JORDAN, District Judge:
Following a two-week trial, a jury found that Vanguard Research, Inc. breached its contract with PEAT, Inc. and appropriated PEAT’s trade secrets in violation of the Alabama Trade Secrets Act, Ala.Code § 8-27-1 et seq. The jury awarded PEAT $325,981.01 in compensatory damages on the breach of contract claim, as well as $1,819,334,00 in compensatory damages and $8,890,000 in punitive damages on the trade secrets claim. The district court denied Vanguard’s motions for judgment as a matter of law and for new trial, but reduced the punitive damages award to $78,000 and entered judgment accordingly.
Vanguard now appeals the adverse judgment on the trade secrets claim, but not the jury verdict or award on the breach of contract claim. Vanguard argues that there is insufficient evidence to support the verdict on the trade secrets claim, that it is entitled to a new trial due to the improper admission of a Rule 1006 summary exhibit purporting to list PEAT’s trade secrets, that the award of compensatory damages is excessive, and that the evidence does not support an award of punitive damages. As explained below, we reject all of Vanguard’s arguments concerning the sufficiency of the evidence, conclude that a new trial is required due to the erroneous and prejudicial admission of the summary exhibit, and do not reach the challenges to the damages awarded by the jury.1
I
Like most commercial litigation, this action was complicated, and had many twists and turns, both procedural and substantive. We summarize the proceedings and facts only insofar as necessary to provide context for our decision.
A
The dispute between PEAT and Vanguard arose out of a marketing and license agreement between the two entities to create plasma energy systems, which use extreme temperatures generated by plasma energy to treat hazardous waste without leaving behind harmful byproducts. PEAT referred to this plasma energy technology as “Thermal Destruction and Recovery” (TDR), while Vanguard called it “Plasma Energy Pyrolosis System” (PEPS).
Vanguard started doing business with PEAT’s prior parent company, Mason & Hanger, in the early 1990s through a series of non-exclusive marketing agreements. Vanguard essentially provided marketing and expertise in the area of government contracting, while Mason & Hanger provided technology for use in plasma energy systems through a subsidiary, Plasma Engineering Applied Technology, Inc. (old PEAT).
In 1994, Dr. Marlin Springer and other old PEAT engineers applied for a patent for a process to treat waste using plasma energy. The patent was issued in 1996 to [1157]*1157Dr. Springer as U.S. Patent No. 5,534,659, and became an asset of old PEAT.
Two years later, in 1996, several investors, including Dr. Springer, bought from Mason & Hanger the assets of old PEAT, including the ’659 patent. The new company retained the PEAT name. Vanguard, which tried unsuccessfully to purchase old PEAT or be included as part of the purchasing group, continued to work with PEAT after it became a separate company.
PEAT and Vanguard entered into a “teaming agreement” in late 1996 to pursue a contract with the Tennessee Valley Authority. In August of 1997, PEAT granted Vanguard an exclusive license for its technology (described as “the system known as PEPS”) for the program involving the TVA and for “all follow-on programs and systems to this initial program.” Shortly thereafter, the parties submitted a proposal to the TVA for a Phase I project to develop a plasma energy system.
PEAT and Vanguard executed a new marketing and license agreement in December of 1997. This agreement did not grant Vanguard a license to make, use, or develop a plasma energy system. Instead, the agreement provided that a separate license would be negotiated for each subsequent contract, that Vanguard would use PEAT as the sole source for manufacturing all plasma energy systems, and that PEAT retained exclusive rights, title, and interest in all of its intellectual property (including technical and proprietary information). The agreement also contained a merger clause. At trial, PEAT asserted that this merger clause superseded the 1997 license it granted to Vanguard, while Vanguard maintained that the merger clause did not affect the 1997 license, which was separately provided for the TVA Phase I project and for all “follow-on programs and systems.”
The prime contract for Phase I — which called for a fixed plasma energy system— was executed in February of 1998, and PEAT and Vanguard entered into a subcontract around that time for their work on Phase I. PEAT designed the system for Phase I in Alabama and assembled it at Vanguard’s facility in Virginia during 1998. Throughout Phase I, Vanguard criticized various aspects of PEAT’s work. Not surprisingly, at trial the parties presented very different versions of PEAT’s performance, as well as what problems existed, how significant those problems were, and who was responsible for them.
In June of 1998, Vanguard made a proposal for Phase II of the project — which called for a mobile plasma energy system — and identified PEAT as a subcontractor. PEAT received a letter subcontract for certain tasks in Phase II from the government’s prime contractor the following month, and thereafter began working with Vanguard on Phase II conceptual issues. Vanguard executed its Phase II contract with the prime contractor in November of 1998. Though the parties traded various proposals for the scope of work on Phase II, Vanguard and PEAT never executed a subcontract between themselves for Phase II.
Some of the work on Phase II was performed simultaneously with the work on Phase I. PEAT claimed that during its work on Phase II it shared much sensitive propriety information with Vanguard (though it also claimed that it had given Vanguard such information throughout the work on Phase I). Vanguard, for its part, denied that PEAT had passed on or divulged any trade secrets.
While undergoing Phase I testing in January of 1999, the plasma energy system designed by PEAT experienced an explosion which blew an 80-pound door off the incinerator. Again, the parties disa[1158]*1158greed as to the causes of, and responsibility for, the explosion. Vanguard blamed PEAT’s poor design — which it claimed was unsafe — for the explosion, while PEAT generally said the explosion was caused by the failure of various safety components and insufficient time to develop one of the relevant processes.
The following month, Vanguard ordered PEAT to stop work on Phase I and Phase II. According to Vanguard, it became clear after the explosion that PEAT could not do the job. PEAT, on the other hand, asserted that Vanguard used the explosion as a pretext for replacing PEAT as the provider of the plasma energy system. PEAT claimed that, following its termination, Vanguard misappropriated its trade secrets and used them to modify the Phase I system and subsequently build the Phase II system.
B
Whether or not information constitutes a trade secret is generally a question of fact under Alabama law. See, e.g., The Soap Co. v. Ecolab, Inc., 646 So.2d 1366, 1372 (Ala.1994). To qualify as a trade secret under the Alabama Trade Secrets Act, information must (1) be used or intended for use in a trade or business; (2) be included or embodied in a formula, pattern, compilation, computer software, drawing, device, method, technique, or process; (3) not be publicly known and not generally known in the trade or business; (4) not be readily ascertained or derived from publicly available information; (5) be the subject of reasonable efforts to maintain its secrecy; and (6) have significant economic value. See AlaEode § 8-27-2(1). Cf. Ex Parte Miltope Corp., 823 So.2d 640, 643-45 (Ala.2001) (granting petition for writ of mandamus and setting aside trial court’s discovery ruling which required disclosure of customer orders received by corporation and of minutes of meetings of corporation’s board of directors, as such information constituted trade secrets under § 8-27-2(1)). As the plaintiff, PEAT had the burden of establishing each of these statutory elements as to each claimed trade secret, see Public Systems, Inc. v. Towry, 587 So.2d 969, 971 (Ala.1991), and a hotly contested issue at trial was whether PEAT satisfied this burden.
C
Before the presentation of evidence, Vanguard filed a motion in limine objecting to PEAT Exhibit 145, a compilation of documents purporting to list PEAT’s trade secrets.2 In its motion, Vanguard argued that Exhibit 145 was not admissible as a business record under Rule 803(6) because Dr. Springer had said in his deposition that it was prepared for PEAT’s counsel after the litigation began. R. 10:200. During argument on the motion, PEAT conceded that Exhibit 145 had been prepared by PEAT personnel, at the request of PEAT’s counsel, in response to Vanguard’s discovery request that PEAT identify what trade secrets were allegedly misappropriated. R. 16:1 at 17.
Although Vanguard reiterated its position that Rule 803(6) did not allow the introduction of Exhibit 145 because it was not prepared in the ordinary course of business, R. 16:1 at 18-19, PEAT nevertheless argued that Exhibit 145 was analogous to a discovery response. Alternatively, PEAT asserted that Exhibit 145 was “[its] best attempt to summarize the entire universe of trade secrets, to summarize voluminous documents that [it] had pre[1159]*1159pared, [it] had ... provided to Vanguard.” R. 16:1 at 19-20. Vanguard responded that PEAT could not introduce its own discovery responses because they were hearsay (though Vanguard could use such responses as admissions against PEAT if it wished), and that there was no list of trade secrets to summarize under Rule 1006: “This is something that was made where different people sat around and said, well, what can we come up with and claim to be trade secrets. It doesn’t meet the criteria for admission. It is fine for discovery, it is fine to use to ask questions ..., but it’s just not the kind of record that is allowed in under the rules of evidence.” R. 16:1 at 20-22.
After PEAT replied that Exhibit 145 was a “classic summary” because the contents were voluminous and could not be examined in court, the district court overruled the objection and told Vanguard it could question witnesses “as to the veracity of the information in Exhibit 145.” R. 16:1 at 22. The district court also permitted Vanguard to have a standing objection to Exhibit 145. R. 16:1 at 22.
Notwithstanding the standing objection allowed by the district court, Vanguard lodged a contemporaneous objection when PEAT sought to introduce Exhibit 145 through Dr. Springer. R. 16:2 at 231 (“Same objection as we discussed previously.”). The district court noted the objection and, consistent with its prior ruling, overruled it. R. 16:2 at 231.
II
After thorough examination of the record, including a review of the entire 2500-page trial transcript, we reject, without further discussion, Vanguard’s argument that there is insufficient evidence to support the verdict on PEAT’s trade secrets claim. The district court’s denial of Vanguard’s motion for judgment as a matter of law on that claim is therefore affirmed. See Eleventh Cir. Rule 36-1.
Reluctantly, however, we conclude that the district court erred in admitting Exhibit 145, and that the error is sufficiently prejudicial to require a new trial. In light of our grant of a new trial, we do not reach Vanguard’s arguments that the compensatory damages are excessive and that the evidence does not permit an award of punitive damages.
We review the admission of Exhibit 145 for abuse of discretion, which, means that we look to see if the district court “made a clear error of judgment ... or ... applied an incorrect legal standard.” Alexander v. Fulton County, 207 F.3d 1303, 1326 (11th Cir.2000) (citation and internal quotation marks omitted). Although Alabama law controlled the substantive issues in this diversity action, “the admissibility of evidence in federal courts is governed by federal law,” Borden, Inc. v. Florida East Coast Ry. Co., 772 F.2d 750, 754 (11th Cir.1985), so we turn to Rule 1006, the provision under which Exhibit 145 was admitted.
In relevant part, Rule 1006 provides that “the contents of voluminous writings, records, or photographs which cannot conveniently be examined in court may be presented in the form of a chart, summary, or calculation^]” Once admitted, a Rule 1006 exhibit constitutes substantive evidence. See, e.g., United States v. Smyth, 556 F.2d 1179, 1184 (5th Cir.1977) (“Although the word ‘evidence’ does not appear in its text we construe the rule as treating summaries as evidenee[.]”). And because “summaries are elevated under Rule 1006 to the position of evidence, care must be taken to omit argumentative matter in their preparation lest the jury believe that such matter is itself evidence [1160]*1160of the assertion it makes.” Id. at 1184 n. 12.
The materials or documents on which a Rule 1006 exhibit is based must be made available for “examination or copying ... by other parties at [a] reasonable time and place,” but need not be admitted into evidence. If they are not introduced, however, those materials or documents must be admissible under the Federal Rules of Evidence. In other words, Rule 1006 is not a back-door vehicle for the introduction of evidence which is otherwise inadmissible. See generally J. MoLaughlin, J. WeinsteiN, & M. BeRgee, 6 Weinstein’s Federal Evidence § 1006.03[3] (2d ed. 2004) (“Charts, summaries, and calculations are only admissible when based on original or duplicate materials that are themselves admissible evidence.”); C.A. Wright & V.J. Gold, 31 Federal Practice and Procedure § 8043, at 527 (2000) (“Rule 1006 evidence may also be excluded where the source materials are inadmissible hearsay or even where just some parts of those materials are inadmissible hearsay.”). For example, we explained in United States v. Goss, 650 F.2d 1336, 1344 n. 5 (5th Cir. Unit A 1981), that Rule 1006 does not permit “the admission of summaries of the testimony of out-of-court witnesses” because such testimony would be hearsay. See also United States v. Francis, 131 F.3d 1452, 1457-58 (11th Cir.1997) (rejecting challenge to admission of Rule 1006 summaries of intercepted phone calls, in part because underlying calls and transcripts were admitted into evidence); United States v. Norton, 867 F.2d 1354 1363 (11th Cir.1989) (assumptions in Rule 1006 exhibit must be “ ‘supported by evidence in the record’ ”); United States v. Atchley, 699 F.2d 1055, 1058-59 (11th Cir.1983) (affirming admission, under Rule 1006, of chart reflecting telephone toll records which had themselves been introduced under Rule 803(6)).
Exhibit 145 fails to satisfy this foundational evidentiary requirement, as it is comprised of classic hearsay — statements made outside of court by persons other than the declarant, introduced for the truth of the matter asserted (that PEAT in fact had trade secrets it shared with Vanguard), and not admissible under any hearsay exception. Exhibit 145 contains numerous self-serving documents prepared by PEAT after this action was filed: (a) a list prepared by a PEAT engineer of 109 “[tjrade secrets shown, discussed, or provided to [Vanguard];” (b) a “recap” of “[tirade [s]ecrets provided to [Vanguard]” dated January 7, 2000, with eonclusory statements like “[e]hamber configuration trade secrets include the size and shape of both rectangular and cylindrical chambers, instrumentation locations, and entrance and exit locations for the various chamber penetrations when used to process a mixture of organic and non-organic waste materials in a reducing atmosphere to produce a syn-gas;” and (c) a checklist, dated December 30, 1999, and devoid of explanation, of 23 PEAT “trade secrets.” Exhibit 145 also includes a copy of the ’659 patent issued to Dr. Springer on July 9, 1996, and, worst of all, two memoranda (each seven pages long) prepared by Dr. Springer, PEAT’s principal and corporate representative, on January 14, 2000, after the litigation began. The memoranda, though similar, are not identical. In the first memorandum, Dr. Springer provides a definition of trade secrets which is broader than that set forth in Alabama law. Dr. Springer prefaces the second memorandum with the following sentence: “The following is a discussion of process capabilities and design and manufacturing issues that include proprietary information and trade secrets provided to Vanguard by PEAT since 1992.”
Dr. Springer goes on, in each of the memoranda, to make the following conclu-[1161]*1161sory statements and claims, many of which go beyond the existence of trade secrets and comment on Vanguard’s alleged misappropriation and intent:
“Since 1992 ... PEAT personnel have continually shared many trade secrets with [Vanguard].”
“It became clear early in the PEPS Phase [I] Project that Vanguard’s intent was to assimilate ... trade secrets associated with the design, manufacturings, and procurement of key equipment and major subsystems (PEAT’s designated role).”
“The source of this knowledge [i.e., the technology] and basis for considering it a trade secret was the lack of information available in the industry.” “[K]nowledge of different vendor capabilities to meet ... specifications is considered a trade secret.”
“The processing chamber and control system is highly proprietary and the detail design of both are considered trade secret.”
“It also became evident that Vanguard employees had been instructed to mount a relentless effort toward gathering sufficient trade secret information in order to take over design, manufacturing, procurement, integration, and operation of future PEPS.”
As noted earlier, Exhibit 145 was not compiled in the ordinary course of business. Instead, PEAT and its employees (including Dr. Springer) prepared Exhibit 145 during the course of this litigation to respond to Vanguard’s discovery request that PEAT identify what trade secrets were allegedly misappropriated. Dr. Springer essentially admitted that Exhibit 145 was overbroad, as he compiled the information “to include everything [he] thought might be a trade secret.” R. 16:2 at 264, 267 (emphasis added). Moreover, as was made clear below, Exhibit 145 was based in large part on the first-hand knowledge of PEAT’s employees. R. 16:1 at 17 (PEAT’s counsel: “it was prepared by PEAT personnel ... in response to a request ... asking PEAT to identify what trade secrets we contended were misappropriated by Vanguard”); R. 16:1 at 264, 445 (testimony to the same effect by Dr. Springer). Thus, Vanguard correctly argued below that the underlying materials or information on which Exhibit 145 was based were not admissible under Rule 803(6) because they were prepared during, and for use in, litigation. R. 16:1 at 19. See Noble v. Alabama Dept. of Environmental Management, 872 F.2d 361, 366 (11th Cir.1989) (document prepared in anticipation of litigation is not “compiled as a matter of regular practice” for purposes of Rule 803(b)). That meant that Exhibit 145 could not be introduced under Rule 1006. “Summaries of records prepared for litigation are indeed inadmissible[,] ... and inadmissible documents are not made admissible by being summarized.” AM-PAT/Midwest, Inc. v. Illinois Tool Works, Inc. 896 F.2d 1035, 1045 (7th Cir.1990). See also Hackett v. Housing Authority, 750 F.2d 1308, 1311-12 (5th Cir.1985) (race of each landlord, which was determined based on conversations with a few owners, constituted inadmissible hearsay, and could not form part of Rule 1006 exhibit); Paddack v. Dave Christensen, Inc., 745 F.2d 1254, 1260 (9th Cir.1984) (audit report should not have been admitted under Rule 1006 because it was based in part on information derived from “[u]nion sources ... not subject to cross-examination,” and witness could not identify what portions of report were not derived from inadmissible hearsay). In sum, Exhibit 145 was based on inadmissible hearsay and contained conclusory allegations concerning Vanguard’s alleged misappropriation and intent. The district court abused its discretion in admitting it under Rule 1006.
[1162]*1162B
Not every evidentiary error, of course, requires reversal. Our cases, consistent with Rule 61 of the Federal Rules of Civil Procedure,3 hold that a new trial is warranted only where the error has caused substantial prejudice to the affected party (or, stated somewhat differently, affected the party’s “substantial rights” or resulted in “substantial injustice”). See, e.g. Hall v. United Ins. Co. of America, 367 F.3d 1255, 1258-59 (11th Cir.2004) (“substantial prejudice”); Alexander, 207 F.3d at 1330 (“substantial injustice”); Noble, 872 F.2d at 367 (“affect substantial rights”). Notwithstanding the difference in terminology, the inquiry is always directed to the same central question — how much of an effect did the improperly admitted or excluded evidence have on the verdict?
To answer this question, we look to a number of factors, including the number of errors, the closeness of the factual disputes (i.e., the strength of the evidence on the issues affected by the error), and the prejudicial effect of the evidence at issue. We also consider whether counsel intentionally elicited the evidence, whether counsel focused on the evidence during the trial, and whether any cautionary or limiting instructions were given. See, e.g., Aetna Cas. & Sur. Co. v. Gosdin, 803 F.2d 1153, 1160 (11th Cir.1986); Nettles v. Electrolux Motor AB, 784 F.2d 1574, 1581 (11th Cir.1986). Every case must of course be evaluated on its own terms, but our decisions provide some guidance as to when error is harmless. Compare, e.g., Alexander, 207 F.3d at 1329-30 (evidentiary error found harmless where there were proper limiting instructions and “the verdicts were the product of such one-sided evidence” such that it was “unlikely, indeed remote, that the jury could have been swayed erroneously by the wrongfully admitted evidence”), with, e.g., U.S. Steel, LLC v. Tieco, Inc. 261 F.3d 1275, 1288 (11th Cir.2001) (improper admission of state judicial opinion required a new trial where opinion was used by one of the parties “throughout the trial” to help establish disputed facts and counsel told the jury in closing argument “to use the opinion to make credibility determinations”); Gosdin, 803 F.2d at 1160 (reversing where there was only circumstantial evidence (though “substantial” according to the opinion), no limiting instructions were given, and counsel asked a key witness about the wrongfully admitted evidence); We do not lightly cast aside the result of a complex two-week jury trial. But we are convinced that the admission of Exhibit 145 caused substantial prejudice to Vanguard and that a new trial on the trade secrets claim is required.
First, Exhibit 145 went to a critical issue in this case — whether PEAT had trade secrets as defined by § 8-27-2(1). As noted earlier, PEAT had the burden of proving each of the statutory elements as to each claimed trade secret. See, e.g., Towry, 587 So.2d at 971.
Second, Exhibit 145 was prejudicial. It contained numerous hearsay assertions by non-testifying PEAT employees with suggestive headings concerning what trade secrets existed and were provided to Vanguard. To make matters worse, Dr. Springer made conclusory accusations against Vanguard in his two memoranda and even commented on Vanguard’s alleged intent. These statements (really ac[1163]*1163cusations) by Dr. Springer meant that Exhibit 145 was not confined to whether PEAT had trade secrets, but carried over to the separate questions of whether Vanguard misappropriated any of those secrets in violation of Alabama law, and whether Vanguard’s misappropriation was “willful and malicious,” the standard for awarding punitive damages under the Alabama Trade Secrets Act. Ala.Code § 8-27-4(3); R. 20:2 at 2528 (jury instruction).
Third, even though PEAT used Exhibit 145 with its witnesses to establish the existence of trade secrets, the evidence on the issue was close. As discussed below, PEAT’s own witnesses admitted that Exhibit 145 was overbroad, and generally were not able to identify trade secrets with much particularity. Moreover, Vanguard presented a fair amount of evidence suggesting that PEAT did not have any protected trade secrets.
Looking first at PEAT’s own evidence, Dr. Springer, who was PEAT’s corporate representative, testified (a) that Exhibit 145 was “a list of everything that PEAT personnel, myself and my engineering staff, felt constituted proprietary and technical information that included, trade secrets,” R. 16:1 at 231 (emphasis added); (b) that Exhibit 145 was a list of “everything [he] thought might be a trade secret,” R. 16:2 at 264, 267 (emphasis added); (c) that one of the items in Exhibit 145— emissions test data sent to state agencies' — was claimed as a trade secret even though it was available for public review, R. 16:1 at 265-66; and (d) that there was “some overlap” in the list of trade secrets in Exhibit 145, R. 16:1 at 273. Thomas Kelly, PEAT’s chairman of the board (and a 20% shareholder), acknowledged that, prior to suing Vanguard, PEAT did not have a list or inventory of its trade secrets. R. 16:3 at 408. Moreover, when PEAT sold its assets to DAE Technologies in late 2001 and early 2002, and conveyed all of its intellectual property, not all of the trade secrets listed in Exhibit 145 were contained in the schedule to the contract detailing the trade secrets being sold to DAE. R. 16:3 at 408-13; Def. Exh. 321. Mr. Kelly had an explanation for this apparent discrepancy based on his reading of the DAE contract and the schedule — e.g., he said that the term “all data” covered what was not specifically listed, R. 16:3 at 413 — but the important point is that his testimony on this issue was not overwhelming.
PEAT’s experts did not fare much better in specifically identifying the trade secrets, and sometimes presented alternative versions of what the trade secrets were. Richard Lindsay, a chemical engineer specializing in thermal dynamics and heat transfer, testified that the items listed in Exhibit 145 were “normally considered to be trade secrets within the practice in [the] industry,” R. 16:4 at 761, but he hedged his answer on cross-examination, saying that the items were “areas that would typically be areas for consideration of the trade secrets,” R. 16:4 at 780 (emphasis added). When asked if the list in Exhibit 145 was vague, he conceded that “trade secrets generally are relatively elusive,” R. 16:4 at 780-81, and when asked what it was about specific items in Exhibit 145 (e.g., rapid quench of gases) that made them trade secrets, he answered as follows: “I probably can’t tell you exactly what it is because I couldn’t. But I can tell you ... the general area.” R. 16:4 at 781. When questioned further, Mr. Lindsay said (somewhat tautologically) that “the result that produces the result you want defines the trade secret,” and admitted he was “just saying this is the area that trade secrets lie in. I can’t give you the exact trade secret.” R. 16:4 at 782, 783. In the end, Mr. Lindsay said that trade secrets might be the “way the whole system fits together,” without further elab[1164]*1164oration. R. 16:4 at 784, 787, 789. Alternatively, he said that, with respect to some items that were repeated in some form or another in Exhibit 145, “the trade secret is where it’s placed in the process.” R. 16:4 at 792. PEAT’s other expert, Thomas Eddy, a mechanical engineer with a master’s degree who did his thesis on plasma, admitted that 10 of the items in Exhibit 145 were included in Dr. Springer’s patent and- therefore were not trade secrets. R. 16:4 at 851. Mr. Eddy identified some specific matters that he believed were trade secrets (e.g., the configuration of the interiors, the design of the chamber, and the refractory information), and testified generally that, aside from the patent, the items listed in Exhibit 145 were trade secrets in the industry. R. 16:4 at 853-54, 875. Yet when asked on direct examination to identify some examples of PEAT trade secrets, Mr. Eddy could not always be specific: “[A] lot of these are tied together. It’s not like you can just separate them. They are kind of in groups.” R. 16:4 at 852. Indeed, on cross-examination, Mr. Eddy could not point to a single specific formula that was a PEAT trade secret and was given to or shared with Vanguard. R. 16:4 at 917.
For its part, Vanguard presented evidence that PEAT did not have any trade secrets under Alabama law. John Sumner, one of Vanguard’s founders and officers, and a deputy program manager for the plasma energy system program, testified that although PEAT personnel discussed proprietary information at meetings with Vanguard (like a meeting in January of 1998), they never referred to such information as trade secrets. Mr. Sumner also said he did not recall anyone at PEAT using the term trade secrets. R. 16:6 at 1121. John Vavruska, Vanguard’s expert on chemical engineering, testified that there were about 24 patents issued, for the use of plasma for waste, and that proprietary information did not necessarily constitute trade secrets. R. 16:6 at 1181, 1187. He also stated that “none” of the items listed in Exhibit 145 were trade secrets, explained why he so opined, told the jury that about two-thirds of the items in Exhibit 145 were covered by claims in the ’659 patent or could be readily inferred from language in the patent by someone with reasonable skill, and testified that the general incantation of a process or formula did not constitute a trade secret unless PEAT provided specifics. R. 16:6 at 1189, 1254, 1266-67. Jay Ramamurthi, a Vanguard industrial engineer and project manager, stated that Vanguard did not use any PEAT trade secrets on Phase II, and explained how Vanguard was able' to complete the system for Phase II without using PEAT’s unique technology or trade secrets. R. 16:8 at 1725-28. John Kantak, a Vanguard senior vice-president, explained that when VRI and PEAT engineers worked together in teams and shared information, the data Vanguard requested from PEAT engineers was not trade secret information. R. 16:8 at 1978-79. Finally, Gordon Smith, Vanguard’s corporate representative, told the jury that the 1998 Phase II work by PEAT was not marked as trade secrets. R. 20:1 at 2334.
Like PEAT’s witnesses on the issue of trade secrets, Vanguard’s witnesses were repeatedly challenged about their testimony on this topic, and we do not mean to suggest otherwise. For example, Mr. Vav-ruska, though not pressed on specifics, conceded on cross-examination that information for use in a specific process (i.e., a unique application to one’s own process) could be a trade secret. R. 16:6 at 1262. And Mr. Ramamurthi admitted, also on cross-examination, that there were “unique” and “proprietary” aspects to Vanguard’s Phase II system, R. 16:8 at 1747, thereby suggesting that PEAT’s system was also unique and proprietary in some [1165]*1165ways, and maybe even included specific trade secrets. What matters for our purposes, however, is that Vanguard put on substantial evidence — from both fact and expert witnesses — to counter the claim that PEAT had trade secrets as defined by § 8-27-2(1).
Fourth, PEAT’s reliance on Exhibit 145 at critical times showed how important it was in the evidentiary puzzle. When Vanguard moved for a judgment as a matter of law on the trade secrets claim, the district court asked “What aré the trade secrets?” PEAT responded that “[t]hey were identified specifically by Dr. Eddy,” and told the district court that Exhibit 145 “listed” the trade secrets. R 16:8 at 2421 (explaining that the trade secrets were “the use of [the process]” or the “knowledge of what not to use”).4 Then, in closing argument, in an effort to convince the jury that it had trade secrets, PEAT invoked Exhibit 145: “You will have ... Exhibit 145 back in the jury room with you. And you heard all the evidence about why it was created and what all is on it and the groups of things that are on it and all that.” R.20:2 at 2549.
Fifth, because disputes between the parties as to Phase I had been submitted to arbitration, PEAT’s trade secrets claim was limited to trade secrets obtained by Vanguard “apart from the Phase I subcontract.” R. 20:2 at 2503 (jury instruction). There was no clear demarcation from the evidence at trial as to which trade secrets, if any, satisfied this limitation, and the general verdict, R. 21:215, does not identify what trade secrets PEAT proved to the jury’s satisfaction.
Sixth, as far as we can tell, the district court did not give any limiting instruction on the portions of Exhibit 145 which contained inadmissible hearsay or Mr. Springer’s conclusory comments and allegations about Vanguard’s alleged misappropriation and intent. The district court read the statutory definition of a trade secret to the jury, and instructed the jury that in deciding what trade secrets existed it was not bound by the testimony of any witness or the notation on any document. R. 16:2 at 168-69, 206-07; R. 20:2 at 2512-13. The district court also told the jury that a patent is a matter of public record, and that information disclosed in a patent cannot be a trade secret. R. 20:2 at 2518. These instructions, however, did not sufficiently eliminate or reduce the prejudicial effect of the inadmissible evidence and statements in Exhibit 145.
Ill
The trial transcript provides no clear answer as to what trade secrets PEAT had, if any, in Phase II. As we see it, the evidence allowed the jury to find for either side on this threshold issue. Because the evidence was so close, the admission of Exhibit 145 constituted reversible error.
The jury verdict and judgment in favor of PEAT on the trade secrets claim is reversed, and that claim is remanded for a new trial. We reject Vanguard’s argument that PEAT presented insufficient evidence on the trade secrets claim, and do not reach Vanguard’s arguments concerning the compensatory and punitive damages awarded by the jury on that claim.5
AFFIRMED IN PART, REVERSED IN PART, AND REMANDED.