was the only woman reviewed in the report. Even though her negative feedback was not unique, she was the only one punished for it. Jacobs, Brockman, and Boyd, who all received negative feedback, some very similar to Pratt’s, were rewarded with performance bonuses at the end of 2018. This differential treatment sup- ports an inference that WAF’s reliance on the Utech report to justify Pratt’s termination is dishonest. See Marnocha v. St. Vin- cent Hosp. & Health Care Ctr., Inc., 986 F.3d 711, 719 (7th Cir. 2021) (quoting McDaniel v. Progress Rail Locomotive, Inc., 940 F.3d 360, 368 (7th Cir. 2019)) (“All things being equal, if an employer takes an action against one employee in a protected class but not another outside that class, one can infer discrim- ination.”). Regarding Pratt’s handling of confidential information and early 2018 errors as justifications for WAF’s decision to fire Pratt, both occurred nearly a year before she was fired. As for the union hiring issue, it is unclear from the record when this occurred. Beyond their staleness, these reasons are incon- sistent with the feedback Pratt received; in her 2018 perfor- mance report, Jacobs stated that she was “good on the benefits side” and made no mention of issues with union hiring or confidentiality. WAF’s reliance on this stale evidence and its conflict with Pratt’s 2018 performance review combined with the culture evidence discussed above could lead a jury to con- clude that the real reason for its decision to terminate Pratt 18 No. 24-1901

was based on her sex. See Joll, 953 F.3d at 932; Loudermilk v. Best Pallet Co., 636 F.3d 312, 315 (7th Cir. 2011) (collecting cases) (“[A]n employer who advances a fishy reason takes the risk that disbelief of the reason will support an inference that it is a pretext for discrimination.”). C. Better Treatment of Employees Outside Pratt’s Pro- tected Group Pratt has also presented evidence of “better treatment of people similarly situated but for [her] protected characteris- tic.” Joll, 953 F.3d at 929. Namely, the male managers in the Utech report. Pratt points to two similarly situated employ- ees, Boyd and Behnke, both men, who were at a similar man- agerial level as Pratt, reported to the same supervisor, and were subject to the same feedback process through the Utech report. Even though Boyd and Behnke also received negative feedback, neither was disciplined or fired, and Boyd received a bonus in 2018. They therefore provide a “meaningful com- parison” to Pratt “to permit a reasonable jury to infer discrim- ination.” Coleman v. Donahoe, 667 F.3d 835, 841 (7th Cir. 2012). While it is true that Boyd and Behnke were not HR man- agers like Pratt, we have been clear that a comparator need not be “identical.” Id. at 846. Pratt must only point us to an analogue, not a twin. See Johnson, 892 F.3d at 895 (quoting Coleman, 667 F.3d at 841) (noting that Coleman “warned against using a mechanical ‘magic formula’ for the similarly- situated inquiry”). Here, it would be impossible to find Pratt’s twin given that Pratt was the sole leader of WAF’s HR depart- ment, and thus, no one else at WAF had the same job descrip- tion as her (or, most likely, comparable experience, education, and other qualifications). We decline to read a gaping hole No. 24-1901 19

into Title VII precluding executives, without direct profes- sional peers, from proving their discrimination claims. Because “this is not a hard and fast test . . . the require- ment to find a similarly situated comparator is really just the same requirement that any case demands—the requirement to submit relevant evidence.” Id. Here, the Utech report pro- vides strong evidence that Behnke and Boyd are similarly sit- uated comparators insofar as they were reviewed in the same category as Pratt under the same criteria. See Mitchell v. Exxon Mobil Corp., 143 F.4th 800, 810 (7th Cir. 2025) (relying on com- pany’s “annual performance assessment,” which “assign[ed] employees to assessment groups based on similar roles and with similar experience, even if employees have different job titles, report to different supervisors, and work in different lo- cations across the country” to identify similarly situated com- parators). The Utech report grouped Boyd, Behnke, and Pratt in the “individual leaders” section and provided a review of each leader’s strengths and weaknesses based on standard- ized interviews with the same staff members. The weaknesses identified for Boyd, Behnke, and Pratt centered around the same key issues: a purported penchant for drama, as well as a lack of professionalism. Boyd was said to “ha[ve] people do his dirty work,” “be intimidating and aggressive,” and “insti- gate[] and perpetuate[] drama.” Behnke, meanwhile was said to “talk[] badly about people” and “be hot-headed and un- professional.” And Pratt was said to “ha[ve] a hidden agenda,” engage in “[g]ossip and backstabbing,” and “hold[] grudges against people.” These reviews suggest a climate of distrust in Boyd’s, Behnke’s, and Pratt’s departments at WAF. Pratt has therefore provided us with “evidence about how [Boyd and Behnke] compared in the [Utech] assessment pro- 20 No. 24-1901

cess,” as well as evidence “show[ing] that [Pratt, Boyd, and Behnke] were assessed unfavorably in comparison to others in their assessment group[],” namely, individual leaders at WAF. Mitchell, 143 F.4th at 812. Boyd and Behnke were not disciplined or demoted based on the Utech report’s findings (like Pratt was). Considering the totality of the evidence, there is “at least one reasonable way to tell the story” in Pratt’s favor, support- ing an inference that she was fired because of her sex. Joll, 953 F.3d at 935. And rather than “weigh[ing] any conflicting evi- dence” ourselves, we must allow a jury to connect the dots. Omnicare, Inc. v. UnitedHealth Grp., Inc., 629 F.3d 697, 705 (7th Cir. 2011). Accordingly, WAF is not entitled to summary judg- ment on Pratt’s Title VII sex discrimination claim. 2

2 Our dissenting colleague does not consider the totality of the evi-

dence and instead insists that we are breaking “our rules on stray remarks, cat’s paw liability, [and] hostile work environment.” Post at 33. But we have made clear that to assess causation in employment discrimination cases, we “ask whether the totality of the evidence shows discrimination, eschewing any framework or formula.” Igasaki, 988 F.3d at 958 (citing Ortiz, 834 F.3d at 765). The “sole question that matters” is “[w]hether a reasonable juror could conclude that [Pratt] would have kept [her] job if [s]he had a different [sex], and everything else had remained the same.” Ortiz, 834 F.3d at 764. We stray from this fundamental inquiry and mis- takenly step into the jury’s role if we dismiss evidence as unworthy of our consideration because it does not fall neatly into a doctrinal bucket. After all, evidence that might support a discriminatory discharge might also support a hostile work environment claim, and vice versa. And evidence of other managers’ remarks and biases might place Pratt’s termination in a “less neutral context.” Merritt, 601 F.3d at 301. We cannot disregard the cumulative impact of such evidence simply by isolating bits and pieces of evidence and labeling it as relevant only to one particular theory of dis- No. 24-1901 21

II. Title VII Retaliation We turn next to Pratt’s retaliation claim. Pratt contends she was fired in retaliation for reporting harassment and discrim- ination at WAF and for complaining to Jacobs that she be- lieved she was being retaliated against. Title VII prohibits employers from “retaliating against an employee for opposing or participating in an investigation of an unlawful employment practice.” Lewis v. Wilkie, 909 F.3d 858, 866 (7th Cir. 2018) (citing 42 U.S.C. § 2000e-3(a)). Our analysis focuses on one fundamental question: “could a rea- sonable trier of fact infer retaliation . . . ?” Castro v. Devry Univ., Inc., 786 F.3d 559, 564 (7th Cir. 2015) (collecting cases). To prevail, Pratt must offer evidence that would permit a rea- sonable juror to find that: (1) she engaged in protected activ-

crimination. A jury must sort through and weigh all the evidence Pratt has put forward. At any rate, this is not a stray remarks case. See supra at n. 1. This is also not a cat’s paw case. Pratt does not suggest that Jacobs was an “un- witting manager” or an “unbiased individual” who terminated her solely because he was beguiled by Pratt’s sexist coworkers, Schandelmeier-Bartels v. Chicago Park Dist., 634 F.3d 372, 379 (7th Cir. 2011); instead, Pratt con- tends Jacobs knew of, and adopted, his colleagues’ sexist bias. Last, while Pratt abandoned her hostile work environment claim on appeal, we need not put blinders on and automatically dismiss evidence merely because it more self-evidently fits into one sort of sex discrimination theory than an- other. Cf. Meritor Sav. Bank, FSB v. Vinson, 477 U.S. 57, 67 (1986) (acknowl- edging overlap between hostile work environment claim and sexual har- assment claim). Rather, per Ortiz’s instruction, we consider all of the evi- dence “in a single pile,” 834 F.3d at 766, and focus on “whether a statuto- rily proscribed factor caused” Pratt’s discharge, Joll, 953 F.3d at 929. More precisely, we consider whether a reasonable jury considering all that evi- dence could reasonably find that the employer discharged Pratt because of her sex. The answer is yes. 22 No. 24-1901

ity; (2) she suffered an adverse employment action; and (3) a causal connection exists between the protected activity and that adverse employment action. Gnutek v. Ill. Gaming Bd., 80 F.4th 820, 824 (7th Cir. 2023). There is no dispute that Pratt meets the second element; she was fired. We focus our analy- sis on the first and third elements. A. Protected Activity For her claim to survive, Pratt must show that she sin- cerely and reasonably believed that she was reporting con- duct prohibited by Title VII. Castro, 786 F.3d at 564. Her claim falls within the opposition clause in Title VII, which protects an employee who opposes “any practice made an unlawful employment practice by [Title VII].” 42 U.S.C. § 2000e-3(a). “Oppose” carries its ordinary meaning: “to resist or antago- nize; to contend against; to confront; resist; withstand.” Craw- ford v. Metro. Gov’t of Nashville & Davidson Cnty., 555 U.S. 271, 276 (2009) (citation modified). “When an employee communi- cates to her employer a belief that the employer has engaged in . . . a form of employment discrimination, that communica- tion virtually always constitutes the employee’s opposition to the activity.” Id. (emphasis in original) (citation and internal quotation marks omitted). Title VII does not just protect employees who can accu- rately divine whether a claim has merit before reporting it to a supervisor. Rather, it protects all who honestly and reason- ably believe they are reporting prohibited conduct. On this point, the district court held Pratt to too high of a standard. It concluded that WAF was entitled to judgment on her retalia- tion claim because she did not engage in a protected activity. The court focused solely on Boyd’s jokes to Goehring during the January 2018 meeting and concluded that those jokes were No. 24-1901 23

not unwelcome or intended to harass and therefore did not amount to sexual harassment. But for a Title VII retaliation claim, we do not consider whether the reported conduct “was persistent or severe enough to be unlawful.” Magyar v. Saint Joseph Reg’l Med. Ctr., 544 F.3d 766, 771 (7th Cir. 2008). Whether Boyd’s conduct constituted sexual harassment un- der Title VII “does not matter.” Castro, 786 F.3d at 564. What does matter is that Pratt reasonably and honestly believed she was reporting prohibited conduct. See id. Pratt honestly believed she was opposing unlawful behav- ior. In February 2018, Pratt wrote that the comments between Boyd and Goehring were “unprofessional and against the law” and that “any witnesses to [it] have the ability to file a sexual harassment complaint with the company.” Her belief that this conduct was unlawful under Title VII was certainly sincere. Her belief also appears reasonable. A joke about an em- ployee exposing her buttocks or being told to sit on another employee’s thumb might be considered actionable sexual har- assment by some. Regardless of whether it was later deter- mined that Boyd did not make the jokes, at the time she re- ported it, Pratt had a reasonable belief that she was reporting conduct prohibited by Title VII. And this was not the only incident Pratt reported. She par- ticipated in a host of protected activity by repeatedly com- municating conduct to Jacobs that she believed was prohib- ited under Title VII. She reported multiple instances where Boyd made inappropriate comments to women, including that he called another employee a “bitch,” and that she be- lieved he treated women employees worse than men. She also 24 No. 24-1901

reported another WAF manager for treating minority employ- ees differently than white employees. And not all of Pratt’s reports were on behalf of other em- ployees. She also reported that a male supervisor called her a “bitch” and a “cunt.” It is not hard to understand how Pratt believed that the use of these blatantly sexist insults against her was more than just inappropriate. Nothing in the record suggests that Pratt did not honestly and reasonably believe she was reporting conduct prohibited by Title VII. See Craw- ford, 555 U.S. at 276. In addition to these reports, Pratt twice told Jacobs that she believed she was being retaliated against: once during her performance review and again in an email one week before she was terminated. Between her reports and her personal complaints of retaliation, Pratt established that she had a rea- sonable and good faith belief that she was opposing unlawful conduct under Title VII. In reaching the opposite conclusion, the district court adopted a unique retaliation rule applicable only to HR em- ployees. Relying on an out-of-circuit case, Littlejohn v. City of New York, 795 F.3d 297, 318 (2d Cir. 2015), the district court concluded that in order to bring a retaliation claim, Pratt needed to show that she was not simply performing her job duties in passing along complaints, but that she was actively supporting other employees in asserting their Title VII rights or personally complaining about discriminatory conduct. We are doubtful that Littlejohn’s heightened standard for retaliation claims by HR employees passes muster. By its text, Title VII requires only that Pratt oppose an unlawful employ- ment practice. See 42 U.S.C. § 2000e-3(a) (making it “unlaw- No. 24-1901 25

ful . . . for an employer to discriminate against any . . . em- ployee[] . . . because he has opposed any practice made an un- lawful employment practice by this subchapter”). We should not graft judge-made requirements onto the text requiring plaintiffs who work in HR departments to show “personal” opposition or “active[] support[]” of other employees. See Lit- tlejohn, 795 F.3d at 318 (quoting Sumner v. U.S. Postal Service, 899 F.2d 203, 209 (2d Cir. 1990)). Further, the Supreme Court has never interpreted the opposition clause to be restricted to only certain types of employees. See Crawford, 555 U.S. at 276. To the extent WAF argues that policy considerations warrant an exception, such concerns “cannot override” the plain text of the statute. Muldrow v. City of St. Louis, 601 U.S. 346, 358 (2024). But we need not decide whether to accept the height- ened standard in Littlejohn here. Pratt vigorously opposed conduct directed both at her and others. WAF also argues that Pratt’s evidence of reporting, which was mostly admitted through her affidavit filed in opposition to summary judgment, cannot be considered based on the sham affidavit rule. WAF argues that when asked in her dep- osition to identify all instances of discrimination and harass- ment that she reported to Jacobs, Pratt did not explain all of the incidents she later cited in her affidavit and never at- tempted to amend or supplement her testimony. Accordingly, WAF contends that Pratt may not rely on this evidence. It is true that “a plaintiff cannot manufacture a genuine dispute of material fact by contradicting her prior deposition testimony with an affidavit.” Leibas v. Dart, 108 F.4th 1021, 1026 (7th Cir. 2024). But an affidavit that “clarifies ambiguous or confusing deposition testimony” is permissible. Perez v. 26 No. 24-1901

Staples Cont. & Com. LLC, 31 F.4th 560, 570 (7th Cir. 2022) (ci- tation omitted). This is no sham affidavit. First, Pratt provided deposition testimony about many of these incidents that she reported to Jacobs. See R.33-1 at 46:1–47:1 (Pratt believed Culp engaged in racial discrimination); 56:7–57:16 (Boyd referred to Rabitz as a “bitch”); 202:6–8 (Behnke retaliated against an employee); 202:9–10 (Pratt believed Boyd was mocking her due to her sex); 203:1–204:14 (a manager sent an email to a woman em- ployee in the HR department which stated “U Smoken Hot”); 208:18–209:5 (Pratt felt disrespected by Jacobs when he yelled at her and believed it was heightened due to her gender); 209:13–25 (women employees were being given secretarial work because they were women). To the extent there were any incidents that Pratt did not testify about in her deposition, her notes, which were available to counsel during her deposi- tion, contained even more descriptions of reports that Pratt made to Jacobs. Pratt therefore has met the first element of her retaliation claim. Whether this claim proceeds to trial thus depends on whether a jury could find that she established a causal con- nection between this protected activity and her termination. B. Causal Connection The “key question” on causation is “whether a reasonable juror could conclude that there was a causal link between the protected activity or status and the adverse action.” Rozumal- ski v. W.F. Baird & Assocs., Ltd., 937 F.3d 919, 924 (7th Cir. 2019). To establish that the protected activity caused the ad- verse employment decision, a plaintiff may offer evidence of “suspicious timing, ambiguous statements, behavior toward No. 24-1901 27

or comments directed at other employees in the protected group, and other bits and pieces from which an inference of discriminatory intent might be drawn.” Boumehdi v. Plastag Holdings, LLC, 489 F.3d 781, 792 (7th Cir. 2007). Of these, cau- sation is most “frequently established by showing that there was a suspiciously short period of time between the em- ployee’s complaint and the adverse employment action.” Id. at 793. To establish a causal link between her protected reporting and her termination, Pratt cites the timing of her termination decision and the lack of non-pretextual explanations for it. As to timing, Pratt was fired one week after raising concerns about retaliation to Jacobs. This is certainly close enough in time to support an inference of causation. But WAF contends that Jacobs decided to fire Pratt before she sent the March 7, 2019 email. In other words, WAF insists that Jacobs decided to fire Pratt before she engaged in any protected activity, and therefore the timing is not suspicious. WAF relies on Jacobs’ declaration filed at summary judgment to support its argument. But this argument fails for two rea- sons. First, the March 7, 2019 email is just one example of the protected activity that Pratt engaged in. She was a serial re- porter of discrimination and harassment to Jacobs throughout her tenure as HR manager. Pratt also raised concerns about being retaliated against during her 2018 performance review, which occurred on December 8, 2018. This was still relatively close in time to her termination and could establish a suffi- cient causal connection. Boumehdi, 489 F.3d at 793; see also Huff v. Buttigieg, 42 F.4th 638, 650 (7th Cir. 2022) (whether five 28 No. 24-1901

months between protected activity and adverse employment action was suspicious timing was a question for the jury). Second, Jacobs’ declaration statement that he decided to fire Pratt before March 7 is in tension with his own deposition testimony. During his deposition, Jacobs could not remember when Pratt was fired. In his declaration, however, he states that he “made the decision to terminate Pratt’s employment prior to March 7, 2019.” One could argue that Jacobs’ later-in- time recollection of the specific timing of the decision, without any corroboration, smacks of sham testimony that a jury should evaluate. Yet, even credited, Jacobs’ declaration does not foreclose Pratt’s causation theory because she engaged in a surplus of protected activity before March 7 too. At the very least, a question of fact remains as to when Jacobs made the decision to terminate Pratt in relation to all her protected ac- tivity. See Perez, 731 F.3d at 707 (genuine issue of fact existed where employer’s statement of facts contradicted its own ev- idence). Pratt also contends that her evidence of pretext supports the causal link. Specifically, she emphasizes that Jacobs wrote in her performance review that she was “good on the benefits side,” but then cited issues with benefits as a reason for her termination. She also contends that no other employee who received negative feedback in the Utech report was termi- nated. If a Title VII retaliation plaintiff “offers specific evi- dence from which the finder of fact may reasonably infer that the proffered reasons do not represent the truth, the case then turns on the credibility of the witnesses.” Collier v. Budd Co., 66 F.3d 886, 893 (7th Cir. 1995). That is precisely the case here, and Pratt’s retaliation claim therefore must survive summary judgment. No. 24-1901 29

III. Pay Discrimination Finally, we turn to Pratt’s pay discrimination claim. Pratt identifies Emery Coonen as a comparator. She argues that even though her job duties initially included those of the en- vironmental health and safety manager, when WAF later hired Emery Coonen to fill that role, he was paid more than she was, raising the inference that she was paid less because of her sex. To establish a claim of pay discrimination under Title VII, Pratt must produce evidence that a similarly situated male employee was treated more favorably. Johnson, 892 F.3d at 895. While similarly situated means “directly comparable in all material aspects,” the “proposed comparator need not be identical in every conceivable way,” and courts “must con- duct a common-sense examination.” Id. (quoting Reed v. Free- dom Mortg. Corp., 869 F.3d 543, 549 (7th Cir. 2017) (internal quotation marks omitted)). We ask, “are there enough com- mon features between the individuals to allow a meaningful comparison?” Id. (quoting Coleman, 667 F.3d at 841). Relevant evidence on this question includes whether the employees were supervised by the same person, were subject to the same standards, and engaged in similar conduct “without such dif- ferentiating or mitigating circumstances as would distinguish their conduct or the employer’s treatment of them.” Id. (cita- tion omitted). The record in this case is simply too undeveloped for us to determine whether Coonen is a sufficient comparator. For ex- ample, neither WAF nor Pratt put forward any evidence that Pratt and Coonen were subject to similar or different work- place standards. The record is similarly devoid of any facts regarding whether Pratt and Coonen had similar qualifica- 30 No. 24-1901

tions and credentials, especially related to environmental health and safety. Pratt stated generally that Coonen was not more experienced or qualified than she was but did not pro- vide any facts supporting this conclusion other than her be- lief. See id. at 895 (“Evidence of what has happened to other employees is only relevant if that employee is in the same boat as the plaintiff.”). Accordingly, Pratt did not meet her burden to survive summary judgment on her pay discrimination claim. CONCLUSION The grant of summary judgment for WAF on Pratt’s pay discrimination claim is AFFIRMED. The grant of summary judgment for WAF on Pratt’s claims for Title VII discrimina- tion and retaliation is REVERSED and the case is REMANDED for trial on those claims. No. 24-1901 31

KIRSCH, Circuit Judge, concurring in the judgment in part and dissenting in part. I agree with the majority that the Wis- consin Aluminum Foundry is entitled to summary judgment on Debra Pratt’s pay discrimination claim. I disagree that her other claims warrant a trial. To salvage Pratt’s suit, the major- ity ignores the distinction between direct discrimination and hostile work environment claims and creates a special com- parator analysis for executives. A faithful application of our precedent should result in summary judgment for the Foundry. Sexist comments by non-decisionmakers, employ- ees who weren’t similarly situated to Pratt, and a delay be- tween some of Pratt’s failures and the decision to fire her do not add up to a Title VII discrimination claim. Similarly, Pratt’s retaliation claim fails because she cannot show that WAF’s stated, legitimate reasons for ending her employment were a lie. I For Debra Pratt’s sex discrimination claim, the only ques- tion that matters is whether a reasonable factfinder could con- clude that the Wisconsin Aluminum Foundry (WAF) fired her because she’s a woman. Ortiz v. Werner Enters., Inc., 834 F.3d 760, 766 (7th Cir. 2016); 42 U.S.C. § 2000e-2. On this record, no reasonable jury could make that finding. A The majority says that a jury could find a culture of sexism existed at WAF and that Senior Vice President Ben Jacobs con- doned that atmosphere. Ante, at 13–16. But there’s insufficient evidence to impute discriminatory motives based on other employees’ conduct to Jacobs, who actually made the decision to fire Pratt. See Coleman v. Donahoe, 667 F.3d 835, 848 (7th Cir. 32 No. 24-1901

2012) (a discrimination claim “requires proof that the deci- sionmaker acted for a prohibited reason”) (citation modified). Jacobs didn’t participate in the problematic conduct Pratt re- ported on, and he didn’t witness it. The majority reasons that perhaps biased employees provided feedback that informed the Utech report (which then informed Jacobs’s decision), and that because Jacobs didn’t take action to address the discrim- inatory conduct of other employees, he must have condoned it. Ante, at 13–15. This is speculation, and these are not reason- able inferences. See Van Houdnos v. Evans, 807 F.2d 648, 655 (7th Cir. 1986) (considering a claim for sexual discrimination under 42 U.S.C. § 1983 and observing that “Silence alone … is not evidence of discriminatory intent.”); E.E.O.C. v. Pipefitters Ass’n Loc. Union 597, 334 F.3d 656, 660 (7th Cir. 2003) (apply- ing Title VII and noting that “inaction, unless invidious, is not discrimination in any accepted sense of the term. Most people don’t take active measures to combat discrimination; their in- action does not condemn them as discriminators.”). It should matter to us that Pratt isn’t suing Eugene Boyd, the subject of many of her complaints. Instead, she’s suing her employer—WAF. And in such a suit, Title VII requires Pratt to prove that WAF fired her because of her sex. See Ortiz, 834 F.3d at 765 (noting that the legal standard “is simply whether the evidence would permit a reasonable factfinder to con- clude that the plaintiff’s race, ethnicity, sex, religion, or other proscribed factor caused the discharge or other adverse em- ployment action”). To be precise, Pratt must show that dis- crimination was a motivating factor in WAF’s decision. See Comcast Corp. v. Nat’l Ass’n of Afr. Am.-Owned Media, 589 U.S. 327, 337 (2020). No. 24-1901 33

Ordinarily, when a plaintiff tries to prove discrimination through evidence about an employer’s culture, we are consid- ering a hostile work environment claim. And when a plaintiff attempts to prove direct discrimination by using the conduct or speech of non-decisionmakers, we apply our rules for stray remarks and cat’s paw liability. The majority says this case is special—that it isn’t governed by our rules on stray remarks, cat’s paw liability, or hostile work environment. See Ante, at 13 n.1 & 20 n.2. But those rules should guide us here, and they demonstrate the ways in which the majority’s novel theory about discriminatory culture evidence extends and disrupts our Title VII doctrine. For instance, our law says that “stray remarks made by nondecisionmakers are not evidence that the decision had a discriminatory motive.” Crabtree v. Nat’l Steel Corp., 261 F.3d 715, 723 (7th Cir. 2001); see Mlynczak v. Bodman, 442 F.3d 1050, 1057–58 (7th Cir. 2006) (finding that a number of discrimina- tory comments made by a non-decisionmaker did not show that an employer discriminated); Brooks v. Avancez, 39 F.4th 424, 439–40 (7th Cir. 2022) (applying similar standards under the ADEA and ADA and concluding that discriminatory com- ments made by three employees did not “singularly influ- ence[]” a decisionmaker to terminate an employee). And, un- der the cat’s paw doctrine, to impute discriminatory remarks or conduct from Boyd and other WAF employees to Jacobs (the decisionmaker), Pratt must show that discrimination was the proximate cause of her termination, which requires “some direct relation between the injury asserted and the injurious conduct alleged,” and excludes links that are “too remote, purely contingent, or indirect.” Staub v. Proctor Hosp., 562 U.S. 411, 419 (2011) (citation modified); see Johnson v. Koppers, Inc., 726 F.3d 910, 915 (7th Cir. 2013). 34 No. 24-1901

Pratt cannot make that showing. Sexist comments made by WAF employees who had nothing to do with Pratt’s termi- nation are remote and indirect from the decision that mat- ters—they are not a proximate cause of WAF’s action. See Staub, 562 U.S. at 419; Vega v. Chi. Park Dist., 954 F.3d 996, 1006 (7th Cir. 2020) (observing that proof of discrimination re- quires “a causal link between an employment decision made by an unbiased individual and the impermissible bias of a non-decisionmaking co-worker”) (citation modified); Rozu- malski v. W.F. Baird & Assocs., Ltd., 937 F.3d 919, 925 (7th Cir. 2019) (“At a minimum, the person with discriminatory animus must influence the ultimate employment decision enough to be a ‘proximate cause’ of that action.”); Hunt v. City of Mark- ham, 219 F.3d 649, 652–53 (7th Cir. 2000) (“[T]he fact that someone who is not involved in the employment decision of which the plaintiff complains expressed discriminatory feel- ings is not evidence that the decision had a discriminatory motivation.”) (citation modified). Relying on precedent from outside our circuit, the major- ity undermines the causal relationship required by Title VII, reasoning that if enough non-decisionmakers make serious enough discriminatory comments, and a decisionmaker doesn’t act or speak out in response, we can impute discrimi- natory motives to the decisionmaker. Ante, at 13–16. That ap- proach ignores our precedent and collapses the distinction be- tween a hostile work environment claim (which Pratt does not raise on appeal) and a claim for direct discrimination. See, e.g., Alexander v. Casino Queen, Inc., 739 F.3d 972, 979–83 (7th Cir. 2014) (discussing the two theories); Nat’l R.R. Passenger Corp. v. Morgan, 536 U.S. 101, 115 (2002) (“Hostile environ- ment claims are different in kind from discrete acts.”). This case isn’t like Perez v. Thorntons, Inc., 731 F.3d 699 (7th Cir. No. 24-1901 35

2013), either, because a biased non-decisionmaker in that case played a central role in the employee’s termination. Id. at 709. I agree with my colleagues that in discrimination cases, evidence (direct or circumstantial) must be considered as a whole. See Ortiz, 834 F.3d at 765. But not everything that hap- pens at a workplace tends to show that an employer violated Title VII. As this court has considered discrimination claims, we’ve carefully drawn the line between evidence that proves that an employer discriminated and evidence that does not. We’ve drawn that line by distinguishing between claims for hostile work environment and direct discrimination, and by applying our rules for cat’s paw liability and stray remarks. My colleagues ignore that careful work. They assert that if evidence doesn’t fit into a “doctrinal bucket” it must still be somehow relevant to our analysis. Ante, at 20–21 n.2. Yet we don’t start with the assumption that all the evidence pre- sented will support a plaintiff’s claim—that’s a determination we make now, in assessing whether WAF is entitled to sum- mary judgment. According to the majority, this is the rare sit- uation where evidence of discriminatory culture that doesn’t involve a decisionmaker is relevant to a Title VII direct dis- crimination claim (rather than to a claim based on a hostile work environment). And they say this is the rare case where our usual doctrines for assessing that kind of evidence (stray remarks, cat’s paw) aren’t adequate. If that sounds suspi- ciously like cover for extending our law in new ways, that’s because it is. Title VII imposes liability on employers only for their own deeds. See Dunn v. Wash. Cnty. Hosp., 429 F.3d 689, 691 (7th Cir. 2005). That’s why we focus on what motivates the decisionmaker who fires an employee—not what other bi- ased employees may have said or done. It’s undisputed that 36 No. 24-1901

Jacobs fired Pratt. And there is nothing to connect the sexism or discriminatory attitudes of other WAF employees to Jacobs. The evidence of sexist comments by others at WAF doesn’t prove that Pratt was fired because she is a woman. B Disparate treatment can prove discrimination where em- ployees (who do not share a protected characteristic) are sim- ilarly situated, but one was treated better. Coleman, 667 F.3d at 845–46. “Employees are similarly situated if they dealt with the same supervisor, were subject to the same standards, and had engaged in similar conduct without such differentiating or mitigating circumstances as would distinguish their con- duct or the employer’s treatment of them.” Dunlevy v. Langfelder, 52 F.4th 349, 353 (7th Cir. 2022) (quotation omit- ted). While these requirements are well-established, my col- leagues suggest that we may ignore them if an employee (here Pratt) has no “direct professional peers.” Ante, at 19. The ma- jority says that two employees at WAF—Eugene Boyd and Tom Behnke—were similarly situated to Pratt and received better treatment, because WAF’s independent consulting re- view (the Utech report) criticized all three, but only Pratt was disciplined. Ante, at 18–20. Neither of these men is a suitable comparator, however. Because WAF’s outside consulting review looked at Pratt, Boyd, and Behnke, the majority assumes that they were sub- ject to the same standards. But that’s not right. To show that her comparators were subject to the same standards, Pratt needs to prove that she, Behnke, and Boyd worked under the same employment policies, and that their professional roles were not “so different from [Pratt’s] as to render the compar- No. 24-1901 37

ison effectively useless.” Coleman, 667 F.3d at 848–49 (quota- tion omitted). Pratt cannot make that showing. The Utech report was a one-off analysis conducted by an outside consulting firm: it wasn’t an employment policy and didn’t include a statement of standards explaining how em- ployees were being assessed. Cf. id. (standards of conduct are policies or rules, set by the employer, that apply to employ- ees). The report centered on feedback that was specific to each employee and their particular work areas. That matters be- cause Boyd and Behnke did different work than Pratt: Boyd was a vice president, Behnke was a foreman, and Pratt was manager of HR. Cf. Mitchell v. Exxon Mobil Corp., 143 F.4th 800, 804 (7th Cir. 2025) (“Assessment groups comprise em- ployees in similar roles with similar experiences.”). Position and rank can show that employees weren’t subject to the same standards “provided that the employer took these factors into account when making the personnel decision in question.” Coleman, 667 F.3d at 849–50 (quotation and emphasis omit- ted). WAF did that in Pratt’s case—in her performance review following the Utech report, Jacobs gave Pratt low marks be- cause “HR is supposed to be a safe confidential place for the organization, and I believed under Deb it was not [a safe space].” And Jacobs fired Pratt because of failings that were specific to her work in HR. Nor did Pratt, Behnke, and Boyd engage in similar con- duct. Pratt needed to prove that she and her co-workers “en- gaged in comparable rule or policy violations.” Id. at 850 (quo- tation omitted). It’s true that all three received negative feed- back in the report, but every other employee did too—that was how the report worked. The negative feedback for these three employees wasn’t comparable. The consultants noted 38 No. 24-1901

that others at the company didn’t trust Pratt, that she con- ducted unnecessary investigations instead of focusing on what was best for the company, and that she had a hidden agenda. By contrast, Behnke was hot-headed, unprofessional, and protective of his people. Boyd was unprofessional, inex- perienced, avoided responsibility, and lacked a leadership presence. Even if receiving negative feedback in the Utech report for different reasons counts as comparable conduct (it doesn’t), Pratt wasn’t disciplined or fired only for the report’s findings. Jacobs gave her a poor review because of the report, a lack of trust in Pratt’s department, and confidentiality issues. He fired her based on a pattern of problems, of which the report was only one example. Pratt makes no argument that Behnke and Boyd had similar trust problems in their departments, or that they engaged in similar conduct that raised questions about their confidentiality. In our circuit, we take a flexible and common-sense ap- proach to identify suitable comparators in discrimination cases. Coleman, 667 F.3d at 846. But our law still requires that similarly situated employees be directly comparable to the plaintiff in all material respects. Id. (citation modified). When a plaintiff cannot make that showing, our job is to say so and consider the discrimination evidence that exists. The majority instead suggests that we should lower the bar for some special employees, and that refusing to do so would create a “gaping hole” in Title VII. Ante, at 18–19. But there’s nothing unfair or unequal about applying Title VII consistently. If there’s a gap- ing hole in our doctrine, it’s the rule my colleagues write to- day—that we examine whether employees dealt with the same supervisor, were subject to the same standards, and en- No. 24-1901 39

gaged in similar conduct without differentiating circum- stances, except when doing so would leave an employee with- out a comparator. Pratt, Boyd, and Behnke weren’t subject to the same standards, didn’t engage in similar conduct, and aren’t suitable comparators. C What’s left of Pratt’s sex discrimination claim? Just legiti- mate reasons for her termination. Pratt was fired because the organization wanted a change in HR leadership and because of ongoing performance problems: lack of trust, an inability to maintain confidentiality, the Utech report, an error related to a benefits plan, and the hiring of an external candidate for a union position. That other employees weren’t punished for negative feed- back in the Utech report isn’t suspicious. Pratt hasn’t shown that she was comparable to any other employees considered in that report, or that other employees committed similar mis- conduct in addition to being written up by the consulting firm. Likewise, the fact that Pratt was told she was “good on the benefits side” at a performance review (which included largely negative feedback) three months before she was fired, isn’t inconsistent with her subsequent termination. Pratt per- formed well in the benefits area at a time when she was per- forming poorly in general, but WAF cited a previous error in that part of her work when it terminated her. Some of the reasons WAF gave for Pratt’s termination weren’t recent, but there’s no rule requiring an employer to fire an employee as soon as they learn of mistakes or miscon- duct. Moreover, Jacobs decided to fire Pratt because of a pat- tern of problems, not because of any particular issue. Even if 40 No. 24-1901

some of the reasons cited were old, delay between cited mis- conduct and termination doesn’t, standing on its own, prove that an employer is lying. See Hitchcock v. Angel Corps, Inc., 718 F.3d 733, 737–42 (7th Cir. 2013) (reversing a grant of summary judgment when evidence of pretext included shifting, incon- sistent explanations (some of which involved stale miscon- duct), an immediate change in treatment after learning of a pregnancy, and potentially offensive statements by a supervi- sor); Peirick v. Ind. Univ.-Purdue Univ. Indianapolis Athletics Dep’t, 510 F.3d 681, 691–94 (7th Cir. 2007) (finding an em- ployer’s reasons pretextual when one of the reasons for termi- nation was stale, the employee exceeded expectations, she wasn’t warned about her misconduct, and management didn’t discuss the employee’s performance with her direct su- pervisor). The majority seems to acknowledge that the evidence of sex discrimination in this case is entirely circumstantial. Ante, at 12. But without connecting Jacobs to others’ sexually dis- criminatory conduct, absent meaningful points of compari- son, and with no evidence that the company was lying, the circumstances of Pratt’s firing do not show that she was dis- criminated against. No reasonable jury could find that Pratt was terminated because she is a woman. II Pratt cannot succeed on her retaliation claim, either. Title VII prohibits firing an employee because she opposes an un- lawful employment practice. 42 U.S.C. § 2000e-3(a). As to whether Pratt engaged in protected activity, I agree with the majority that we need not decide in this case whether an HR professional must show something more than that she was merely passing along complaints of discrimination affecting No. 24-1901 41

others. See id. (protected action is opposition to “any practice made an unlawful employment practice by this subchapter”). Pratt opposed conduct that was directed at her. Rather, at is- sue is whether there’s “sufficient evidence to permit a reason- able fact finder to conclude that retaliatory motive caused the discharge[.]” See Igasaki v. Ill. Dep’t of Fin. & Pro. Regul., 988 F.3d 948, 959 (7th Cir. 2021) (quotation omitted). Terminating an employee close on the heels of her protected activity can indicate that an employer’s stated reasons for the firing are pretextual. See Daza v. Indiana, 941 F.3d 303, 309 (7th Cir. 2019). But suspicious timing alone is rarely enough, and usu- ally requires other evidence to support the inference of a causal link. See id. In this case, all we have is timing. Viewing the facts in the light most favorable to Pratt, WAF decided to fire her eight days after protected activity (the March 2019 email to Jacobs). There is no set legal rule for deciding when events are close enough together for an inference of causation to be drawn solely on the basis of suspicious timing. Kidwell v. Eisenhauer, 679 F.3d 957, 966 (7th Cir. 2012). In such a case, “we typically allow no more than a few days to elapse between the pro- tected activity and the adverse action,” but the inference still “depends on context.” Id. (quotation omitted). Here, the gap between Pratt’s email and the company’s decision to fire her is small. But drawing an inference of causation in this case would be inappropriate given the context: a complete absence of other evidence of pretext, Pratt’s poor performance review just a few months prior, and evidence that Pratt’s perfor- mance issues were ongoing at the time of her termination. See id. 42 No. 24-1901

The majority says there’s other evidence of pretext: that Jacobs told Pratt she was doing well handling benefits in De- cember 2018 but cited a failure in that area when he decided to fire her, and that other employees received negative feed- back in the Utech report but weren’t fired. Ante, at 28. For rea- sons already discussed, I disagree. Neither Pratt’s perfor- mance evaluation nor the experience of other, non-similarly situated employees shows that WAF lied about why it de- cided to fire Pratt. Because “a suspicion is not enough to get past a motion for summary judgment,” Kidwell, 679 F.3d at 966 (quotation omitted), I would affirm the district court’s grant of summary judgment as to Pratt’s retaliation claim.