delivered the opinion of the Court.
Respondents are three black bricklayers who sought employment with petitioner Furnco Construction Corp. Two of the three were never offered employment. The third was employed only long after he initially applied. Upon adverse findings entered after a bench trial, the District Court for the Northern District of Illinois held that respondents had not proved a claim under either the “disparate treatment” theory of McDonnell Douglas Corp. v. Green, 411 U. S. 792 (1973), or the “disparate impact” theory of Griggs v. Duke Power Co., 401 U. S. 424 (1971). The Court of Appeals for the Seventh Circuit, concluding that under McDonnell Douglas respondents had made out a prima facie case which had not been effectively rebutted, reversed the judgment of the District Court. 551 F. 2d 1085 (1977). We granted certiorari to consider important questions raised by this case regarding the exact scope of the prima facie case under McDonnell Douglas and the nature of the evidence necessary to rebut such a case. 434 U. S. 996 (1977). Having concluded that the Court of Appeals erred in its treatment of the latter question, we reverse and remand to that court for further proceedings consistent with this opinion.
I
A few facts in this case are not in serious dispute. Petitioner Furnco, an employer within the meaning of §§ 701 (b) and (h) of Title VII of the 1964 Civil Rights Act, 42 U. S. C. §§2000e(b) and (h) (1970 ed., Supp. V), specializes in refractory installation in steel mills and, more particularly, the rehabilitation or relining of blast furnaces with what is called in the trade “firebrick.” Furnco does not, however, maintain a permanent force of bricklayers. Rather, it hires a superintendent for a specific job and then delegates to him *570the task of securing a competent work force. In August 1971, Furnco contracted with Interlake, Inc., to reline one of its blast furnaces. Joseph Dacies, who had been a job superintendent for Furnco since 1965, was placed in charge of the job and given the attendant hiring responsibilities. He did not accept applications at the jobsite, but instead hired only-persons whom he knew to be experienced and competent in this type of work or persons who had been recommended to him as similarly skilled. He hired his first four bricklayers, all of whom were white, on two successive days in August, the 26th and 27th, and two in September, the 7th and 8th. On September 9 he hired the first black bricklayer. By September 13, he had hired 8 more bricklayers, 1 of whom was black; by September 17, 7 more had been employed, another of whom was black; and by September 23, 17 more were on the payroll, again with 1 black included in that number.1 From October 12 to 18, he hired 6 bricklayers, all of whom were black, including respondent Smith, who had worked for Dacies previously and had applied at the jobsite somewhat earlier. Respondents Samuels and Nemhard were not hired, though they were fully qualified and had also attempted to secure employment by appearing at the jobsite gate. Out of the total of 1,819 man-days worked on the Interlake job, 242, or 13.3%-, were worked by black bricklayers.
Many of the remaining facts found by the District Court and the inferences to be drawn therefrom are in some dispute between the parties, but none was expressly found by the Court of Appeals to be clearly erroneous. The District Court elaborated at some length as to the “critical” necessity of insuring that only experienced and highly qualified fire-*571bricklayers were employed. Improper or untimely work would result in substantial losses both to Interlake, which was forced to shut down its furnace and lay off employees during the relining job, and to Furnco, which was paid for this work at a fixed price and for a fixed time period. In addition, not only might shoddy work slow this work process down, but it also might necessitate costly future maintenance work with its attendant loss of production and employee layoffs; diminish Furnco’s reputation and ability to secure similar work in the future; and perhaps even create serious safety hazards, leading to explosions and the like. App. to Pet. for Cert. A13-A15. These considerations justified Furnco’s refusal to engage in on-the-job training or to hire at the gate, a hiring process which would not provide an adequate method of matching qualified applications to job requirements and assuring that the applicants are sufficiently skilled and capable. Id., at A18-A19. Furthermore, there was no evidence that these policies and practices were a pretext to exclude black bricklayers or were otherwise illegitimate or had a disproportionate impact or effect on black bricklayers. Id., at A17-A18. From late 1969 through late 1973, 5.7% of the bricklayers in the relevant labor force were minority group members, see 41 CFR § 60-11 et seq. (1977),2 while, as mentioned before, *57213.3% of the man-days on Furnco’s interlake job were worked by black bricklayers.
Because of the above considerations and following the established practice in the industry, most of the firebricklayers hired by Dacies were persons known by him to be experienced and competent in this type of work. The others were hired after being recommended as skilled in this type of work by his general foreman, an employee (a black), another Fumco superintendent in the area, and Furnco’s General Manager John Wright. Wright had not only instructed Dacies to employ, as far as possible, at least 16% black bricklayers, a policy due to Furnco’s self-imposed affirmative-action plan to insure that black bricklayers were employed by Furnco in Cook County in numbers substantially in excess of their percentage in the local union,3 but he had also recommended, in an effort to show good faith, that Dacies hire several specific bricklayers, who had previously filed a discrimination suit against Furnco, negotiations for the settlement of which had only recently broken down, see n. 3, supra.
From these factual findings, the District Court concluded that respondents had failed to make out a Title VII claim under the doctrine of Griggs v. Duke Power Co., 401 U. S. 424 (1971). Furnco’s policy of not hiring at the gate was racially neutral on its face and there was no showing that it had a disproportionate impact or effect. App. to Pet. for Cert. A20-A21. It also held that respondents had failed to *573prove a case of discrimination under McDonnell Douglas Corp. v. Green, 411 U. S. 792 (1973). App. to Pet. for Cert. A21. It is not entirely clear whether the court thought respondents had failed to make out a prima facie case of discrimination under McDonnell Douglas, see App. to Pet. for Cert. A20-A21, but the court left no doubt that it thought Furnco’s hiring practices and policies were justified as a “business necessity” in that they were required for the safe and efficient operation of Furnco’s business, and were “not used as a pretext to exclude Negroes.” Thus, even if a prima facie case had been made out, it had been effectively rebutted. Id., at A21.
“Not only have Plaintiffs entirely failed to establish that Furnco’s employment practices on the Interlake job discriminated against them on the basis of race or constituted retaliatory conduct but Defendant has proven what it was not required to. By its cross-examination and direct evidence, Furnco has proven beyond all reasonable doubt that it did not engage in either racial discrimination or retaliatory conduct in its employment practices in regard to bricklayers on the Interlake job.” 4 Id., at A22.
The Court of Appeals reversed, holding that respondents had made out a prima facie case under McDonnell Douglas, supra, at 802, which Furnco had not effectively rebutted. Because of the “historical inequality of treatment of black workers” 5 and the fact that the record failed to reveal that *574any white persons had applied at the gate, the Court of Appeals rejected Furnco’s argument that discrimination had not been shown because a white appearing at the jobsite would have fared no better than respondents. That court also disagreed with Furnco’s contention, which the District Court had adopted, that “the importance of selecting people whose capability had been demonstrated to defendant’s brick superintendent is a ‘legitimate, nondiscriminatory reason’ for defendant’s refusal to consider plaintiffs.” 551 F. 2d, at 1088. Instead, the appellate court proceeded to devise what it thought would be an appropriate hiring procedure for Furnco, saying that “[i]t seems to us that there is a reasonable middle ground between immediate hiring decisions on the spot and seeking out employees from among those known to the superintendent.” Ibid. This middle course, according to the Court of Appeals, was to take written applications, with inquiry as to qualifications and experience, and then check, evaluate, and compare those claims against the qualifications and experience of other bricklayers with whom the superintendent was already acquainted. We granted certiorari to consider whether'the Court of Appeals had gone too far in substituting its own judgment as to proper hiring practices in the case of an employer which claimed the practices it had chosen did not violate Title VII.6
*575II
A
We agree with the Court of Appeals that the proper approach was the analysis contained in McDonnell Douglas, supra.7 We also think the Court of Appeals was justified in concluding that as a matter of law respondents made out a prima facie case of discrimination under McDonnell Douglas. In that case we held that a plaintiff could make out a prima facie claim by showing
“(i) that he belongs to a racial minority; (ii) that he applied and was qualified for a job for which the employer was seeking applicants; (iii) that, despite his qualifications, he was rejected; and (iv) that, after his rejection, the position remained open and the employer continued to seek applicants from persons of complainant’s qualifications.” 411 U. S., at 802 (footnote omitted).
This, of course, was not intended to be an inflexible rule, as the Court went on to note that “[t]he facts necessarily will vary in Title VII cases, and the specification ... of the prima facie proof required from respondent is not necessarily appli*576cable in every respect to differing factual situations.” Id., at 802 n. 13. See Teamsters v. United States, 431 U. S. 324, 358 (1977). But McDonnell Douglas did make clear that a Title VII plaintiff carries the initial burden of showing actions taken by the employer from which one can infer, if such actions remain unexplained, that it is more likely than not that such actions were “based on a discriminatory criterion illegal under the Act.” 431 U. S., at 358. See also id., at 335 n. 15. And here respondents carried that initial burden by proving they were members of a racial minority; they did everything within their power to apply for employment; Furnco has conceded that they were qualified in every respect for the jobs which were about to be open;8 they were not offered employment, although Smith later was; and the employer continued to seek persons of similar qualifications.
B
We think the Court of Appeals went awry, however, in apparently equating a prima facie showing under McDonnell Douglas with an ultimate finding of fact as to discriminatory refusal to hire under Title VII; the two are quite different and that difference has a direct bearing on the proper resolution of this case. The Court of Appeals, as we read its opinion, thought Furnco’s hiring procedures not only must be reasonably related to the achievement of some legitimate purpose, but also must be the method which allows the employer to consider the qualifications of the largest number of minority applicants. We think the imposition of that second *577requirement simply finds no support either in the nature of the prima facie case or the purpose of Title VII.
The central focus of the inquiry in a case such as this is always whether the employer is treating “some people less favorably than others because of their race, color, religion, sex, or national origin.” Teamsters v. United States, supra, at 335 n. 15. The method suggested in McDonnell Douglas for pursuing this inquiry, however, was never intended to be rigid, mechanized, or ritualistic. Rather, it is merely a sensible, orderly way to evaluate the evidence in light of common experience as it bears on the critical question of discrimination. A prima facie case under McDonnell Douglas raises an inference of discrimination only because we presume these acts, if otherwise unexplained, are more likely than not based on the consideration of impermissible factors. See Teamsters v. United States, supra, at 358 n. 44. And we are willing to presume this largely because we know from our experience that more often than not people do not act in a totally arbitrary manner, without any underlying reasons, especially in a business setting. Thus, when all legitimate reasons for rejecting an applicant have been eliminated as possible reasons for the employer’s actions, it is more likely than not the employer, who we generally assume acts only with some reason, based his decision on an impermissible consideration such as race.
When the prima facie case is understood in the light of the opinion in McDonnell Douglas, it is apparent that the burden which shifts to the employer is merely that of proving that he based his employment decision on a legitimate consideration, and not an illegitimate one such as race. To prove that, he need not prove that he pursued the course which would both enable him to achieve his own business goal and allow him to consider the most employment applications. Title VII prohibits him from having as a goal a work force selected by any proscribed discriminatory practice, but it does not impose a duty to adopt a hiring procedure that maximizes hiring of *578minority employees. To dispel the adverse inference from a prima facie showing under McDonnell Douglas, the employer need only “articulate some legitimate, nondiscriminatory reason for the employee’s rejection.” 411 U. S., at 802.
The dangers of embarking on a course such as that charted by the Court of Appeals here, where the court requires businesses to adopt what it perceives to be the “best” hiring procedures, are nowhere more evident than in the record of this very case. Not only does the record not reveal that the court’s suggested hiring procedure would work satisfactorily, but also there is nothing in the record to indicate that it would be any less “haphazard, arbitrary, and subjective” than Fumco’s method, which the Court of Appeals criticized as deficient for exactly those reasons. Courts are generally less competent than employers to restructure business practices, and unless mandated to do so by Congress they should not attempt it.
This is not to say, of course, that proof of a justification which is reasonably related to the achievement of some legitimate goal necessarily ends the inquiry. The plaintiff must be given the opportunity to introduce evidence that the proffered justification is merely a pretext for discrimination. And as we noted in McDonnell Douglas, supra, at 804-805, this evidence might take a variety of forms. But the Court of Appeals, although stating its disagreement with the District Court’s conclusion that the employer’s hiring practices were a “legitimate, nondiscriminatory reason” for refusing to hire respondents, premised its disagreement on a view which we have discussed and rejected above. It did not conclude that the practices were a pretext for discrimination, but only that different practices would have enabled the employer to at least consider, and perhaps to hire, more minority employees. But courts may not impose such a remedy on an employer at least until a violation of Title VII has been proved, and here none had been under the reasoning of either the District Court or the Court of Appeals.
*579c
The Court of Appeals was also critical of petitioner’s effort to employ statistics in this type of case. While the matter is not free from doubt, it appears that the court thought that once a McDonnell Douglas prima facie showing had been made out, statistics of a racially balanced work force were totally irrelevant to the question of motive. See 551 F. 2d, at 1089. That would undoubtedly be a correct view of the matter if the McDonnell Douglas prima facie showing were the equivalent of an ultimate finding by the trier of fact that the original rejection of the applicant was racially motivated: A racially balanced work force cannot immunize an employer from liability for specific acts of discrimination. As we said in Teamsters v. United States, 431 U. S., at 341-342:
“[T]he District Court and the Court of Appeals found upon substantial evidence that the company had engaged in a course of discrimination that continued well after the effective date of Title VII. The company’s later changes in its hiring and promotion policies could be of little comfort to the victims of the earlier post-Act discrimination, and could not erase its previous illegal conduct or its obligation to afford relief to those who suffered because of it.”
See also Albemarle Paper Co. v. Moody, 422 U. S. 405, 412-413 (1975.) It is clear beyond cavil that the obligation imposed by Title VII is to provide an equal opportunity for each applicant regardless of race, without regard to whether members of the applicant’s race are already proportionately represented in the work force. See Griggs v. Duke Power Co., 401 U. S., at 430; McDonald v. Santa Fe Trail Transportation Co., 427 U. S. 273, 279 (1976).
A McDonnell Douglas prima facie showing is not the equivalent of a factual finding of discrimination, however. Rather, it is simply proof of actions taken by the employer from which *580we infer discriminatory animus because experience has proved that in the absence of any other explanation it is more likely than not that those actions were bottomed on impermissible considerations. When the prima facie showing is understood in this manner, the employer must be allowed some latitude to introduce evidence which bears on his motive. Proof that his work force was racially balanced or that it contained a disproportionately high percentage of minority employees is not wholly irrelevant on the issue of intent when that issue is yet to be decided. We cannot say that such proof would have absolutely no probative value in determining whether the otherwise unexplained rejection of the minority applicants was discriminatorily motivated. Thus, although we agree with the Court of Appeals that in this case such proof neither was nor could have been sufficient to conclusively demonstrate that Furnco’s actions were not discriminatorily motivated, the District Court was entitled to consider the racial mix of the work force when trying to make the determination as to motivation. The Court of Appeals should likewise give similar consideration to the proffered statistical proof in any further proceedings in this case.
Ill
The parties also press upon the Court a large number of alternative theories of liability and defense,9 none of which were directly addressed by the Court of Appeals as we read its opinion. Given the present posture of this case, however, *581we think those matters which are still preserved for review are best decided by the Court of Appeals in the first instance. Accordingly, we decline to address them as an original matter here. The judgment of the Court of Appeals is reversed, and the case is remanded for further proceedings consistent with this opinion.
It is so ordered.
Respondents contend that two of these four blacks were not actually “hired,” but merely “transferred” from another Furnco job. Brief for Respondents 7-8. Both the District Court and the Court of Appeals spoke only of “hiring” bricklayers, however, and those parts of the record to which respondents point do not persuade us that this is a mischaracterization.
Respondents attempted to introduce a study conducted in late 1973 by the local union which matched members’ names and race in an effort to show what percentage of the union membership was black. The study concluded that approximately 500 of the 3,800 union members were black. The District Court excluded this evidence because the study had been conducted two years after Furnco completed its job. App. to Pet. for Cert. A16 n. 1. The Court of Appeals thought rejection of this evidence was an abuse of discretion, but in dealing with the merits did not rely on the racial proportions in the labor force, so did not remand the ease to permit introduction of that testimony. The Court of Appeals also noted that in any event respondents suffered no prejudice by the court’s refusal to admit the study because it would not have demonstrated discrimination. The study showed that 13.7% of the membership of the union was black, *572while the evidence demonstrated that 13.3% of the man-days were worked by black bricklayers, Furnco had set a goal of 16% black bricklayers, and 20% of the individuals hired were black. 551 F. 2d 1085, 1090 (1977).
According to the District Court, this affirmative-action program was initiated by Fumco following a job performed in 1969-1970 from which charges of racial discrimination in hiring were filed by several black bricklayers. These claims are apparently still pending on appeal in the Illinois courts and the merits of a parallel federal action remain to be adjudicated. See App. to Pet. for Cert. A15; Batiste v. Furnco Construction Corp., 503 F. 2d 447 (CA7 1974).
The District Court also found that certain other plaintiffs never attempted to apply for work at Interlake or were fired or not hired for valid reasons, such as insubordination or poor workmanship. App. to Pet. for Cert. A17-A19. The Court of Appeals, concluding that the District Court’s findings were not clearly erroneous, affirmed the judgment against these particular plaintiffs. 551 F. 2d, at 1087-1088. These rulings are not challenged here.
The court stated:
“The historical inequality of treatment of black workers seems to us to *574establish that it is prima jade racial discrimination to refuse to consider the qualifications of a black job seeker before hiring from an approved list containing only the names of white bricklayers. How else will qualified black applicants be able to overcome the racial imbalance in a particular craft, itself the result of past discrimination?” 551 F. 2d, at 1089.
The petition for certiorari set out three questions:
“1. Whether the Seventh Circuit, in reversing the judgment of the District Court, erred in finding as irrelevant to the issue of racial discrimination in hiring, statistics demonstrating that in hiring highly skilled bricklayers, the employer hired Negroes in a percentage far in excess of their statistical presence in the relevant labor force.
“2. Whether a court may find an employer guilty of racial discrimination *575in employment due to alleged disparate treatment in hiring without a finding of discriminatory intent or motive.
“3. Whether a hiring practice not shown to result in disparate impact or treatment of prospective minority employees and found by the District Court to be justified by business necessity and legitimate business reasons may be found to be racially discriminatory by the Court of Appeals merely because it is subjective and because the Court of Appeals substitutes its judgment for that of the District Court as to what constitutes legitimate business reasons.” Pet. for Cert. 2.
This case did not involve employment tests, which we dealt with in Griggs v. Duke Power Co., 401 U. S. 424 (1971), and in Albemarle Paper Co. v. Moody, 422 U. S. 405, 412-413 (1975), or particularized requirements such as the height and weight specifications considered in Dothard v. Rawlinson, 433 U. S. 321, 329 (1977), and it was not a “pattern or practice” case like Teamsters v. United States, 431 U. S. 324, 358 (1977).
We note that this case does not raise any questions regarding exactly what sort of requirements an employer can impose upon any particular job. Furnco has conceded that for all its purposes respondents were qualified in every sense. Thus, with respect to the McDonnell Douglas prima facie case, the only question it places in issue is whether its refusal to consider respondents’ applications at the gate was based upon legitimate, nondiscriminatory reasons and therefore permissible.
Respondents, for example, argue that regardless of the propriety of Furneo’s general refusal to hire at the gate or of a general policy of hiring only bricklayers known to the superintendent or referred to him by an insider, a foreman, or another bricklayer, Dacies’ particular method of hiring was discriminatory. Thus, the general hiring practice, though perhaps legitimate in the abstract, was discriminatorily applied in this case, and cannot be used to rebut the prima facie case. Brief for Respondents 19-26. In particular, respondents argue that the evidence proved that Dacies hired from a “list” he had prepared, which allegedly included *581competent firebricklayers with whom he had worked, but in fact included only white firebricklayers with whom he had worked. Exclusion from this list of competent blacks with whom he had worked, such as respondents Smith and Samuels, was itself discriminatory and thus cannot be used to rebut respondents’ prima facie case.
Fumco, on the other hand, vigorously disputes that Dacies hired only from this list and that the hiring process can be as neatly broken down into various components as respondents would like. It argues that even if most of the people with whom Dacies was familiar were white, Dacies made a concerted effort to speak with people who were familiar with competent black bricklayers and then hired a large number of black bricklayers. In fact, argues Fumco, the statistics indicate that he hired a disproportionately large number of blacks, thus clearly indicating that his so-called “list” certainly could not have been the exclusive source of potential employees even if it had been all white. It further disputes the notion that Fumco or Dacies had in any way put some sort of ceiling on the maximum number of blacks they were willing to hire. It asserts there is absolutely nothing in the record to support such a conclusion.
The District Court made no findings which would support respondents’ view of the evidence. The Court of Appeals mentioned the existence of such a list, 551 F. 2d, at 1086, but we do not read its opinion as expressly relying on this point either. Rather, as we read its opinion, the court found only that respondents had made out a prima facie case under McDonnell Douglas and that, for the reasons outlined in the text, Furnco had failed to rebut that prima facie case. On remand, respondents are of course free to pursue any such contentions which have been properly preserved.