KANNE, Circuit Judge.

Erika Langenbach worked for Wal-Mart for over ten years, moving her way up the internal hierarchy. Her progress was consistent until she sought promotion to an Assistant Manager position; Wal-Mart did not promote her until more than a year after her initial application submission. Eventually, she completed the company Management-In-Training program and began work as an Assistant Manager.

Although she had impressed her bosses with her previous work, the Assistant Manager position proved to be a challenge. Langenbach struggled with delegation, organization, and time management. After receiving several negative performance reviews, she was fired. Her termination happened to come five months after she returned from FMLA leave.

She now alleges that Wal-Mart retaliated against her for exercising her FMLA rights and discriminated against her because of her sex by delaying her promotion to Assistant Manager, paying her less than her male counterparts, and refusing to promote her further. The district court dismissed the suit following Wal-Mart's motion for summary judgment, and we now affirm.

I. BACKGROUND

Langenbach began working for Wal-Mart in 1998, when she was hired to stock shelves in Mukwonago, Wisconsin. The next year, she requested and was transferred to the Saukville, Wisconsin store, where she was promoted to Jewelry Department Sales Coordinator. In 2001, she was promoted to Jewelry Department Manager. In 2006 or 2007, Langenbach began seeking an Assistant Manager position, applying to Wal-Mart's Management-In-Training program several times without success. She was finally admitted to the Management-In-Training program at a store in West Baraboo, Wisconsin in February 2008. To qualify for promotion to Assistant Manager, a Wal-Mart employee had to meet one of the following minimum requirements: (1) one year retail experience and one year supervisor experience, (2) two years general work experience and one year supervisor experience, or (3) completion of two or more years of college.

Upon completing the program, Langenbach began work as an Assistant Manager at the West Bend, Wisconsin store. She was initially assigned to the day shift but was scheduled for the night shift about a year later. This change-of-shift was in line with her job description, which noted that Assistant Managers were required to work overnight shifts as required.

In 2009, Langenbach received her first annual evaluation as an Assistant Manager. She was given an overall "Solid Performer" rating, but management also noted a number of deficiencies she needed to work on, including her assertiveness and time management skills. Later that year, Langenbach was placed on her first Performance Improvement Plan (the "2009 PIP"). Wal-Mart used these plans to improve sub-par performance and gathered information from managers and supervisees to compile them. The 2009 PIP described several problems with Langenbach's performance, including a lack of leadership, a tendency to push decisions off on associates, spending too much time in the office rather than on the sales floor, not following appropriate overnight shift procedures, and professionalism issues. This PIP was never completed, however, as Langenbach's managers did not hold the anticipated follow-up meetings.

Around January 2010, Langenbach returned to the day shift. On January 21st, Wal-Mart filed a written discipline comment (a "Coaching for Improvement" in Wal-martese) concerning Langenbach. "Coachings" were used for discrete disciplinary problems, while PIPs were used to address ongoing performance issues. According to the comment, Langenbach was not following management routines and frequently failed to complete her duties on time.

Langenbach's annual review took place in April 2010. The Store Manager, Mike Dooley, prepared the review, which gave Langenbach a competency score of 2.63 out of 5 and a rating of "Development Needed." The review noted that Wal-Mart needed to see a "complete turn around" from Langenbach and a renewed sense of "urgency and time management." It described specific issues complying with the overnight stocking program, attendance, and holding underperforming associates accountable.

Later that month, Langenbach discovered that she needed surgery to remove fibroid tumors in her uterus. In July 2010, she submitted a written request for continuous leave under the Family and Medical Leave Act ("FMLA") from July 30, 2010 to August 26, 2010. Wal-Mart approved the request. A few days before her scheduled return to work, Langenbach had some complications involving the incision from her surgery. She spoke with Joanne Massopust, the Market Human Resources Manager, seeking an extension of her leave or an accommodation by which she could work only part-time until the incision healed. Additionally, she provided a return to work certification form from her doctor, which indicated that she would be able to return to work without restrictions on September 13, 2010. In response, Wal-Mart extended her continuous leave through September 13, 2010.1

Upon return from leave, Langenbach was again assigned to the overnight shift. While this shift could be more physically demanding than the day shift, Assistant Managers could delegate the heavy lifting to the associates they supervised. Around this time, Langenbach also had a brief discussion with her supervisor, Courtney Wilcox, where she expressed concerns about her medical condition. Wilcox told Langenbach that she needed to go back on leave or request an Americans with Disabilities Act accommodation if she was concerned about her condition's effect on her job performance. Langenbach did not follow up on the conversation.

Langenbach's mid-year evaluation took place after she returned from leave. Although the evaluation was prepared in July, before Langenbach took her leave, Wal-Mart decided to deliver it after she returned. This evaluation assessed Langenbach's overall competency rating at 2.26 out of 5 and assigned her an overall performance rating of "Development Needed." As in the 2009 evaluation, the managers' comments reflected poor leadership skills and insufficient organization and planning.

Following this poor evaluation, Langenbach was again placed on a Performance Improvement Plan (the "2010 PIP"). The 2010 PIP identified a number of issues with Langenbach's performance. For instance, it described Langenbach's time management problems, noting that she was "not organized to be able to give clear cut deadlines on calendar events, etc. Too many things left until the last minute. [She gets] focused on one task and forget[s] or lose[s] track of what else is going on." It further explained that Langenbach had not consistently implemented company policies or complied with Assistant Manager routines. The 2010 PIP went on to identify actions Langenbach could take to improve her performance, measurement standards by which her performance would be evaluated, and a time frame in which she was expected to improve.

Wilcox and Dooley held the first follow-up session for the 2010 PIP in November 2010. Langenbach's progress was rated as "Below Expectations" and the follow-up document explained that she had failed to improve her time management or the consistency with which she implemented management routines and company policies. At the second follow-up session in January 2011, Wilcox and Dooley once more rated Langenbach's progress as "Below Expectations," again citing her poor time management skills and inconsistent implementation of routines and policies. The final follow-up session took place on March 1, 2011. Prior to this session, Wilcox and Dooley met with Massopust and determined that Langenbach had not made the improvements specified in the plan. As a result, they decided to fire her.

Several other Wal-Mart employees' career paths are also relevant to this appeal. Joe Frankiewicz, who, like Langenbach, had only a high school education, began working at the West Bend Wal-Mart as a meat cutter in 1998. Before that, he managed the meat department at a local Piggly Wiggly. Seven months after he started, he was promoted to Manager of the West Bend meat department, an Assistant Manager position. He had to apply for the position, but he was the only applicant. No Management-In-Training program was involved because no such program existed for the food departments. He was uniquely qualified for this position because he knew how to cut meat, a skill he described as a "lost art."

Adam Zastrow began working at Wal-Mart in 2001 as a produce sales associate in the West Bend store. He started out with a part-time schedule because at the time, he was attending community college. Eventually, he completed two years of community college, but did not obtain an Associate's Degree. Eventually, he transitioned to full-time work. A year-and-a-half later, he attended the Management-In-Training program in Beaver Dam, Wisconsin and was promoted to Assistant Manager at another store. He got the job because he mentioned to his manager that he was looking for work that paid more and she said she would recommend him for the program. About six years later, he was promoted to Shift Manager. A little more than a year after that promotion, he became a Store Manager.

There were also a number of male college graduates who were hired directly into the Management-in-Training program — C.J. Tabasky, Blake Hamel, Preston St. John, LaTravis Henry, and Troney Shumpert. They all played football for Chris Meidt, Wal-Mart's Regional Manager in Wisconsin. Meidt had them e-mail their resumes to him and forwarded them on to the human resources manager.

II. ARGUMENT

On appeal, Langenbach argues that the district court erred in granting summary judgment on her FMLA retaliation claim and her Title VII failure-to-promote and disparate pay claims. We will address each of these claims in turn, mindful that we review the district court's summary judgment determination de novo and will affirm only if, drawing all reasonable inferences in Langenbach's favor, there is no genuine issue of material fact. Celotex Corp. v. Catrett, 477 U.S. 317, 322-23, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986).

A. FMLA Retaliation Claim

Langenbach first asserts that Wal-Mart illegally retaliated against her for exercising her rights under the FMLA by taking several adverse actions against her after she returned from leave. To make out a prima facie retaliation case, an employee can proceed using either a direct or indirect method of proof. Scruggs v. Carrier Corp., 688 F.3d 821, 826 (7th Cir. 2012).

1. Direct Method

To establish a prima facie case using the direct method, Langenbach had to "present evidence of (1) a statutorily protected activity; (2) a materially adverse action taken by the employer; and (3) a causal connection between the two." Id. at 826. The parties do not dispute that Langenbach's leave was statutorily protected. They do, however, dispute which of Wal-Mart's actions were "materially adverse" and whether there was a causal connection between Langenbach's FMLA leave and the materially adverse actions.

Langenbach argues that Wal-Mart took four materially adverse actions against her: giving her the worst performance rating she had ever received, placing her on a performance improvement plan, assigning her to the overnight shift, and terminating her. The first two of these actions — her performance review and the 2010 PIP — are not materially adverse. See Haywood v. Lucent Techs., Inc., 323 F.3d 524, 532 (7th Cir.2003) (negative performance evaluations alone are not an adverse employment action); Cole v. Illinois, 562 F.3d 812, 816 (7th Cir.2009) (implementation of a performance improvement plan not materially adverse action).

Langenbach's assignment to the night shift was likewise not materially adverse. Where there is no evidence the defendant sought to exploit a "known vulnerability" by altering a plaintiff's work schedule upon return from FMLA leave, a schedule change is not a materially adverse action. Porter v. City of Chi., 700 F.3d 944, 955 (7th Cir.2012); cf. Washington v. Ill. Dep't of Revenue, 420 F.3d 658, 662 (7th Cir.2005) (finding that, in altering plaintiff's schedule, employer sought to exploit a "known vulnerability" — the fact that she used her prior flex time schedule to care for her disabled son). Langenbach asserts that the night shift was more physically demanding, and that she had alerted her supervisors to ongoing limitations following her return to work. But Langenbach also did not seek special accommodations after Massopust told her she could, and her doctor had cleared her to work without restrictions. Based on these facts, we don't think it is a reasonable inference that Wal-Mart assigned her to the night shift to exploit a known vulnerability. See McCann v. Iroquois Memorial Hosp., 622 F.3d 745, 754 (7th Cir.2010).

This leaves Langenbach's termination, which the parties agree is a materially adverse action. Proceeding to step three of the direct method analysis, we must determine if there is a causal connection between Langenbach's FMLA leave and Wal-Mart's decision to fire her. Langenbach can establish a causal nexus either with a direct admission from Wal-Mart or "a convincing mosaic of circumstantial evidence." Pagel v. TIN Inc., 695 F.3d 622, 631 (7th Cir.2012). This mosaic can include suspicious timing, ambiguous statements from which retaliatory intent can be inferred, evidence of similar employees being treated differently, or evidence that the employer offered a pretextual reason for the termination. Id. On summary judgment, this circumstantial evidence must point "directly to the conclusion that an employer was illegally motivated, without reliance on speculation." Good v. Univ. of Chi. Med. Ctr., 673 F.3d 670, 676 (7th Cir.2012).

Langenbach cannot lead us to conclude that she was fired for retaliatory reasons without the aid of unbridled speculation. She asserts both that the timing of her termination was suspicious and that her managers offered pretextual reasons for it. But she cannot deny that she had a history of performance issues that preceded her 2010 leave. Before taking leave, she had already received low performance evaluations and been placed on a PIP (although her supervisors did not follow up with it). And the comments her supervisors made about her shortcomings as an Assistant Manager were consistent both before and after her leave; at all relevant times, they complained that she lacked leadership and time management skills. We do not see anything that suggests a different conclusion.

2. Indirect Method

Alternatively, Langenbach asks us to consider whether she can survive summary judgment using the indirect method of proof. To make out such a claim, she must present evidence that she was meeting Wal-Mart's legitimate expectations, suffered an adverse employment action, and was treated less favorably than similarly situated employees who did not request FMLA leave. Caskey v. Colgate-Palmolive Co., 535 F.3d 585, 592 (7th Cir. 2008).

Langenbach did not present evidence that she was meeting Wal-Mart's legitimate expectations. When determining if this is so, "the issue is not the employee's past performance but whether the employee was performing well at the time of [her] termination." Peele v. Country Mut. Ins. Co., 288 F.3d 319, 329 (7th Cir.2002) (internal quotation marks omitted). The only evidence Langenbach points to that her work at the time she was fired was satisfactory is a comment by Wilcox that she was "doing great" and "doing fine." These comments are relevant, as Wilcox was involved in the decision to fire Langenbach. Cf. Young v. James Green Mgmt., Inc., 327 F.3d 616, 622 n. 2 (7th Cir.2003) (in employment discrimination cases, declarant must be involved in the decision-making process affecting the challenged employment action). But they are also consistent with the written evaluations received at the 2010 PIP follow-up meetings, in which Wilcox noted that Langenbach had "made strides to improve her performance" but that they ultimately were not enough to meet company expectations. And, in any event, there is voluminous evidence that Langenbach was not meeting Wal-Mart's expectations — the notes from the 2010 PIP, her evaluations, and the deposition testimony of her co-workers. Without some other evidence of the quality of Langenbach's work, we cannot say she was meeting Wal-Mart's expectations. She therefore cannot make out a case for FMLA retaliation using the indirect method.

B. Title VII

Langenbach also makes two claims of sex discrimination under Title VII: that Wal-Mart failed to promote her because of her sex and that she was paid less than her male counterparts.

Her failure-to-promote claim has two factual bases: first, the ten-year delay in promoting her to Assistant Manager and second, Wal-Mart's failure to promote her past Assistant Manager. The latter can be dismissed out of hand: Langenbach was fired from her Assistant Manager position because she was not meeting Wal-Mart's performance expectations. Further, she admitted in her deposition that she was not qualified for promotion past Assistant Manager, and she never applied for any higher position. See Grayson v. City of Chi., 317 F.3d 745, 748 (7th Cir. 2003) (one of the requirements for making out a prima facie case of discrimination is that the plaintiff applied for the job sought).

The former takes a little more analysis.2 As in the FMLA analysis above, a Title VII discrimination plaintiff can proceed under either the direct or indirect method of proof.

We begin with the indirect method. This method employs a burden-shifting approach: the employee must first make out a prima facie case of discrimination before the burden of disproving discrimination falls to the employer. Collins v. Am. Red Cross, 715 F.3d 994, 1000 (7th Cir.2013). A prima facie case of employment discrimination requires evidence that: (1) the employee is a member of a protected class, (2) she was meeting the employer's legitimate expectations, (3) she suffered an adverse employment action, and (4) similarly situated employees outside of the protected class were treated more favorably. Naficy v. Ill. Dep't of Human Servs., 697 F.3d 504, 511 (7th Cir.2012). Langenbach has presented sufficient evidence of the first three requirements to survive summary judgment. There is no serious question that she is a member of the protected class or that she was meeting the employer's legitimate expectations at the time she applied for promotion to assistant manager; all her reviews from her time as a Jewelry Department Manager were positive. And a significant delay in promotion can be an adverse employment action. Bannon v. Univ. of Chi., 503 F.3d 623, 628 (7th Cir.2007) (recognizing that delayed promotion can be the basis for a discrimination suit but rejecting that basis where the gap between plaintiff's application for and receipt of a promotion was only two months).

This leaves us to consider whether Langenbach pointed to sufficient evidence that similarly-situated male employees were promoted more quickly than she was. To be "similarly situated," co-workers must be "directly comparable to the plaintiff in all material aspects, but they need not be identical in every conceivable way." Coleman v. Donahoe, 667 F.3d 835, 846 (7th Cir.2012) (citing Patterson v. Ind. Newspapers, Inc., 589 F.3d 357, 365-66 (7th Cir.2009)) (internal quotation marks omitted). Whether a co-worker is similarly situated is typically a question for the factfinder, but summary judgment is appropriate where no reasonable jury could find the plaintiff has met her burden. Srail v. Vill. of Lisle, 588 F.3d 940, 945 (7th Cir.2009).

For comparison, Langenbach relies primarily on two male Wal-Mart employees: Joe Frankiewicz and Adam Zastrow. Both were promoted to Assistant Manager less than two years after being hired and upon their first application to the position.3 Langenbach, by contrast, applied several times for a management position in her district beginning in 2006 or 2007. Her applications were denied. She eventually applied outside her district and was accepted into a Management-In-Training program in another district in 2008.

But Frankiewicz and Zastrow both differ in significant ways from Langenbach. Frankiewicz has a special skill — the "lost art" of meat cutting — that uniquely qualified him to become an Assistant Manager heading up the meat department. He also had three years of management experience at a local Piggly Wiggly before starting work at Wal-Mart. Langenbach does not allege that she also had such qualifications. Likewise, Zastrow had two years of community college experience, which alone allowed him to meet the minimum requirements for the Assistant Manager position. Langenbach has no education past a high school diploma. Differences in experience, education, and qualifications are relevant to the similarly-situated analysis so long as the employer took them into account when making the relevant employer decisions. Patterson v. Avery Dennison Corp., 281 F.3d 676, 680 (7th Cir.2002). Wal-Mart's minimum requirements for the Assistant Manager position depend heavily on schooling and experience. We do not think, given these significant differences, that a reasonable fact-finder would find Frankiewicz or Zastrow an adequate comparator for Langenbach.

This brings us to the direct method. To prevail, a plaintiff must produce either direct or circumstantial evidence that would permit a jury to infer that discrimination motivated an adverse employment action. Diaz v. Kraft Foods Global, Inc., 653 F.3d 582, 587 (7th Cir. 2011). Langenbach lacks direct evidence, and so must rely on circumstantial evidence. Circumstantial evidence can take a number of forms, such as suspicious timing, behavior or comments directed at members of the protected group, evidence showing that similarly-situated employees outside the protected group received systematically better treatment, and evidence that the reason the employer gave for the adverse action was pretextual. Id. She relies exclusively on evidence showing that similarly-situated male employees were promoted more quickly than she was. But the analysis for whether the men were similarly situated is substantially the same as the analysis under the indirect method. See Good, 673 F.3d at 675-75 (7th Cir. 2012). Because Zastrow and Frankiewicz are inappropriate comparators under the indirect method, they are similarly inappropriate for the direct method of proving discrimination.

Langenbach also contends that she was paid less than her male counterparts because she was not promoted when she should have been. This argument is inextricably intertwined with her delayed-promotion argument, as Wal-Mart's pay structure is linked to job title and seniority; essentially, her argument is that she was paid less than her male counterparts because she was not promoted at the same rate. Because we find the grant of summary judgment was appropriate on her delayed-promotion claim, we need not address her disparate pay claim.

III. CONCLUSION

On the undisputed facts, we do not believe Langenbach can establish either that she was retaliated against for taking FMLA leave or that she was discriminated against because of her sex. We therefore AFFIRM the district court's decision.