An appellate court’s review of a trial court’s grant of certification should be a two-step process. First, the focus of the appellate court’s review should be on whether the order appealed is “final” as defined by R.C. 2505.02. The reviewing court should concentrate on answering that predominantly legal question of whether the order sought to be appealed affects a substantial right and whether it in effect determines an action and prevents a judgment. It is in this first step of the review process that the court of appeals plays its most important role.
Second, the appellate court should review the trial court’s determination, required by Civ.R. 54(B), that “there is no just reason for delay.” As this court has held in the past, the phrase “no just reason for delay” is not a mystical incantation which transforms a nonfinal order into a final appealable order. Chef Italiano Corp. v. Kent State Univ. (1989), 44 Ohio St.3d 86, 541 N.E.2d 64. Such language can, however, through Civ.R. 54(B), transform a final order into a final appealable order.
In deciding that there is no just reason for delay, the trial judge makes what is essentially a factual determination — whether an interlocutory appeal is consistent with the interests of sound judicial administration, i.e., whether it leads to judicial economy. Trial judges are granted the discretion to make such a determination because they stand in an unmatched position to determine whether an appeal of a final order dealing with fewer than all of the parties in a multiparty case is most *355efficiently heard prior to trial on the merits. The trial court can best determine how the court’s and the parties’ resources may most effectively be utilized. The trial court is most capable of ascertaining whether not granting a final order might result in the case being tried twice. The trial court has seen the development of the case, is familiar with much of the evidence, is most familiar with the trial court calendar, and can best determine any likely detrimental effect of piecemeal litigation. More important than the avoidance of piecemeal appeals is the avoidance of piecemeal trials. It conserves expense for the parties and clarifies liability issues for jurors when cases are tried without “empty chairs.”
In making its factual determination that the interest of sound judicial administration is best served by allowing an immediate appeal, the trial court is entitled to the same presumption of correctness that it is accorded regarding other factual findings. An appellate court should not substitute its judgment for that of the trial court where some competent and credible evidence supports the trial court’s factual findings. Seasons Coal Co. v. Cleveland (1984), 10 Ohio St.3d 77, 10 OBR 408, 461 N.E.2d 1273. Likewise, regarding Civ.R. 54(B) certification, where the record indicates that the interests of sound judicial administration could be served by a finding of “no just reason for delay,” the trial court’s certification determination must stand. An appellate court need not find that the trial court’s certification is the most likely route to judicial economy, but that it is one route which might lead there. Trial courts, however, should be careful not to breach the duty entrusted to them, and should avoid a mechanical application of the Civ.R. 54(B) language.
The first step that should have been taken by the court of appeals in this case was to determine whether the order appealed was “final,” as defined by R.C. 2505.02. The present order fits squarely within the statute. The statute’s first clause addresses the type of order at issue in this case — an “order that affects a substantial right in an action which in effect determines the action and prevents a judgment.” The order given in this case certainly affects a substantial right, that being potential recovery against an alleged tortfeasor. The order also determines the action against the appellees. All of the causes of action between appellants and appellees are resolved by the summary judgment. The summary judgment prevents a judgment against appellees for appellants. That cross-claims exist between appellees and other defendants is irrelevant — at this point, appellants cannot recover from appellees.
Since the summary judgment granted at the trial level was a final order pursuant R.C. 2505.02, the appellate court next should have determined whether the record indicates that a finding of “no just reason for delay” could lead to judicial economy. The paramount consideration to be made is whether the court’s determination serves judicial economy at the trial level.
*356In this case, the trial court did not act reflexively in finding “no just reason for delay.” The court acted only after appellants filed a motion seeking an order to certify nunc pro tunc. Thus, the trial judge did not mechanically sign a boilerplate summary judgment entry — he realized the import of his certification. The record reveals that that reasoned decision could result in judicial economy.
First, the life of the cross-claims against Grinnell and JDB independent of the original claims against them is tenuous, at best. The viability of the cross-claims depends on the alleged negligence of Grinnell and JDB, and the issue of their negligence would be at the heart of an immediate appeal on the original claims. A resolution of that issue would appear to resolve all parties’ claims against Grinnell and JDB, or at least would be very helpful in settlement. If summary judgment against plaintiffs were upheld, Grinnell and JDB might also decide to seek summary judgment against Sverdrup and Graham, respectively, which they failed to do specifically earlier. If the failure to seek summary judgment against the complaining cross-defendants was purposeful, an immediate appeal prevents a party from keeping alive a shaky cross-claim in order to avoid trial on a more meritorious claim.
Also, an immediate appeal presents the only possible way to achieve the most efficient and straightforward trial, one with all of the parties present with an ability to present evidence against each other. Absent an immediate successful appeal, Grinnell and JDB would remain in the trial as defendants on the cross-claims, but not as defendants on the original claims, and appellants would be unable to present evidence against them. If appellants were to attempt to present evidence through the “back door” against Grinnell and JDB, that might deflect liability away from the remaining defendants. Since Grinnell and JDB can only be liable through Sverdrup and Graham, respectively, it would serve their interests for Sverdrup and Graham to prevail against appellants, and each might decide not to present evidence against Sverdrup and Graham. The unresolved appeals allow for a myriad of tangled scenarios full of duplicitous machinations which could only serve to muddle the issues and present the jury with less than the full story.
If, on the other hand, the appeals are unsuccessful, the parties may at least operate from certainty. A resolved appeal, whether successful or unsuccessful, will go far to clarify the issues for the purpose of settlement between the parties. That may well lead to the most efficient resolution possible — no trial.
A successful appeal subsequent to trial would lead to the possibility of the case being tried twice, both times with “empty chairs.” That possibility is enough to demonstrate that the trial court reasonably found that there was “no just reason for delay” for appellants’ appeal.
*357Therefore, since the summary judgments in favor of Grinnell and JDB were “final orders” as defined by R.C. 2505.02, and since the record reflects that the interests of sound judicial administration could be served through a finding of “no just reason for delay,” we reverse the judgment of the court of appeals dismissing appellants’ appeal.
Judgment reversed.
Moyer, C.J., A.W. Sweeney, Wright and F.E. Sweeney, JJ., concur. Douglas and Resnick, JJ., separately dissent.