HARLINGTON WOOD, Jr., Circuit Judge.

Mary Nell Little, plaintiff-appellant, brought this action against her former employer, Cox’s Supermarkets (Cox’s), alleging that her discharge was due to Cox’s desire to terminate her employee benefit plan rights in violation of the Employee Retirement Income Security Act (ERISA), 29 U.S.C. § 1001 et seq. Little also brought a supplemental claim of defamation under the laws of Indiana. Cox’s responded with a state law counterclaim for conversion. The district court entered summary judgment in favor of Cox’s on the ERISA claim and declined to exercise jurisdiction on the state law claims. The district court also assessed costs and a small portion of Cox’s attorney’s fees against Little and her counsel pursuant to the ERISA statute. Little appeals both the summary judgment and cost and fee awards. We affirm.

I. BACKGROUND

Mary Nell Little began employment as a eashier for Cox’s Supermarkets in Richmond, Indiana in 1974. She worked for this small chain in various locations as both a cashier and courtesy office clerk. She was retained as a full time employee, with retirement and health insurance benefits, despite working less than the requisite number of hours per week for substantial periods. Cox’s was *640 aware that her husband, the late A1 Little, 1 suffered from severe health problems and that the couple relied on Little’s health insurance coverage. The premium for this coverage was shared equally by Little and her employer, amounting to a weekly payment of $55.70 each. The health plan, in turn, paid for her husband’s medical expenses, which totalled approximately $170 per month.

On November 4, 1991, three cashiers reported to Cox’s office manager, Janice Anderson, that they suspected Little of stealing merchandise — namely, panty hose and a turkey. Anderson notified the store manager, Fornie Benge. Four days later, on November 8, Little used her break time to select three dry-good items from the shelves, which she brought to the front of the store and placed on an open register. She then selected three steaks and put them in the store’s office. Two cashiers observed Little with the steaks and reported the matter to Steve Ross, the night manager, who phoned Benge at home. Benge returned to the store and examined the office. Little then gathered her purchases from their separate locations and paid for them. Following this Benge confronted Little and asked to see her receipt, which she produced. Benge thereupon informed her that he would have to let her go for theft, mentioning in particular the previous reports regarding the turkey. Upon Little’s request, Benge phoned the store’s vice president, Richard Cox, who declined to speak with Little and who concurred in Benge’s decision. Little’s employment was thus ended.

Little brought suit in March, 1993, alleging that her termination was pretextual for the purpose of interfering with her rights under Cox’s employee benefit plans in violation of § 510 of the ERISA statute, 29 U.S.C. § 1140. The district court applied the appropriate burden-shifting analysis and concluded that Little had not established a prima facie case of discrimination, nor supported her claim that the reason given was pretextual. Accordingly, the court entered summary judgment in favor of Cox’s. On Cox’s motion entered in accordance with ERISA’s fee-shifting provision, 29 U.S.C. § 1132(g)(1), the court awarded to Cox’s the costs of the action plus a small portion of attorney’s fees ($1000). Little now appeals both the summary judgment and cost and fee orders.

II. ANALYSIS

A. Application of Local Rule 56.1

Little first contends that the district court abused its discretion by holding her to a “higher procedural standard” than that applied to Cox’s regarding its local rule for summary judgment motions. The rule in question, S.D.Ind.L.R. 56.1, provides that a party moving for summary judgment must file a “Statement of Material Facts,” supported by appropriate citations. The opposing party must then file any material controverting the movant’s position, “together with an answer brief that shall include ... a ‘Statement of Genuine Issues’ setting forth, with appropriate citations ... all material facts.” S.D.Ind.L.R. 56.1.

The rule states that in determining the motion for summary judgment, the district court will assume that the facts as claimed and supported by the moving party are admitted to exist without controversy, except to the extent that such facts are controverted in the “Statement of Genuine Issues” filed by the opposing party. See S.D.Ind.L.R. 56.1.

In this case, Cox’s, as movant, filed its factual assertions, together with appropriate citations, but mistitled the document “Proposed Undisputed Facts.” Little’s response, however, consisted of a “Statement of Genuine Issues” in the form of a list of questions only. In this document, no effort was made to identify precise factual issues or to controvert Cox’s statements, and no indication was given as to where corresponding evidence in the record might be found. Little points to the “Statement of Facts” found elsewhere in her answer brief, which contained citations to the record, and urges that the district court should have considered this document in its determination of whether issues existed for trial. She opines that Cox’s mistitled sub *641 mission was accepted for this purpose, and therefore she is due equal flexibility since neither party complied fully with L.R. 56.1.

It is generally held that a “failure to draw the district court’s attention to an applicable legal theory waives pursuit of that theory in this court.” Teumer v. General Motors Corp., 34 F.3d 542, 546 (7th Cir.1994) (citations omitted). Little did not raise the issue of the adequacy of Cox’s L.R. 56.1 filings with the district court, and thus this argument is waived. Even viewing the issue more broadly, however, we find no abuse of discretion. This court has repeatedly recognized the importance and usefulness of this and similar local rules throughout the circuit, 2 as well as “the exacting obligation these rules impose on a party contesting summary judgment.” Waldridge v. American Hoechst Corp., 24 F.3d 918, 921-22 (7th Cir.1994) (and eases cited therein). Moreover, it is clear that the decision whether to apply the rule strictly or to overlook any transgression is one left to the district court’s discretion. Id. at 923. A district court’s interpretation of its own rule is accorded considerable deference by us. Id. at 923 n. 4 (citations omitted).

The district court, in its order granting Cox’s motion for summary judgment, declared Cox’s version of events “undisputed” in accordance with the terms of L.R. 56.1, but nevertheless assembled the facts and considered the elements of the prima facie ease by drawing upon Little’s Statement of Facts in her answer brief. Thus the court did not turn a blind eye to the facts elsewhere available, though it is permitted to do so by non-compliance with the local rule.

Furthermore, Little’s claim that she was held to a “higher” procedural standard simply overlooks the allocation of burdens— and the important rationale — represented by summary judgment procedure and rules such as L.R. 56.1. These rules serve to notify the parties of the factual support for their opponent’s arguments, but more importantly inform the court of the evidence and arguments in an organized way — thus facilitating its judgment of the necessity for trial.

The requirements of such rules are not onerous, but they are exacting. See Waldridge, 24 F.3d at 922-23. Local Rule 56.1 makes explicit the responding party’s burden of controverting the movant’s position with adequate citations to the record. While “the non-movant need not match the movant witness for witness,” Anderson v. Liberty Lobby, 477 U.S. 242, 248-49, 106 S.Ct. 2505, 2510, 91 L.Ed.2d 202 (1986), there still exists the obligation of “demonstrating that there is a pending issue of material fact.” Id. (emphasis supplied). Such a duty is usually more involved than that faced by the movant. General principles of advocacy suggest, and we have often repeated, that a party contesting summary judgment has a responsibility under such rules to “highlight which factual averments are in conflict as well as what record evidence there is to confirm the dispute.” Id. at 922. It is reasonable to assume that just as a district court is not required to “scour the record looking for factual disputes,” id., it is not required to scour the party’s various submissions to piece together appropriate arguments. A court need not make the lawyer’s case. We find the district court did not abuse its discretion in the manner in which it enforced its local rule.

B. Summary Judgment

Little also claims that the district court erroneously granted Cox’s motion for summary judgment. We review this grant de novo, viewing all the evidence in the light most favorable to the non-moving party; summary judgment will be upheld if the record reveals no genuine issue of material fact and the moving party is entitled to a judgment as a matter of law. Fed.R.Civ.P. 56(c); Anderson v. Liberty Lobby, 477 U.S. 242, 247, 106 S.Ct. 2505, 2509, 91 L.Ed.2d 202 (1986).

The statute at issue here, § 510 of ERISA, states in pertinent part that “[i]t shall be unlawful for any person to discharge, fine, suspend, expel, discipline, or discriminate *642 against a participant or beneficiary ... for the purpose of interfering with the attainment of any right to which such participant may become entitled under the plan.” 29 U.S.C. § 1140. In analyzing Little’s claim, the parties have utilized the familiar burden-shifting analysis first announced in McDonnell Douglas v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973). The use of this approach was neither briefed nor argued. For the purposes of this case, we do not decide whether it is appropriate. According to this analysis, a plaintiff must first establish a prima facie case of discrimination. Having done so, the burden then shifts to the defendant, who must offer a legitimate and non-diseriminatory reason for the employment action taken. Once this is established, the burden then shifts back to the plaintiff to show that the reason(s) offered by the defendant was merely pretextual. E.g., St. Mary’s Honor Center v. Hicks, — U.S. -, -, 113 S.Ct. 2742, 2747, 125 L.Ed.2d 407 (1993).

Courts differ regarding the required content of a prima facie case in the ERISA context. 3 We agree with the district court’s well-reasoned choice of the “Turner/Dister approach,” see Turner v. Schering-Plough Corp., 901 F.2d 335, 347 (3d Cir.1990) (citing Dister v. Continental Group, Inc., 859 F.2d 1108, 1111-12 (2d Cir.1988)); Massie v. Indiana Gas Co., 752 F.Supp. 261, 268 (S.D.Ind.1990), and note that this is the appellant’s preferred analysis.

In order to establish a prima facie case under this framework, a plaintiff must show that she (1) belongs to the protected class; (2) was qualified for the position involved; and (3) was discharged or denied employment under circumstances that provide some basis for believing that the prohibited intent was present. Turner, 901 F.2d at 347. There is a substantial question in this case whether Little has carried this burden. The first two elements of the prima facie case are satisfied. There is no dispute that Little belongs to the relevant class as she is entitled to ERISA protection in connection with her employer’s benefit plans. Nor is there any real dispute that she met the minimal qualifications for her position. At issue is the third element concerning intent. Little claims that “the pretextual nature” of her termination, “combined with the circumstantial evidence of Cox’s desire to reduce its benefit costs” provides a sufficient basis for believing that the prohibited intent was present. On the facts of this case, neither of these arguments is sufficient to provide the requisite basis for a rational inference that a discriminatory motive guided Cox’s decision.

Here Little is attempting to build her prima facie case from circumstantial evidence usually reserved for the “pretext” stage of analysis. This method is permissible; we have previously stated that “proof of pretext allows, but does not require, the fact-finder to infer discrimination.” Petfetti v. First Nat’l Bank of Chicago, 950 F.2d 449, *643 451 n. 4 (7th Cir.1991), cert0 denied, 505 U.S. 1205, 112 S.Ct. 2995, 120 L.Ed.2d 871 (1992). But while an inference is enough to establish a prima facie case, this nevertheless may not be created by mere subjective belief or debatable assertions. Proof of pretext is required. Such proof may be supplied indirectly, by direct or circumstantial evidence. The sufficiency of that evidence, however, clearly remains within our review. See Perfetti, 950 F.2d at 451-52. Moreover, all such evidence is not equal; we have described circumstantial evidence of pretext as the “weakest category” of all, id. at 451, and stated that in evaluating evidence, courts are “not required to evaluate every conceivable inference which can be drawn from evidentiary matter, but only reasonable ones.” Parker v. Federal Nat’l Mortg. Ass’n, 741 F.2d 975, 980 (7th Cir.1984).

As “direct evidence of pretext,” Little claims that Cox’s justification for its decision to terminate her has been inconsistent and that the policies on which Cox’s relies for that decision are applied unevenly. Regarding the first claim, while different store personnel have referred to the justification for the dismissal variously as “theft,” “concealment,” and “assembling purchases,” we agree with the district court’s assessment that these are merely attempts to apply a formal label to the same conduct. The district court observed correctly that the undisputed facts involved more than the mere assemblage of purchases. Rightly or wrongly, Little was suspected of theft and was being watched closely. Fellow employees reported her possession of the steaks, and Little was found to have placed the steaks in the store office rather than at the register with the other items she intended to purchase. The relatively minor differences among the labels attached to Little’s conduct, considered together with the undisputed facts, do not suggest that these are pretextual labels or reasons, see Anderson v. Baxter Healthcare Corp., 13 F.3d 1120, 1124 (7th Cir.1994), nor do their differences rise to a level of inconsistency that permits an inference of discriminatory motive.

A similar conclusion must be reached concerning the descriptions of the order of payment for the merchandise in question. While there is some discrepancy between Benge’s statement on the “Employee Counseling Form” (which was prepared by another employee) and his later deposition about whether Little paid for the dry good items and the steaks as one transaction or two, this is hardly a material conflict. In any event both statements describe one short, continuous set of events, and interpreting them in the light most favorable to Little adds no support to her theory. We agree with the district court that any attempt to construe the discrepancy as direct evidence of pretext is a strained one.

Regarding her claim that store policies are applied unevenly, Little points to deposition statements suggesting that employees assembled purchases “routinely.” None of the relevant statements cited by Little, however, are attributable to employees at the north store location, from which Little was discharged. More importantly, this assertion again ignores context; more was involved here than the mere assembling of purchases. Furthermore, the fact that employees did assemble purchases proves nothing about management’s reaction to the practice. Other depositions indicate that management, upon request, granted permission to certain employees to assemble purchases; Little had no such permission. There is also evidence that many employees were terminated for violating a wide array of similar store policies.

Finally, Little argues that Cox’s efforts to save money provides “circumstantial evidence of pretext.” She offers a variety of factors as proof: (1) that Cox’s is motivated by small amounts of money; (2) that Cox’s had an interest in saving benefit costs; (3) that Cox’s group health insurance premium had increased by 392%; (4) that the insurer was concerned about liability to disabled dependents; and (5) that Little’s highest earnings occurred within the five years prior to her termination, thus making her retirement plan more expensive to her employer. We have examined each of these factors and conclude that they provide no basis for an inference of discriminatory intent. The factors most nearly relevant to this dispute are *644 those involving insurance liability and premiums, yet Little offers no significant evidence of a causal connection between these factors and her termination. For example, though Cox’s insurance premiums rose, many factors may account for this, including inflation in the health care industry generally. Furthermore, these rates are established on the basis of the claims experience of the past twelve months, yet Little’s husband had no heart problems for more than three years prior to her termination. Indeed, after subtracting Little’s deductible and co-payments, the difference between Little’s claims and her premiums was a mere $216 annually, or $18 per month. Such potential savings can not be viewed realistically as a motivating factor in Little’s discharge. No other evidence of the economic consequences to Cox’s from the termination of Little was offered, and thus we find an insufficient basis for a belief that impermissible concerns motivated this employment decision. We conclude that Little has failed to establish a prima facie case of ERISA discrimination, and affirm the district court’s grant of summary judgment.

C. Award of Costs and Fees

Little also appeals the district court’s award of costs and fees to Cox’s as prevailing party. ERISA provides that “the court in its discretion may allow a reasonable attorney’s fee and costs of action to either party.” 29 U.S.C. § 1132(g)(1). Such awards are subject to an exception for cases in which “special circumstances make an award unjust,” Bittner v. Sadoff & Rudoy Indus., 728 F.2d 820, 830 (7th Cir.1984), but the district court also entertains a “modest presumption” that prevailing parties are entitled to a reasonable attorney’s fee. Id. We review such awards for abuse of discretion. Hooper v. Demco, Inc., 37 F.3d 287, 293 (7th Cir.1994) (citations omitted). Under this standard of review, a district court’s determination will not be disturbed if it has a basis in reason. Marcus v. Shalala, 17 F.3d 1033, 1037 (7th Cir.1994) (citing Pirus v. Bowen, 869 F.2d 536, 540 (9th Cir.1989)).

A variety of approaches has been used to determine whether fee awards in ERISA actions are justified. Cf. Bittner, 728 F.2d at 830 (“substantially justified” test) with Nichol v. Pullman Standard, Inc., 889 F.2d 115, 121 n. 9 (7th Cir.1989) (five factor test). We have stated that whichever approach is used, the “bottom-line question” is the same: was the losing party’s position substantially justified and taken in good faith, or was that party simply out to harass its opponent? Hooper, 37 F.3d at 294 (citations omitted). In answering this question, the district court is not compelled to choose either of these labels; the formal test, as announced in its source, the Equal Access to Justice Act, ultimately asks only whether the losing party’s position was “substantially justified.” 28 U.S.C. § 2412(d)(1)(A). The Supreme Court has defined this standard as a “genuine dispute” that is “ ‘justified in substance or in the main’ — that is, justified to a degree that could satisfy a reasonable person.” Pierce v. Underwood, 487 U.S. 552, 565, 108 S.Ct. 2541, 2550, 101 L.Ed.2d 490 (1988).

Before awarding a small portion of attorney’s fees, the district court examined several factors. These included its finding that no genuine issue of material fact had been established, despite considerable discovery, and that a defamation claim had been sent to state court. The court also considered the total time expended by defendant’s counsel, which resulted in charges of approximately $50,000, and Cox’s reasons for requesting fees totalling $1000 to $5000 against Little and $2500 to $7500 against her counsel.

Addressing the issue of special circumstances, the court recognized Little’s meager assets and the fact that her counsel is a solo practitioner. In view of these factors, the court made a single fees award at the bottom of the requested range, $1000, and assessed this against both Little and her counsel, jointly and severally. 4 Given our *645 findings regarding the merit of the plaintiffs ease and the explanations provided by the district court, we conclude that an adequate basis existed for the court’s finding of no substantial justification, and for the award of costs and minimal attorney’s fees. Though we might not have reached precisely the same award, we recognize that these difficult determinations are entrusted to the discretion of the district court. We find no abuse of that discretion.

AFFIRMED.

Notes

. A1 Little, originally named as a plaintiff in this suit, died while the case was pending in the district court. Pursuant to Fed.R.Civ.P. 25(a)(2), his name was deleted from the caption, and the case proceeded in favor of his surviving spouse.

. See, e.g., Local General Rule 12, Northern District of Illinois; Local Rule 11, Northern District of Indiana; Local Rule 6.05, Eastern District of Wisconsin; "Rule on Procedure to be Followed on Motions for Summary Judgment,” Western District of Wisconsin.

. Because direct evidence of discrimination is difficult to obtain, the burden-shifting approach used in Title VII and Age Discrimination in Employment Act (ADEA) cases has been suggested for cases brought under § 510 of ERISA. See Turner v. Schering-Plough Corp., 901 F.2d 335, 347 (3d Cir.1990) (citing Gavalik v. Continental Can Corp., 812 F.2d 834 (3d Cir.1987)). Initially, a prima facie case was established under these statutes by proving that the plaintiff: (1) belongs to a protected group; (2) was qualified for the position; and (3) was discharged or denied employment under circumstances that give rise to an inference of discrimination. See Dister v. Continental Group, Inc., 859 F.2d 1108, 1114-15 (citations omitted). Over time, elaboration has caused these requirements to diverge to some degree. To establish a prima facie case under Title VII in this circuit, a plaintiff must show that: (1) she belongs to some protected class; (2) she performed her job satisfactorily; (3) she suffered an adverse employment action; and (4) her employer treated similarly-situated employees outside her classification more favorably. Lenoir v. Roll Coater, Inc., 13 F.3d 1130, 1132 (7th Cir.1994). Under ADEA, a plaintiff is required to establish: (1) she belongs to the protected class; (2) she was doing her job well enough to meet her employer's legitimate expectations; (3) in spite of her performance she was discharged; and (4) the employer sought a replacement for her. LaMontagne v. American Convenience Prod., 750 F.2d 1405, 1409 (7th Cir.1984).

While both of these sets of requirements are imprecise in the ERISA context, they also neglect the fact that a plaintiff in an ERISA action must demonstrate that the employer had the “specific intent” to violate the statute. See 29 U.S.C. § 1140. The Tumer/Dister approach, adopted here by the district court, is also somewhat imprecise, but nevertheless restates the basic requirements involved.

. Creating such liability is not prohibited, and is a matter within the discretion of the district court. The allocation of a fee award between a party and the party’s counsel is not unprecedented, see Pelletier v. Zweifel, 921 F.2d 1465, 1520-23 (11th Cir.) (finding joint and several liability of attorney and client under Fed.R.Civ.P. 11 and 38), cert. denied, 502 U.S. 855, 112 S.Ct. 167, 116 L.Ed.2d 131 (1991), and is apparently subject only to the limitation that the award must be *645 consistent with the pre-existing background of substantive liability rules. See Kentucky v. Graham, 473 U.S. 159, 171, 105 S.Ct. 3099, 3108, 87 L.Ed.2d 114 (1985). We find no abuse of discretion regarding the manner in which the district judge allocated liability in this case.