This employment discrimination case presents two questions for review. First, does a plaintiff have to allege specific facts to state a cause of action for intentional racial discrimination? Second, must an “employment practice” be more than a single decision by an employer to be actionable under a disparate impact theory? The answer to both questions is no. We therefore reverse.
I.
In the mid-1980s, unemployment in Illinois dropped to felicitously low levels. Federal spending on unemployment benefits dropped as well, forcing the Illinois Department of Employment Security (IDES or the Department) to cut its budget. In response, the (former) Director of IDES, Sally Ward,1 decided to lay off large numbers of the Department’s staff in several stages throughout 1985. She decided to concentrate the cuts in - certain divisional offices of IDES that are mostly located in Chicago, and in the three regional field offices that serve the Chicago metropolitan area. Although a greater percentage- of the staff of IDES statewide was black (43%) than was the population of Illinois, black employees are concentrated in the field and divisional offices affected by the cuts.
In January 1985, Gail Bradshaw, the Department’s Affirmative Action/Equal Opportunity officer, prepared an adverse impact analysis on the first round of the proposed layoffs for submission to the Illinois Department of Human Rights (IDHR). Her report compared the layoff rate among white employees to the layoff rate among black employees and concluded that the planned layoff would have an adverse impact on black employees. Ward discussed this analysis with Bradshaw, who suggested that the analysis could compare retention rates rather than layoff rates. Ward agreed, and the resulting numbers looked much better. A March 7, 1986, analysis prepared by Bradshaw summarizes the difference between the two approaches. As a whole, the 1985 layoffs resulted in the layoff of 8.6 percent of the Department’s black staff and only 3.0 percent of the Department’s white staff. Conversely, 91.4 percent of the Department’s black staff were retained as were 97.0 percent of its white staff.
Of course, this number-juggling had no substantive content, but it served a purpose. The Equal Employment Opportunity Commission (EEOC) generally regards any selection procedure that , picks blacks and whites by rates that are different by one *376fifth or more to have an adverse impact. EEOC, Uniform, Guidelines on Employee Selection Procedures (1978), 29 C.F.R. § 1607.4D (1991). Use of a layoff rate analysis triggered the “four-fifths” rule, since (again using the figures for all of 1985) the layoff rate among white employees was only 35 percent (3.0 percent/8.6 percent) of the black layoff rate. Use of a retention rate analysis, on the other hand, showed no adverse impact, as black employees were retained at 94 percent (91.4 percent/97.0 percent)' of the white rate. Bradshaw eventually submitted retention rate analyses to the IDHR and to the Illinois Department of Central Management Services. Neither department disapproved the layoff plan.
The first round of layoff notices was sent on January 31, 1985 — these layoffs are not directly at issue here. The layoffs challenged in this lawsuit took place between April 15 and August 5 of the same year. In the course of the challenged layoffs IDES laid off 217 employees; 130 were black.
All of the employees affected by the layoffs are members of Council 31 of the American Federation of State, County and Municipal Employees, AFL-CIO (Council 31 or the Union). In March 1985, the Union sued IDES in the Circuit Court of Cook County to enjoin the layoffs. Among other complaints, the Union argued that the layoffs were racially discriminatory. This suit was eventually dismissed for want of prosecution after the Union decided to proceed through the EEOC instead. Council 31 also filed various grievances under the collective bargaining agreement, which were eventually settled.
The present lawsuit was brought by Council 31 and by a separate class consisting solely of the black IDES employees affected by the 1985 layoffs.’ The complaint included claims of intentional discrimination brought under Title VII (42 U.S.C. §§ 2000e et seq. (1988)) and under 42 U.S.C. §§ 1981 & 1983 (1988). The complaint also presented a claim of -disparate impact under Title VII.
The case was first assigned to Judge Leighton of the Northern District of Illinois. Judge Leighton dismissed the claims under sections 1981 and 1983 for failure to allege “well pleaded facts that would give rise to an inference that the layoff procedures were the result of intentional discrimination.” Mem.Op. at 5, 1987 WL 4915 (April 27, 1987). The judge appears to have overlooked the fact that the plaintiffs also brought a claim of intentional discrimination under Title VII. Nonetheless, the parties have been assuming that the claim of intentional discrimination under Title VII was dismissed as well.2
The plaintiffs amended their complaint, and the parties proceeded with the litigation on a disparate impact theory before Judge Plunkett, to whom the case was reassigned. After a certain amount of procedural skirmishing, Judge Plunkett granted the defendants’ motion for summary judgment. The court reasoned that a .disparate impact case requires the identification of a “specific employment practice.” In such a case, an “employment practice” must be a “repeated, customary method of operation”: a single layoff decision does not count. Mem.Op. and Order, 771 F.Supp. 247, 251. (July 30, 1991) (hereinafter Mem.Op.). Therefore the plaintiffs had failed to establish a prima facie case of disparate impact. Mem.Op. at 249, 251. The plaintiffs appealed, and the defendants cross-appealed. We reverse as to the direct appeal and affirm as to the cross-appeal.
II.
We begin with the issues presented by the plaintiffs’ appeal.
A. Pleading Intentional Discrimination
The Federal Rules of Civil Procedure establish a system of notice pleading. In addition to a jurisdictional statement and a demand for judgment, a complaint need present only “a short and plain statement of the claim showing that the pleader is entitled to relief.” Fed.R.Civ.P. 8(a)(2). *377Well-pleaded facts are not required. American Nurses Ass’n v. Illinois, 783 F.2d 716, 723 (7th Cir.1986).
In their original complaint, the plaintiffs identified the layoffs at issue and then alleged that “Defendants knowingly, intentionally and maliciously discriminated against Negro employees when effectuating said layoffs____” Complaint 1112. This was “quite enough to state a claim.” American Nurses, 783 F.2d at 724.
B. Disparate Impact
As the district court noted, there is “precious little” case law on the meaning of “employment practice” as that term is used in disparate impact cases. Mem.Op., 771 F.Supp. at 250. Parties may not have litigated the issue, however, because the meaning seems fairly clear. The relevant portion of the statute that authorizes disparate impact suits, section 703 of the Civil Rights Act of 1964, reads as follows:
(a) It shall be an unlawful employment practice for an employer—
(1) to fail or refuse to hire or to discharge any individual, or otherwise to discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual’s race, color, religion, sex, or national origin; or
(2) to limit, segregate, or classify his employees or applicants for employment in any way which would deprive or tend to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual’s race, color, religion, sex, or national origin.
42 U.S.C. § 2000e-2(a); see Griggs v. Duke Power Co., 401 U.S. 424, 91 S.Ct. 849, 28 L.Ed.2d 158 (1971).
As used in the statute, the term “employment practice” must refer to any of the enumerated acts, most of which could be characterized as single decisions of an employer. There is no question that single decisions are actionable under this same section when consciously motivated by race. See, e.g., Artis v. Hitachi Zosen Clearing, Inc., 967 F.2d 1132 (7th Cir.1992) (single recall decision). It is difficult to. see why the result should be any different when the decision is challenged on the grounds of disparate impact. Certainly there are appellate cases dealing with disparate impact challenges to single decisions of employers. For instance, we have found two appellate opinions dealing with the disparate impact of layoffs that may be characterized as challenges to single decisions of employers. Sengupta v. Morrison-Knudsen Co., 804 F.2d 1072, 1073-74 (9th Cir.1986) (company decided to use performance ratings based on subjective criteria for purposes of single layoff); Nolting v. Yellow Freight System; Inc., 799 F.2d 1192, 1194 (8th Cir.1986) (disparate impact case under Age Discrimination in Employment Act; decision to use performance ratings for single layoff decision). Challenges to testing procedures may also be characterized as single decision cases, especially when the test is administered only once. See, e.g., Bridgeport Guardians, Inc. v. Bridgeport, 933 F.2d 1140, 1142-43 (2d Cir.), cert. denied, — U.S. —, 112 S.Ct. 337, 116 L.Ed.2d 277 (1991). Even Griggs, the prototypical disparate impact case, could be characterized as a single decision case — at some point, someone in the management of Duke Power decided to impose a high school diploma requirement for transfers to other departments. 401 U.S. at 427, 91 S.Ct. at 851. In some sense, the distinction drawn by the district court is analytically unmanageable — almost any repeated course of conduct can be traced back to a single decision.
The district court relied on our decision in Beard v. Whitley County REMC, 840 F.2d 405 (7th Cir.1988), to support its corn elusion. Mem.Op., 771 F.Supp. at 250-51. In Beard, however, this court decided only that an employer does not engage in an “employment practice” by inaction. In Beard the challenged disparity between men’s and women’s compensation packages was the result of separate negotiations with the two unions that represented male and female employees. 840 F.2d at 409. Since there was no evidence that the employer had any hand in the division of its *378workforce into primarily male and primarily female. bargaining units, we held that the disparities were not the result of a “practice.” Id.; see also EEOC v. Chicago Miniature Lamp Works, 947 F.2d 292, 305 (7th Cir.1991) (company’s passive reliance on word-of-mouth to fill occasional vacancies did not constitute “employment practice”). Notably, although we held that there was no “employment practice” to be challenged, the employer’s “practice” of separate negotiations was its regular course of conduct. Beard, 840 F.2d at 406 (“Historically, the two work forces have negotiated their own ‘group-wide’ agreements concerning wage and benefit increases.”). Under the district court’s analysis here, the panel in Beard should have concluded that the employer had engaged in an employment practice; obviously enough, the panel did not do so.
The district court’s opinion (and the argument of IDES) appears to reflect some misapprehension about the difference between an “employment practice” prohibited by section 703 and a “pattern or practice or resistance to the full enjoyment of any of the rights secured by this subchapter” that authorizes a civil enforcement suit by the Attorney General under section 707 of Title VII. 42 U.S.C. § 2000e-6(a) (1988). As the Supreme Court explained in International Bhd. of Teamsters v. United States, 431 U.S. 324, 97 S.Ct. 1843, 52 L.Ed.2d 396 (1977), a “pattern or practice” suit is a method of proving discrimination: by showing that an employer regularly and purposefully discriminates against a protected group, a plaintiff may shift the burden of proof to the employer to demonstrate that it did not discriminate against a particular member of that group. Id. at 360-63, 97 S.Ct. at 1867-69; see also King v. General Electric Co., 960 F.2d 617, 628 (7th Cir. 1992) (individual plaintiffs may bring “pattern or practice” action). This means of proving intentional discrimination is distinct from a disparate impact case, where the plaintiff need not make any showing of the employer’s intent. Teamsters, 431 U.S. at 335-36 n. 15, 97 S.Ct. at 1854-55 n. 15 (contrasting claims of disparate treatment, including pattern-or-practice cases, with claims of disparate impact); see also Chicago Miniature, 947 F.2d at 297 (noting potential for confusion between disparate treatment and disparate impact theories when a pattern or practice of discrimination is alleged).
Finally, the district court was concerned that if single employer decisions are actionable then “every single act, intentional or not, which has an adverse impact on a protected class is actionable under Title VII.” Mem.Op. 771 F.Supp. at 251. But we fail to see the obvious evil in the court’s hypothesis. Intentionally discriminatory acts are of course actionable under Title VII, single or otherwise. As for unintentionally discriminatory acts that have an adverse impact on a protected class, they too are actionable under Title VII unless they have a business justification. But it is difficult to make out a prima facie showing of adverse impact: the affected group may be too small for any valid statistical comparisons, Watson v. Ft. Worth Bank & Trust, 487 U.S. 977, 996-97, 108 S.Ct. 2777, 2790, 101 L.Ed.2d 827 (1988) (plurality opinion); id. at 1001 n. 2, 108 S.Ct. at 2792 n. 2 (Blackmun, J., concurring in part) (agreeing with plurality’s discussion of issue), and obvious or common non-discriminatory reasons for the apparent disparities must be taken into account. Mozee v. American Commercial Marine Serv. Co., 940 F.2d 1036, 1045 (7th Cir.1991); Allen v. Seid-man, 881 F.2d 375, 378 (7th Cir.1989); Coates v. Johnson & Johnson, 756 F.2d 524, 540 (7th Cir.1985). These requirements immunize most single decisions from disparate impact challenges. To the extent that members of a protected class can show significant disparities stemming from a single decision, however, there is no reason that decision should not be actionable. It is worth noting that IDES regularly prepares disparate impact analyses on proposed layoffs. State regulations require that IDHR and the Department of Central Management Services review layoffs for disparate impact. Ill.Admin.Code § 302.-520 (1985). This lawsuit cannot have come as a surprise to IDES.
*379In sum, we conclude that the plaintiffs stated a claim for intentional racial discrimination and, at least as to the present issue, established a prima facie case of disparate impact.
III.
The IDES defendants argue that there are several other grounds on which we can and should affirm the grant of summary judgment. Reed v. AMAX Coal Co., 971 F.2d 1295, 1298 (7th Cir.1992) (appellate court may affirm on any ground finding support in the record).
A. Failure of Proof of Disparate Impact
IDES argues that we should affirm because the plaintiffs’ statistics were insufficient to support their claim. For the most part, this argument is based on a mischaracterization of the plaintiffs’ case. IDES notes that the statistics presented to the district court are based on the racial makeup of IDES's state-wide staff. This analysis omits “critical factors and variables,” IDES argues, because it fails to take the concentration of black employees in the Chicago area into account. But the plaintiffs are not claiming (or at least we do not understand them to claim) that the layoffs within Chicago were discriminatory. Instead, they are claiming that the initial decision to concentrate the layoffs in Chicago was discriminatory, precisely because black employees are concentrated there.
Another argument along these lines has • more merit. IDES argues that its retention rate analysis is the right approach to this case and that the retention rate analysis shows conclusively that the layoffs did not have a disparate impact. The plaintiffs insist on the layoff rate analysis. The Supreme Court has told us, however, that no particular mathematical- formula can be adopted for disparate impact claims. Cases must be evaluated on their own terms. Watson, 487 U.S. at 995-96 n. 3, 108 S.Ct. at 2789-90 n. 3. Accordingly, we think it appropriate to give the district court the first opportunity to decide what method of . proof is most appropriate and whether the plaintiffs’ statistics are adequate.3
B. Bona Fide Seniority System
IDES argues that it followed the dictates of a bona fide seniority system in implementing the layoffs. This argument is based on the same mischaracterization of the plaintiffs’ claim that inspires IDES’s challenge to the plaintiffs’ use of statewide employment figures. It may be true that IDES implemented the layoffs in Chicago in accordance with the seniority system, but again the issue is not how the layoffs were handled in Chicago but rather how Ward chose the Chicago offices to bear their brunt.
C. Waiver by Settlement
IDES contends that Council 31 waived its right to pursue this lawsuit. The circumstances are as follows. In addition to the 1985 layoffs at issue in this case, Council 31 also challenged layoffs that IDES planned to implement in 1986. Council 31 and the state settled this dispute with a document that states: “The Union agrees to refrain from filing any court action and appeals to arbitration concerning the propriety of the layoff.” Mem. of Understanding between the Dep’t of Cen*380tral Management Servs and AFSCME Council 31 at 2 (effective April 23, 1986) (R. 94, Ex. J). Contrary to IDES’s assertion, there is no indication that “the layoff” referred to is the 1985 layoff. Indeed, the substantive provisions of the Memorandum apply by their terms only to the 1986 layoffs. Id. at 3. And an addendum to the same agreement expressly reserves Council 31’s right to pursue the lawsuit now before us. Addendum to the Mem. of Understanding between the Dep’t of Central Management Servs and AFSCME Council 31 at 2 (undated) (R. 94, Ex. J). We have difficulty understanding how counsel could argue that the Union waived its claims based on the 1985 layoffs.
D. Preclusion Arguments
IDES suggests two more reasons for affirming the grant of summary judgment in whole or in part: 1) the Circuit Court of Cook County decided that the layoffs were not discriminatory and 2) one member of the plaintiff class has already sued IDES about the layoffs and lost, precluding him from participation in this lawsuit. These arguments have no merit. The Circuit Court of Cook County never gave judgment on the merits of the Union’s claims, and McCalpine, the individual in question, challenged IDES’s failure to transfer him to another position, not the layoff. McCal-pine v. Foertsch, 870 F.2d 409 (7th Cir. 1989).
IV.
We turn now to the cross-appeal. The defendants challenge the order certifying a class, alleging that the class representatives are inadequate. They also appeal from Judge Plunkett’s refusal to prevent counsel for the class from working with counsel for the Union. The cross-appeal raises an interesting jurisdictional issue.
A class certification order is interlocutory. Under normal circumstances, it may not be appealed until there is a final judgment. Coopers & hybrand v. Livesay, 437 U.S. 463, 476, 98 S.Ct. 2454, 2462, 57 L.Ed.2d 351 (1978). Orders involving the qualification and conduct of class attorneys are similarly interlocutory, and, therefore, are ordinarily beyond this court’s jurisdiction until a final judgment is rendered. In re Fine Paper Antitrust Litigation, 617 F.2d 22, 26-28 (3d Cir.1980); see generally 15B Charles A. Wright et al., Federal Practice and Procedure § 3914.19 at 76-77 (1992) (collecting cases).
Although there was a final judgment in this case, the IDES was initially barred from appealing to this court because the final judgment in the district court was in its favor and one cannot appeal from a victory. See Turner v. Chicago Housing Authority, 969 F.2d 461, 463 (7th Cir.1992). This court, however, together with most of the other circuits, treats conditional cross-appeals differently from unconditional appeals. Nominally prevailing parties are entitled to file such cross-appeals against the contingency that this court will reverse an otherwise thoroughly satisfactory judgment. See, e.g., Hartford Acc. and Indem. Co. v. Sullivan, 846 F.2d 377, 385 (7th Cir.1988); see generally 15A Charles A. Wright et al., Federal Practice and Procedure § 3902 at 78-79 (1992) (collecting cases). Because we have reversed the district court, the IDES has indeed become adversely affected by the district court’s orders certifying the class and permitting consultation between counsel for the class and counsel for the Union. This development has the effect of invigorating the cross-appeal and supporting our jurisdiction. Accordingly, we reach the merits of the IDES’s contentions.
Appellate review of class certification decisions is deferential, and this court will not reverse an order certifying a class absent an abuse of the district court’s discretion. Rand v. Monsanto Co., 926 F.2d 596, 598 (7th Cir.1991); Susman v. Lincoln Am. Corp., 561 F.2d 86, 90 (7th Cir.1977). In light of the district court’s explicit findings that the class representatives were committed to the litigation and aware of their responsibilities, see Mem.Op. and Order, No. 87 C 356,1990 WL 103226 (N.D.Ill. July 16, 1990) (order certifying class), we affirm the class certification as within the *381district court’s discretion. We are similarly persuaded that the district court did not abuse its discretion in permitting class and Union counsel to cooperate with and assist each other to the extent their cases against the IDES are coextensive.
V.
For the foregoing reasons, in appeal number 91-2980, the judgment of the district court is Reversed and the cause is Remanded for further proceedings not inconsistent with this opinion. In the defendants’ cross-appeal, number 91-3077, the judgment is Affirmed.
. Loleta Didrickson has succeeded Ms. Ward as Director of the IDES. Under Fed.R.App.P. 43(c), Didrickson automatically substitutes for Ward to the extent that Ward was sued in her official capacity. Furthermore, appellants conceded at oral argument that Ms. Ward is no longer a proper defendant in this case. Accordingly, appellants’ complaint is dismissed to the extent it alleges causes of action against Ms. Ward in her individual capacity.
. This assumption raises some interesting jurisdictional issues which we need not reach.
. At this early juncture, we find it difficult to believe that either the retention rate analysis or the layoff rate analysis is appropriate. Mathematically speaking, retention rate analysis and layoff rate analysis are identical, like describing a glass half-full or half-empty. Preferring one approach over another makes little sense. The important questions would seem to be: (1) what percentage of employees laid off in a random selection would be black; and (2) what are the confidence intervals surrounding that outcome? If the deviation from the expected outcome of the actual percentage of blacks laid off is statistically significant, it may be possible to infer that the layoff decision caused a disparate impact and that such an impact is not mere happenstance. We note, in addition, that in defining the relevant population, the parties must be mindful of the Supreme Court's holding in Firefighters v. Stotts, 467 U.S. 561, 104 S.Ct. 2576, 81 L.Ed.2d 483 (1984) that Title VII protects bona fide seniority systems. Answering the questions we pose would seem to require a more sophisticated statistical analysis than that presented by the plaintiffs.