concurring in part and dissenting in part.
I join the opinion of the court with respect to its disposition of Silva’s motion to amend his complaint to add a claim under 29 U.S.C. § 1132(a)(3). However, because I believe that the district court properly granted summary judgment in favor of MetLife and Savvis on Silva’s claim under 29 U.S.C. § 1132(a)(1)(B), I respectfully dissent from Part III.A of the opinion of the court.
In considering a § 1132(a)(1)(B) claim, we review a plan administrator’s benefits determination for abuse of discretion when, as here, the “plan grants the administrator discretion to determine eligibility for benefits and to interpret the plan’s terms.” Green v. Union Sec. Ins. Co., 646 F.3d 1042, 1050 (8th Cir.2011). Under this standard, a plan “administrator’s decision should be affirmed if it is reasonable.” Id. (internal citations and quotations omitted). *729As the court observes, the ERISA plan in this case provides that Abel’s supplemental-life-insurance coverage would not commence if he did not give MetLife “evidence of insurability or the evidence of insurability is not accepted by [MetLife] as satisfactory.” See ante at 714-15. MetLife interpreted this provision to require Abel to submit a Statement of Health, which he did not do. The dispositive question in this case, then, is whether MetLife’s interpretation was reasonable. See ante at 718-19.
In reviewing [a plan administrator’s] interpretation of its plan language, we generally examine the following factors [often known as the Finley factors]: (1) whether the interpretation is consistent with the goals of the plan; (2) whether it renders any language in the plan meaningless or internally inconsistent; (3) whether it conflicts with the substantive or procedural requirements of ERISA; (4) whether [the plan administrator] has interpreted the provisions at issue here consistently; and (5) whether the interpretation is contrary to the clear language of the plan.
McClelland v. Life Ins. Co. of N. Am., 679 F.3d 755, 759 (8th Cir.2012) (citing Finley v. Special Agents Mut. Benefit Ass’n, 957 F.2d 617, 621 (8th Cir.1992)).14 “[W]e do not search for the best or preferable interpretation of a plan term.” Hutchins v. Champion Intern. Corp., 110 F.3d 1341, 1344 (8th Cir.1997). “[W]e may not find the interpretation invalid merely because we disagree with it, but only if it is unreasonable.” Id.
Consideration of the Finley factors shows that MetLife reasonably interpreted the plan to require submission of a Statement of Health. First, MetLife’s interpretation is consistent with the clear language of the plan. The plan requires participants to submit evidence of insurability that is “satisfactory” to MetLife. This phrasing necessarily contemplates that MetLife would require some evidence— such as a Statement of Health — that is not specified by the plan’s terms and that Met-Life would be the judge of the adequacy of that evidence. Silva has not provided any reason to doubt that requiring the submission of a Statement of Health is a fair method of assessing the health of a life-insurance applicant. Second, the record demonstrates that MetLife has applied its interpretation consistently. As the court observes, approximately 200 other Sawis employees were found not to have completed a Statement of Health. See ante at 715-16. MetLife applied the same plan interpretation to these employees as it did to Silva and required them to submit Statements of Health. See Appellant’s Appendix at 565 (containing MetLife internal claims-determination file) (“[W]e have determined that we will not be grandfathering these individuals into the plan at the amounts they’ve requested. We will require ALL past participants to submit a form.”). This consistent application supports the reasonableness of MetLife’s interpretation. Third, requiring submission of a Statement of Health also comports with at least one plan goal, that is, ensuring affordability for participants. As the court notes, the evidence-of-insurability *730provision can prevent “those who may be very ill from taking out a large life insurance policy shortly before death.” Ante at 718. By limiting MetLife’s risk exposure, such a requirement could in turn reduce premium costs for other participants. Refusing to enforce this provision in the reasonable manner selected by MetLife could impose future premium increases on other plan participants. Finally, none of the other Finley factors undermines MetLife’s interpretation. In light of these factors, which the court does not specifically analyze, I conclude that MetLife reasonably interpreted the plan to require submission of a Statement of Health, a requirement with which Abel did not comply.
The court objects that the administrative record omits certain facts that the court deems necessary to determining whether MetLife abused its discretion. But I am not convinced that these eviden-tiary lacunae are so glaring. In particular, the administrative record shows that Met-Life investigated Sawis’s method of notifying employees that submitting a Statement of Health is required. And Sawis responded that when employees attempt to enroll for benefits through its online enrollment system, the system prompted them to complete a Statement of Health and directed them to the company’s benefits department, from which the form could be obtained. See Appellant’s Appendix 573-76. Silva did not present evidence to contradict the results of MetLife’s investigation. Thus, the uncontradicted evidence in the administrative record supports the conclusion that Abel received notice that he was required to complete a Statement of Health. See Green, 646 F.3d at 1050 (explaining that a plan administrator’s benefits determination is reasonable if “it is supported by substantial evidence”). As such, I believe that the district court correctly granted summary judgment in favor of MetLife and Sawis on Silva’s § 1132(a)(1)(B) claim.
As noted above, I agree with the court that Silva was permitted to plead simultaneously claims under both § 1132(a)(1)(B) and § 1132(a)(3). The court correctly explains that at this early stage of litigation, an ERISA plaintiff may plead claims under both provisions in the alternative. See ante at 725-29. Critically, though, a § 1132(a)(3) claim — even if pled in the alternative — cannot survive a motion to dismiss if the plaintiff seeks to enforce rights that arise under the terms of an ERISA plan; § 1132(a)(1)(B) provides the exclusive remedy for vindicating rights arising under the terms of the plan. See Pilger v. Sweeney, 725 F.3d 922, 927 (8th Cir.2013). Here, however, Silva premises his § 1132(a)(3) claim on alleged breaches of the defendants’ fiduciary duties that prevented him from becoming eligible for benefits under the plan. See Gearlds v. Entergy Servs., Inc., 709 F.3d 448 (5th Cir.2013); McCravy v. Metro. Life Ins. Co., 690 F.3d 176 (4th Cir.2012). Accordingly, I join the court’s disposition of Silva’s motion to amend his complaint to add a claim under § 1132(a)(3).
. Although we also must consider MetLife's inherent conflict of interest as both insurer and plan administrator, that conflict will be most relevant " 'as a tiebreaker’ when the issue is close” or " 'where circumstances suggest a higher likelihood that it affected the benefits decision.’ ” Jones v. ReliaStar Life Ins. Co., 615 F.3d 941, 946 (8th Cir.2010) (quoting Metropolitan Life Ins. Co. v. Glenn, 554 U.S. 105, 117, 128 S.Ct. 2343, 171 L.Ed.2d 299 (2008)). I do not find this case so close as to require a tiebreaker, and Silva has not pointed to any evidence that MetLife’s conflict of interest was particularly likely to affect this specific benefits decision.