MEMORANDUM OPINION & ORDER
In December 2005, Plaintiff Babcock & Wilcox Company (“B&W”) obtained a contract from Kansas City Power & Light Company (“KCP&L”) to design and construct a Selective Catalyst Reduction (“SCR”) system to control emissions from one of KCP&L’s coal-burning power stations in La Cygne, Kansas. Complaint, Doc. 1, p. 3, ¶ 8. Thereafter, B&W issued a purchase order to Defendant Cormetech, Inc. (“Cormetech”) to obtain catalyst modules to be used in the SCR. Id. ¶¶ 7, 9-10. Cormetech guaranteed that the catalyst it supplied would not need to be replaced for 24,000 operating hours. However, the catalyst reached the end of its useful life well before the guaranteed number of hours. Id., p. 4, ¶ 14. KCP&L made claims against B&W and B&W paid KCP&L $3.5 million pursuant to a settlement reached in mediation. Id., ¶ 16. In this diversity action,1 B&W seeks to recover the settlement amount from Cormetech, asserting two claims: breach of warranty and indemnification.2
Cormetech has filed a Motion for Summary Judgment (the “Motion”) pursuant to Fed. R. Civ. P. 56 arguing that B&W’s *1019claims are time-barred and that the claims also fail because B&W has no evidence of a defect in the catalyst. Doe. 98. B&W filed a brief in opposition (Doc. 114) and Corme-tech filed a reply (Doc. 117).3
The Court GRANTS Cormetech’s Motion. B&W’s breach of warranty claim is time-barred and there is no genuine issue of material fact that would allow B&W to recover on its contractual indemnity claim. The contract’s indemnity provision requires B&W to prove that its settlement loss was caused by a defect or an act or omission attributable to Cormetech. As more fully set forth below, B&W has pointed to no evidence of such a defect or act or omission and B&W’s expert testified that no catalyst would have worked given the conditions that existed at the relevant time at KCP&L’s LaCygne power station.
I. Background
B&W, a Delaware corporation having its principal place of business in Ohio, is in the business of designing, manufacturing, and constructing power generation facilities and emissions control equipment and services, including SCR systems, for power companies such as KCP&L. Doc. 1, pp. 1-3, ¶¶ 2, 6. Cormetech, a Delaware corporation with its principal place of business in North Carolina, manufactures catalysts for SCR systems. Id, pp. 2-3, ¶ 2, 7.
In its Motion, Cormetech provides the following summary of the purpose and operation of an SCR:
An SCR is designed to reduce emission of nitrogen oxides (“NOx”), an undesirable component of the flue gas that results from burning coal in a power plant. When a power plant’s turbine burns coal, it collects all the gases from the burning process and expels them through a large system of ductwork. The ductwork leads the flue gas into the SCR, which is roughly the size of a three story home. The “floors” of this home are [ ] comprised of honey-combed catalyst material. The flue gas enters the SCR from the top, and a mist of ammonia is added. As the gas-ammonia mix makes it way down through the “floors,” a chemical reaction occurs. The gas-ammonia mix passes over the surface of the catalyst and the resulting reaction produces nitrogen and water, reducing the output of NOx significantly. Under any circumstance, the catalyst has limited useful life and must be changed out at regular intervals.
Doc. 98-1, pp. 1-2. In his report, B&W’s expert, Dr. Stephen Niksa (“Niksa”) agrees with this description of an SCR system (114-17, p. 4) and states, “Even under the best of circumstances, the active sites on SCR catalysts are slowly taken out of service by poisons and masking agents.... Such processes are said to deactivate the catalyst....” Id, p. 6. For that reason, SCRs are designed to facilitate the periodic replacement of catalyst modules. See Doc. 103-3, pp. 46-47 (Deposition of Mark Rohner, B&W’s purchasing agent).
In December 2005, B&W entered into a contract with KCP&L for the design and construction of an SCR system at KCP&L’s power station in La Cygne. Doc. 1, p. 3, ¶ 8. B&W thereafter provided design specifications to Cormetech for the catalyst modules to be used in the SCR system at the La Cygne plant and later contracted with Cormetech to purchase the catalyst. Id, ¶ 9. On the basis of the parties’ assumptions as to, inter alia, the type of fuel KCP&L would use at the power station, Cormetech made certain *1020guarantees regarding the catalyst, including that the ammonia slip (aka “NH slip”)4 would not surpass 2 parts per million (“ppm”) and that the catalyst would last for 24,000 operating hours. See, e.g., Doc. 114-4, p. 6 (B&W’s Technical Specifications setting forth the ammonia slip level); Doc. 114-6, p. 3 (Cormetech’s Performance Guarantee Language describing the catalyst life expectancy of 24,000 operating hours); Doc. 114-17, pp. 4-5 (Niksa’s report).
Niksa’s report provides the following information about the catalyst supplied by Cormetech for the LaCygne power station:
The catalyst was provided in the form of hundreds of modules, which are rectangular blocks with internal honeycomb walls that form numerous square channels along the length of each block. The modules are closely stacked into scaled layers that process all the flue gas that enters the SCR. The SCR at La Cygne has two identical SCRs installed side-by-side that each process about half of the flue gas stream.
Doc. 114-17, p. 4.
KCP&L began operating the SCR at the La Cygne plant in April 2007. Doc. 1, p. 4, ¶ 13. B&W conducted an Initial Performance Test in June 2007 at 1,200-1,500 hours of operation, which revealed that the ammonia slip was at a higher level than expected.5 Doc. 114-17, p. 7; Doc. 98-15, p. 3. In other words, more ammonia was being emitted from the SCR than expeet-ed, albeit still within the guaranteed limits. The catalyst was otherwise operating to reduce the emission of NOx as expected.6 Doc. 114, p. 7; Doc. 98-1, p. 4; Doc. 114-17, pp. 4-5, 7-8; Doc. 114-18, p. 5, 1120.
B&W’s expert report explains the significance of rising levels of ammonia slip:
As the NO reduction rate diminishes, the [ammonia] slip increases. This inverse proportionality explains why monitoring [ammonia] slip is widely regarded as the most reliable means to assess the extent of catalyst deactivation ....
Doc. 114-17, p. 6.
On October 8, 2007, Cormetech reported to B&W the results of its analysis of catalyst samples extracted from the SCR at “approximately 3000 hours of field service.”7 Doc. 114-9, p. 2. Such testing provides even “more conclusive evidence” regarding the degree of catalyst deactivation than measuring ammonia slip, according to B&W’s expert report. Doc. 114-17, p. 8. This testing showed that the catalyst had already reached 50% of its useful life. Id, p. 9.
Slightly more than one year after the Initial Performance Test, on August 5, 2008, KCP&L reported to B&W that its testing showed that the catalyst was then at the end of its useful life and that the ammonia slip was well beyond the guaranteed level. Doe. 1, p. 4, ¶ 14; Doc. 114-12, p. 4. KCP&L’s testing was conducted at 8,000 *1021hours and, as noted above, the catalyst had been expected, and was guaranteed, to last 24,000 hours. Doc. 1, p. 4, ¶ 14; Doc. 114-12, p. 4.
Thereafter, B&W and Cormetech, per their contract, undertook a root cause evaluation to determine why the catalyst failed to meet its expected life term and ammonia slip guarantee. Doc. 98-2, p. 4. On February 7, 2009, B&W issued a report, the Root Cause Analysis (“RCA”), summarizing its findings and conclusions. Doc. 98-2. The RCA determined that there were seven factors that contributed to the early deactivation of the catalyst, none of which was attributable to Cormetech:
1. Fuels Burned
Conclusion: It is B&W’s belief that burning fuels outside the specified range has led to increased catalyst deactivation.
2. Ash Loading — Isokinetic Test Results
Conclusion: It is B&W’s belief that the catalyst is seeing higher ash loading than it was designed and specified for and thus has experienced increased catalyst deactivation.
3. Sonic Horn Operation
Conclusion: It is B&W’s belief that the inoperable sonic horns as detailed in Appendix C and as noted above have allowed ash to accumulate on the catalyst and led to increased catalyst deactivation.
4. Tube Leaks
* * *
Conclusion: It is B&W’s belief that severe tube leaks and continued unit operation with these tube leaks has led to conditions which negatively affect catalyst activity.
5. Operation Conditions
* * *
Conclusion: It is B&W’s belief that KCP&L’s continued operation of the SCR, without removal of the ash from the catalyst, caused increased catalyst deactivation.
6. Catalyst Cleaning Procedures
* * :|C
Conclusion: It is B&W’s belief that KCP&L’s continued operation of the SCR, without removal of the ash from the catalyst, caused increased catalyst deactivation.
7. Unit Operations
Conclusion: It is B&W’s belief that poor combustion as evidence by higher than average LOI and the instances of carburization, led to increased ash loading to catalyst and perhaps increased the risk of phosphorous poisoning, and in turn led to increased catalyst deactivation.
Doc. 98-2, p. 6-10 (emphasis in original).8
Significantly, B&W’s RCA found that Cormetech had properly sized the catalyst (id., p. 10) and, in its overall conclusion, stated:
B&W feels confident that sufficient catalyst was supplied for this unit to meet all performance guarantees, if the unit had been operated as intended.
Id., p. 11.
After KCP&L determined in 2008 that the catalyst was at the end of its useful life, it contracted directly with Cormetech to obtain replacement catalyst. That cata*1022lyst also failed before the end of its expected life. Doe. 114-17, p. 11. Ultimately, according to B&W’s expert report, the phosphorus poisoning problem that allegedly caused the early deactivation of both the original and replacement catalyst “could only be eliminated by replacing the cyclone burners [in the power- station’s furnace] that were in place while both of Cormetech’s catalyst installations were completely deactivated.” Doc. 114-17, p. 16.
KCP&L made a commercial claim against B&W for about $60 million based on the unsatisfactory performance of the SCR, which resulted in a mediation pursuant to which the claim was settled by B&W for $3.5 million. Doc. 98-3, pp. 5-6.
II. Legal Standard
Under Federal Rule of Civil Procedure 56(a), a “court shall grant summary judgment if the movant shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed. R. Civ. Pro. 56(a). The movant “bears the initial responsibility of informing the district court of the basis for its motion, and identifying those portions of the pleadings, depositions, answers to interrogatories, and, admissions on file, together with affidavits, if any, which it believes demonstrates the absence of a genuine issue of material fact.” Celotex Corp. v. Catrett, 477 U.S. 317, 323, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986) (internal quotations omitted).
Once the moving party has met its burden, the non-moving party “must present affirmative evidence in order to defeat a properly supported motion for summary judgment.” Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 257, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986). “[A] party opposing a properly supported motion for summary judgment may not rest upon mere allegation or denials of his pleading, but must set forth specific facts showing that there is a genuine issue for trial.” Id. at 256, 106 S.Ct. 2505; Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 586, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986) (“When the moving party has carried its burden under Rule 56(c), its opponent must do more than simply show that there is some metaphysical doubt as to the material facts.”). “[T]he mere existence of some alleged factual dispute between the parties will not defeat an otherwise properly supported motion for, summary judgment; the requirement is that there be no genuine issue of material fact.” Anderson, 477 U.S. at 247-48, 106 S.Ct. 2505 (emphasis in original). “Only disputes over facts that might affect the outcome of the suit under the governing law will properly preclude the entry of summary judgment.” Id. at 248, 106 S.Ct. 2505.
III. Analysis
A. Ohio’s UCC statute of limitations bars B&W’s breach of warranty claim.
Cormetech asserts that the catalyst is a “good.” Sales of goods are governed by chapter 2 of the Uniform Commercial Code, adopted in Ohio as Ohio Rev. Code Ch. 1302.9 Cormetech argues that B&W’s claims are time-barred under the UCC’s four-year statute of limitations, set forth in R.C § 1302.98.10 Doc. 98-1, pp. 13-16. B&W contends that the catalyst is not a “good” subject to the UCC and that, therefore, the fifteen-year limitations period found in R.C. § 2305.06 applies to its claims.11 Doc. *1023114, p. 13. B&W further argues that, even if the UCC’s four-year limitations period applies, a genuine issue of material fact exists as to when the limitations period began to run, precluding summary judgment. Doc. 114, pp. 8-20.
1. The catalyst is a good and the UCC applies to B&W’s breach of warranty claim.
a. The catalyst is a good.
A “good” is defined as: all things (including specially manufactured goods) which are movable at the time of identification to the contract for sale.... “Goods” also includes... other identified things attached to realty as described in section 1302.03 of the Revised Code.
R.C. 1302.01(A)(8); UCC § 2-105.
Section 1302,03, referenced in R.C. § 1302.01(A)(8) provides, in pertinent part,
(A) A contract for the sale of minerals or the like, including oil and gas, or a structure or its materials to be removed from realty is a contract for the sale of goods within sections 1302.01 to 1302.98 of the Revised Code, if they are to be severed by the seller....
(B) A contract for the sale apart from the land of growing crops or other things attached to realty and capable of severance without material harm thereto but not described in division (A) of this section...is a contract for the sale of goods within sections 1302.01 to 1302.98 of the Revised Code, whether the subject matter is to be severed by the buyer or by the seller even though it forms part of the realty at the time of contracting, and the parties can by identification effect a present sale before severance.
Id.; UCC § 2-107.
Under these provisions, the catalyst is a “good” subject to the UCC. The catalyst was movable at the time it was identified to the contract. It was meant to be installed and removed from the SCR. See Docs. 98-6, p. 8 (deposition of Scott Hiedeman, KCP&L air quality control engineer, agreeing that a catalyst is not permanently affixed to the SCR and is designed to be removed periodically); 117-1, pp. 2-4 (deposition of Jason Chitwood, B&W process engineer, agreeing that catalyst layers are designed in a modular fashion and are removable). The catalyst is capable of severance from the SCR without material harm. See id. (Chitwood agreeing that the SCR contains doors, rails and other apparatus to facilitate the removal of catalyst blocks).
B&W’s case law, cited in a footnote in its opposition brief, is not persuasive because none of the cases involved a part intended to be installed and uninstalled in a structure designed to accommodate the replacement of such parts. See Doc. 114, p. 9, n. 4 (citing Weiss v. MI Home Prods., Inc., 376 Ill.App.3d 1001, 315 Ill.Dec. 690, 877 N.E.2d 442 (2007) (homeowner sued alleging faulty windows; the court found that windows, once annexed to a home, become part of the real estate and are no longer goods); Kennedy v. Vacation Internationale, Ltd., 841 F.Supp. 986, 990 (D.HI.1994) (tile affixed to a lanai is not a good because it is not capable of severance without material harm to the realty); Keck v. Dryvit Sys., Inc., 830 So.2d 1, 8-9 (Ala. 2002) (homeowners sued alleging that the exterior insulation finishing system feature of their home was a good; the court found that the system was not a good because it could not be detached from the house without causing damage but instead was an integral part of the structure of the home); Loyd v. Ewald, 1988 WL 37484, *2 (Oh.Ct.App. Apr. 6, 1988) (in-ground swimming pool was not a good because it was incapable of severance without causing material harm to the realty)).
*1024b. Applying the predominant purpose test, the contract between B&W and Cormetech was one for the sale of goods.
B&W argues-that, even if the catalyst was a good, the UCC does not apply because its contract with Cormetech was a mixed contract for goods and services, the predominant purpose of which was the purchase of services,-not goods. Doc. 114, pp. 9-10. In support of its argument it cites C.J. Mahan Constr. Co. v. Valspar Corp, 30 Fed.Appx. 381 (6th Cir.2002).
When a contract is a mixed contract for goods and services, Ohio courts use the predominant purpose test to determine whether the contract is for the sale of goods or services; if the contract is for services the UCC does not apply. Id at 383 (citing Allied Indus. Serv. Corp. v. Kasle Iron & Metals, Inc., 62 Ohio App.2d 144, 405 N.E.2d 307, 310 (1977)). In Mecanique C.N.C., Inc. v. Durr Envtl., Inc., 304 F.Supp.2d 971 (S.D.Ohio 2004), the District Court for the Southern District of Ohio applied Ohio’s predominant purpose test to a contract for the fabrication and installation of ductwork and related equipment for an SCR system. The court considered “whether the predominant factor and purpose of the contract is the rendition of service, with goods incidentally involved, or whether the contract is for the sale of goods, with labor incidentally involved.” Id. at 976 (quoting Allied Indus., 405 N.E.2d at 310). Whether a contract “predominantly involves goods or services is ordinarily a question of fact for the jury[,]” but only “if there is a true factual dispute, not if the division between goods and services merely involves a close call.” Id. at 976-977 (internal citation omitted). In other words, “where ’there are no disputed facts that raise issues to be decided by the jury, it is proper for the trial court to rule as a matter of law on whether the contract is covered by Article Two [of the UCC].”’ Id. quoting Valleaire Golf Club, Inc. v. Conrad, 2003 WL 22900451, at *1-2 (Oh.Ct. App. Dec. 10, 2003). See also Action Grp., Inc. v. NanoStatics Corp., 2013 WL 6708395, *8 (Oh.Ct.App. Dec. 17, 2013); Tekfor, Inc. v. SMS Meer Serv., Inc., 2014 WL 5456525, *4-5 (N.D.Oh. Oct. 27, 2014). The burden of proving that a contract is predominantly for the purchase of goods is on the party asserting that the contract is governed by the UCC. Id.
B&W and Cormetech do not dispute the facts surrounding the contract. Accordingly, the Court must determine, as a matter of law, whether the predominant purpose of the contract between B&W and Corme-tech was for the sale of goods or services. Mecanique, 304 F.Supp.2d at 977 (finding, as a matter of law, that the contract for the fabrication and installation of ductwork and related equipment for an SCR system was governed by the UCC because it was predominantly for the sale of goods).
The Court finds the factors enumerated in Boardman Steel Fabricators, Ltd. v. Andritz, Inc., 2015 WL 5304293, *4 (E.D.Ky. Sept. 9, 2015), to be a useful guide when performing the predominant purpose test: a court considers (1) the language of the contract; (2) the payment terms; (3) the mobility of the goods; (4) the value of the goods and services; and (5) the business of the seller. M (citing BMC Indus., Inc. v. Barth Indus., Inc., 160 F.3d 1322 (11th Cir.1998); Fab-Tech., Inc. v. E.I. DuPont De Nemours & Co., 311 Fed. Appx. 443, 445 (2d Cir.2009)).12
*1025Here, the catalyst was movable; it was fabricated offsite and transported to the plant in La Cygne. The language of the contract and the value of the goods also indicate that the predominant purpose was for the sale of goods. For example, Corme-tech’s Proposal is “for the supply of SCR catalyst.” Doc. 114-1, p. 2 (Proposal). The “Scope of Supply” included a catalyst, performance warranties, drawings, installation procedures, and an operations and maintenance manual. Id., p. 5. The cost of these items was $4,481,000. Id. B&W’s Purchase Order referenced Cormetech’s Proposal and the purchase price: $4,481,000. Doc. 114-3, p. 2 (Purchase Order). Eighty percent (80%) of the purchase price was to be paid upon “completion of catalyst production”; 10% was due upon delivery of the catalyst to the jobsite; and 10% was due upon acceptance of the catalyst. Doc. 114-1, p. 6.
Mark Rohner, B&W’s purchasing agent, testified that his job was to purchase equipment, not services. Doc. 103-3, p. 44. The Purchase Order he generated for the Cormetech catalyst was for the purchase of “equipment.” Id. at 36-37, 39. He signified this by using the letters “EQ” in the Charge Number on the face of the Purchase Order. Id. Moreover, the term “Purchase Order” is “almost exclusively used for transactions in goods.” Boardman Steel, 2015 WL 5304293, at *5 (quoting BMC Indus., 160 F.3d at 1331). Thus, the primary item offered and paid for was the catalyst, not services.
The Purchase Order described the “Field Services” to be performed by Cor-metech as consisting of “on-site supervision or installation, refurbishment, performance testing, and other field service work,” which were to be billed only when services were required. Doc. 114-1, p. 6. The cost of services began at $950 per person per day, exclusive of overtime, holiday rates and transportation costs. Id. These Field Service charges are very small compared to the $4,481,000 purchase price for the catalyst itself. See Tekfor, 2014 WL 5456525, *5 (contract was a service repair contract because 75% of the invoiced amount was for services, not parts); Boardman Steel, 2015 WL 5304293, *5 (price for “field assistance” billed at $1,000 per day and extra design costs approaching one million dollars “are still a drop-in-the-bucket compared to the costs of the fabricated steel” contracted for, $5,317,750). B&W’s Technical Specifications contain a provision that states, “In the event site service is required, [Corme-tech] shall furnish a qualified representative for such service at the stipulated per diem rate in [Cormetech’s] proposal.” Doc. 114-4, p. 4. The language, “In the event site service is required,” is inconsistent with a contract predominantly for services.
Finally, regarding the business of the seller, B&W, in its Complaint, describes Cormetech as a manufacturer: “Cormetech manufactures catalyst for [SCR] systems for companies operating in the power generation industry.” Doc. 1, p. 3, ¶ 7. The Complaint also indicates that what B&W purchased from Cormetech was “catalyst modules.” Id., ¶¶ 9-10. See Action Grp., 2013 WL 6708395, at *8 (when applying the predominant purpose test, a court may look to the complaint to determine how the parties characterize their dealings). In sum, an analysis of the factors set forth in Boardman Steel overwhelmingly supports a finding that the predominant purpose of the contract between B&W and Cormetech was the sale of goods.
In support of its argument that the contract was predominantly for services, B&W points out the following: (1) the Purchase Order references final inspection to be made by Cormetech after installation; (2) Cormetech was responsible for catalyst testing after installation; and (3) *1026the Purchase Order incorporates additional documents that describe services to be performed by Cormetech, including a warranty clause that references services and a warranty guarantee that required Cor-metech and B&W to undertake a root cause analysis and perform repairs to the catalyst in the event of a breach. Doc. 114, pp. 11-12.
The evidence cited by B&W does not support a conclusion that the predominant purpose of the contract was the purchase of services. Taking B&W’s arguments in turn: the fact that the Purchase Order referenced a final inspection by Cormetech of its own catalyst (Doc. 114-3, p. 6) does not render the sale of the catalyst one for services. It is not unusual for a manufacturer to inspect its own product. Indeed, in Mecanique, the court observed, “virtually all commercial goods involve some type of service, “whether design, assembly, installation, or manufacture.’” 304 F.Supp.2d 971 at 977 (quoting Neibarger v. Universal Coops., 439 Mich. 512, 486 N.W.2d 612, 622 (1992)). As for testing, B&W cites Home Ins. Co. v. Detroit Fire Extinguisher Co., 212 Mich.App. 522, 538 N.W.2d 424, 427-428 (Mich.Ct.App.1995) in support of its claim that a contract requiring regular testing and service after installation is predominately one for services. Doc. 114, p. 11. Home Ins. Co. is distinguishable for two reasons. First, there appeared to be disputed issues of fact that the state court of appeals found precluded the trial court from ruling on the predominant purpose as a matter of law; second, the seller in that case contracted to engineer, design, manufacture, and install a fire extinguisher system in a manufacturing plant. Home Ins. Co., 538 N.W.2d at 426-428. Here, Corme-tech, the seller, supplied the catalyst to be used in the SCR system that B&W, the buyer, engineered, designed and installed at the power plant. Thus, although the contract between B&W and Cormetech provided that Cormetech would perform additional testing of its catalyst, the Court does not find that such testing made the contract primarily one for services. Finally, the fact that the warranty provisions required Cormetech, in the event of a breach, to perform a root cause analysis and, if appropriate, make repairs, is not indicative of a service contract but merely describes a remedy in the case of a breach.
Other capes B&W relies upon in support of its position (Doc. 114, p. 12, n. 8) are also unpersuasive. In Allied Indus., the defendant had a scrap iron processing plant that was non-compliant with air pollution standards. 405 N.E.2d at 308-09. Defendant hired the plaintiff to design a pollution control system and recommend equipment to use for this purpose. Id. at 308. The court applied the predominant purpose test and held that the contract was predominantly for services because the plaintiff designed and installed pre-purchased goods and acted, in effect, as the defendant’s agent. Id. at 147. The role of the plaintiff in Allied Indus. most closely resembles the role of B&W, not Corme-tech, in this case. It was B&W that contracted with KCP&L to design and build the SCR using component parts purchased from manufacturers like Cormetech.
In Ankle & Foot Care Ctrs. v. Info-Cure, 164 F.Supp.2d 953, 956-957, 959-960 (N.D.Ohio 2001), the court found that there were disputed issues of fact regarding a contract for medical billing software. The defendant provided and installed the software (onto computers that the plaintiff purchased from a separate supplier) and trained the plaintiffs staff on how to use the software. In Franklin Publ’ns, Inc. v. General Nutrition Corp., 2007 WL 2071914 (S.D.Ohio July 13, 2007), the plaintiff, a magazine publisher, entered into agreements with the defendant, a seller of nutritional products, whereby the defendant would supply the plaintiff with *1027names and mailing addresses of members of its discount club and the plaintiff would distribute to those members copies of two of its magazines. Id. at *1. The court determined that the contracts were predominantly ones for services because, most significantly, the contracts provided the defendant with free advertising space in the plaintiffs magazines and the ability to purchase additional advertising at a reduced rate. Id. at *5. The court explained that a contract for advertising is a contract for services, not goods. M And in C.J. Mahan Constr. Co., the court stated, “It is nonsensical to suggest that the provision of free technical support and advice makes the provision of services the predominate purpose of a contract for the sale of thousands of gallons of paint.” 80 Fed.Appx. at 383. These cases do not persuade the Court that the contract between B&W and Cormeteeh was predominately one for services. The contract was predominantly one for the purchase of goods. Accordingly, the UCC governs B&W’s breach of warranty claim.
2. The UCC four-year statute of limitations began to run in 2007; B&W’s breach of warranty claim filed in 2012 is time-barred.
The statute of limitations begins to run on a breach of warranty claim “when the breach is or should have been discovered.” R.C. § 1302.98(B); UCC § 2-725(2). Cormeteeh argues that B&W should have discovered that the catalyst would not reach its guaranteed life no later than August 21, 2007, as evidenced by an email and attached letter sent by B&W notifying Cormeteeh of the likely early failure of the catalyst. Doc. 98-1, pp. 16-17. B&W disagrees; it asserts that the limitations period began to run in September 2008, when it received notice from KCP&L “of performance warranty issues,” i.e., that the catalyst “did not (or would not) meet the performance guarantees.” Doc. 114, p. 16. In his August 21, 2007, correspondence, B&W’s Rohner stated,
The attached letter is to notify Corme-tech of the potential performance failure your [sic] catalyst at the La Cygne job-site.
Doc. 98-15, p. 2. The attached letter, addressed to Cormeteeh and also dated August 21, 2007, advised, in pertinent part,
As Cormeteeh is aware, B&W has conducted SCR-System testing at the La Cygne jobsite. Test results have been forwarded to your Jeremy Freeman for review and comment.
The tests were conducted at approximately the 1,500 hour mark in boiler operation. B&W’s analysis of the test results, which include input from Cor-metech, indicates that the Cormeteeh catalyst is failing to perform at expected values at the beginning of its active life. * * *
Therefore this letter serves to put Cor-metech on notice that, based on the results, B&W has reason to believe that the catalyst will not reach the end of life guarantees.
B&W would like to work with Corme-tech to determine the reasons for advanced deactivation....
Doc. 98-15, p. 39 (emphasis supplied).
On deposition, Rohner agreed that, at the time he sent the August 21, 2007, letter, B&W “was aware that there was a problem potentially with the catalyst at La Cygne.” Doc. 98-13, p. 9 And, referring to the June 2007 test results that were addressed in Ro'hner’s letter, B&W’s expert report states that “accelerated deactivation of the catalyst was clearly apparent in the much-greater-than-expected [ammonia] slip after 1,200 hrs on-stream.... ” Doc. 114-17, p. 11. Thus, the Initial Performance Test performed in June 2007 and the Rohner correspondence describing the results of that test serve to show that B&W *1028knew, on or before August 21, 2007, that the catalyst was deactivating early and would not likely reach its end of life guarantees.
B&W argues that it only learned the catalyst would not meet its expected guarantees in 2008, at or shortly before the date KCP&L notified B&W that the catalyst had reached the end of its life. B&W cites representations that Cormeteeh made to it after the August 2007 correspondence indicating that the catalyst had not failed to meet its guarantees at that time and that additional testing was necessary. Doc. 114, p. 19 (citing Doc. 114-9 (letter from Cormeteeh to B&W dated October 8, 2007)). Rather than support B&W’s contention, the October 2007 letter only serves to bolster Cormetech’s position. The letter explicitly states that the catalyst was deactivating more quickly than expected. See Doc. 114-9, p. 2 (“Measured SCR catalyst performance is above the design performance threshold for the La Cygne facility, but is below the expected performance for the SCR catalyst at the La Cygne facility for the operational hours accumulated by the SCR at the time the catalyst elements were removed.”).
B&W’s expert, Niksa, states in his report that the Initial Performance Test (the basis for B&W’s August 21, 2007 letter) “indicates much faster catalyst deactivation than Cormeteeh anticipated” and that the catalyst “would not meet Cormetech’s written guarantees.... ” Doc. 114-17, p. 8. Niksa states that the testing of catalyst samples reported by Cormeteeh to B&W in early October 2007 was even “more conclusive” and “confirm[ed] the indication of excessive deactivation....” Indeed, Niksa stated that the testing reported in October indicated “a loss of 50% of the service life” of the catalyst after only 2,880 hours of operation, i.e., at only 12% of the guaranteed life of 24,000 hours. Id., p. 9. That Cormeteeh recommended additional testing to pinpoint the exact cause of the premature catalyst deactivation does not change the fact that B&W knew of the excessive early deactivation.
Next, B&W identifies specific performance guarantees, including NOx reduction, ammonia slip emissions, and the catalyst operational life, that it asserts it only learned would not be met in September 2008. Doc. 114, p. 16; Doc. 114-4, p. 13. However, based on the elevated ammonia slip levels and excessive deactivation that B&W knew of in August and October 2007, it should have known that the catalyst would not meet its end of life guarantees and, without a working catalyst, NOx emissions would not be reduced as guaranteed.
B&W submits that it did not know and could not have known “of these performance breaches until September, 2008, when it received notice from KCP&L.. .that the catalyst would not reach the end of its 24,000 hour catalytic life guarantee.” Doc. 114, p. 16. The Court disagrees. The 2008 letter from KCP&L to B&W notified B&W that the catalyst had actually reached the end of its useful life. Doc. 1, p. 4, ¶ 14; Doc. 114, p. 16. It stated, “This letter serves as Notice that the SCR does not meet the warranties and Performance Guarantees specified in the Contract” and referenced an increased ammonia slip. Doc. 114-12, p. 2. However, this was not B&W’s first notice of the accelerated catalyst deactivation. As set forth above, its August 2007 letter to Cormeteeh, drafted by Rohner, discussed its Initial Performance Test results confirming an elevated ammonia slip and indicated that, based on the test results, B&W “has reason to believe that the catalyst will not reach the end of life guarantees.” Doc. 98-15, p. 3; see also Doc. 114-17, p. 8 (Niksa’s expert report: “Ammonia slip is the best indirect indication of a catalyst deactivation rate for SCRs on-stream.”). In October 2007, Cormeteeh in*1029formed B&W of test data that showed the catalyst had already reached 50% of its useful life at only 2,880 hours. Doc. 144-17, p. 9.
Given these facts, the date that KCP&L informed B&W that the catalyst had actually reached the end of its useful life and that KCP&L intended to assert its own remedies under its contract with B&W is not the date B&W first knew or should have known that the catalyst would not achieve its guaranteed life. See Miles v. Kohli & Kaliher Assocs., Ltd., 917 F.2d 235, 239, 256 (6th Cir.1990) (Under R.C. § 1302.98, the statute of limitations on a breach of warranty claim for the construction of a bridge began to run when the steel company’s product manager visited the bridge site, identified there was a problem with the bridge that could lead to the eventual collapse of the bridge, and informed the county authority of his finding; not when the bridge actually collapsed).
Because the four-year statute of limitations began to run in 2007, by October 2007 at the latest, it expired no later than October 2011, ten months before B&W filed its first complaint in August 2012. B&W’s breach of warranty claim is therefore time-barred.
B. B&W’s indemnification claim fails because it has not identified a genuine issue of material fact in support of that claim.
The parties’ Standard Terms and Conditions contain the following indemnification provision:
Seller shall defend and indemnify Buyer against all damages, liabilities, claims, losses and expenses (including attorney’s fees) arising out of, or resulting in any way from any defect in the goods or services purchased hereunder or from any act or omission of Seller, its agents, employees or subcontractors.
Docs. 117-2, p. 3, ¶ 4; 114-5, p. 4, ¶ 4.
Cormetech argues that B&W’s indemnity claim is governed by the UCC and is barred by the four-year statute of limitations for the same reason its breach of warranty claim is barred. Doc. 98-1, pp. 17-18. B&W contends that a claim for indemnification is separate from a breach of contract claim and that it does not, therefore, fall under the UCC. Doc. 114, p. 13. The parties agree that Ohio has not decided whether an indemnification claim arising from a sale of goods is governed by the UCC. B&W argues that the majority of jurisdictions that have decided the issue have found that an indemnification claim is not governed by the UCC. Doc. 114, p. 14; Doc. 117, p. 10. Although B&W appears to have the stronger argument, the Court need not decide this issue because, even if B&W’s indemnification claim is not time-barred, it fails on the merits.
On the merits of the claim, Cormetech contends that, in order to prevail, B&W must prove that there was a defect in the catalyst that caused its settlement loss,13 which B&W cannot do. Doc. 98-1, p. 19; Doc. 117, p. 12. Cormetech submits that there is no evidence that its catalyst was defective (Doc. 98-1, pp. 10-12) and cites the RCA, quoted above, in support.14 In *1030response, B&W argues: (1) that its indemnity claim “arises from more than just mere defects in the catalyst or services provided by Cormetech”; and (2) that circumstantial evidence demonstrates that the catalyst was defective. Id., pp. 22-23. The Court considers both of B&W’s' arguments.
1. B&W’s indemnity claim is separate from its warranty claim; it must prove a defect or an act or omission.
B&W’s first argument, that its indemnification claim “arises from more than just mere defects in the catalysts or services provided by Cormetech” (Doc. 114, pp. 22-28), conflates its breach of warranty and indemnification claims. It cites the performance warranties made by Corme-tech, including the warranty that the catalyst would last for 24,000 operating hours, and attempts to tie the performance warranties to the “act or omission” language in the contract’s indemnification provision. Id. In other words, B&W seeks to incorporate its claim for breach of warranty into its indemnification claim. This it cannot do for at least two reasons.
First, this argument is flatly contradicted by other statements made by B&W in its brief. In arguing against application of the UCC statute of limitations to its indemnity claim, it states, “’[Ijndemnity is a separate equitable cause of action” independent from warranty liability under the UCC. Doc. 114, p. 14 (quoting Am. Premier Underwriters Inc. v. General Elec. Co., 866 F.Supp.2d 883, 916 (S.D.Ohio 2012)). B&W also states that it “negotiated [with Cormetech] an indemnity provision wholly apart from the guarantees provided for in the Contract. In other words, the indemnity provision provides separate obligations and protections.” Id., p. 21. Accordingly, B&W may not now use its indemnification claim to circumvent the UCC’s time bar under which its breach of warranty claim fails. '
Second, B&W does not explain what “act or omission” on the part of Cormetech the performance warranties implicate. See Portsmouth Ins. Agency v. Med. Mut. of Ohio, 188 Ohio App.3d 111, 934 N.E.2d 940, 944-945 (2009) (courts look to the language in the indemnity agreement to ascertain a party’s duty to indemnify; thus, to recover under an indemnity provision for unauthorized acts, the indemnitee must show unauthorized acts); Orville Prods., Inc. v. MPI, Inc., 1992 WL 10281, at *2 (Oh.Ct.App.1992). B&W does not identify an act or omission by Cormetech that caused the performance guarantees not to be met. Nor does it even show that the guarantees were not in fact met. For instance, although the contract guarantees the life of the catalyst for 24,000 hours, this warranty is dependent upon the plant firing coal within the specified analysis. The RCA indicates that the plant did not fire coal within the specified analysis. See Doc. 98-2, p. 6 (RCA, explaining that over 30% of the coal samples provided for testing contained an ash content over the maximum specification analysis and concluding, “It is B&W’s belief that burning fuels outside the specified range has led to increased catalyst deactivation.”); Doc. 114-14, p. 6 (deposition of Mark Low, B&W business development manager and vice-president of service products, in which Low stated that, when KCP&L burned fuel outside the specified range, it contributed to the deactivation of the catalyst).
B&W has not identified an act or omission by Cormetech that allegedly caused a *1031breach of the performance guarantees. Even if it is assumed that the performance guarantees were not met, that alone would not show that there is a genuine issue of material fact regarding an “act or omission” that would support B & W’s indemnification claim. Portsmouth Ins., 934 N.E.2d at 944-945.
2. B&W has no evidence of a defect.
B&W’s second argument with respect to the merits of its indemnification claim is that circumstantial evidence “clearly demonstrates that the catalyst was defective.” Doc. 114, p. 23. The two items of circumstantial evidence it points to are: the 2007 Initial Performance Test; and its expert evidence. Id., p. 24.
B&W contends that Ohio law permits a defect to be proven by circumstantial evidence “ ‘[w]here direct evidence is unavailable” and “a preponderance of [the circumstantial] evidence establishes that the loss was caused by a defect and not other possibilities, although not all other possibilities need to be eliminated.’ ” Id., p. 23 (quoting State Farm Fire & Cas. Co. v. Chrysler Corp., 37 Ohio St.3d 1, 523 N.E.2d 489 (1998)).
B&W’s application of the law regarding circumstantial evidence to the facts of this case is misplaced. The mere fact that B&W has no direct evidence does not mean such evidence is unavailable; for example, B&W’s own report, the RCA, shows that the catalyst deactivation was accelerated by operating issues at KCP&L’s plant.15 The RCA and the early failure of the second catalyst supplied by Cormetech, which was of a different size and pitch,16 rule out sizing and pitch as causes of the higher than expected ammonia slip. B&W now attempts to raise a question of fact about the formulation of the catalyst, citing the depositions of its employees John Monacélli, Low, and Chit-wood. Id., p. 23, n. 6. This attempt is unsuccessful for two reasons. First, those employees simply reasoned backwards from the fact that the catalyst failed earlier than expected to engage in rank speculation that the catalyst may not have been formulated properly. See Doc. 114-14, p. 10 (Low’s deposition in which he stated, “Because of the rapid deactivation that we saw....I believe that the formulation probably wasn’t optimum for the conditions that were specified” but also testified that he had no idea what the problems .with the formulation could be); Doc. 114-15, p. 9 (Monacelli’s deposition in which he states that the formulation of the catalyst is “Cormetech’s area of expertise and we do not know that”); Doc. 114-16, p. 8 (Chit-wood’s deposition in which he stated, “I don’t know how they formulate it. I can’t speak to that. All I know is it didn’t make the catalyst life at the end of the day.”)17 Speculation is not evidence.- Lewis v. Philip Morris Inc., 355 F.3d 515, 533 (6th Cir.2004) (non-moving party must demonstrate sufficient probative evidence, not *1032“mere speculation, conjecture, or fantasy,” quoting Godfrey v. Pulitzer Publ’g. Co., 276 F.3d 405, 412 (8th Cir.2002)). The deposition testimony of Mopacelli, Low, and Chitwood do not establish a genuine issue of material fact regarding whether B&W’s settlement loss was caused by a defect in the catalyst, nor do they support that there are no other possibilities.
Second, the speculation regarding a possible defect in the formulation of the catalyst is entirely negated' by the testimony of Niksa. Niksa’s report says that phosphorus poisoning of the catalyst was the cause of its early deactivation. Doc. 114-17, p. 16.18 The report states that there was an “oversight by Cormetech” in its failure to recognize the potential for phosphorus poisoning of the catalyst. Doc. 114-17, p. 15. However, when pressed during his deposition to identify a defect in the catalyst, Niksa was unable to do so. See Doc. 98-18, pp. 13-14, 16, 18, 20-24.19 His inability to identify a defect in the catalyst is consistent with the conclusion in his report that the problem of phosphorus poisoning of the catalyst could only be solved by replacing the cyclone burners in the boiler (furnace) at KCP&L’s power station:
... Scott Heideman, KCP&L’s air quality control engineer, said that the SCR problems associated with P [phosphorus] poisoning were eliminated only after the cyclone burners were replaced in the fall of 2010. Evidently, the contribution of P poisoning to the excessive deactivation of Cormetech’s first and second catalyst installations was predominant, and P poisoning could only be alleviated by replacing the cyclone burners, which would have significantly diminished the extents of locally reducing zones compared to those with the burners in place when the Cormetech catalysts were deactivated at an excessive rate.
Doc. 114-17, p. 15 (emphasis supplied).
Finally, when asked to say what Corme-tech could or should have done differently, Niksa gave the following testimony:
By throwing up a flag saying that[,] given the potential for phosphorous deactivation in this application, no catalyst is going to work. That’s what they should have done. That’s exactly what they should have done.
Doc. 98-18, p. 24 (emphasis supplied).
The statement in Niksa’s report as to an “oversight by Cormetech,” when read in context with the statement in the same report concluding that the phosphorus poisoning could only be corrected by replacing the plant’s cyclone burners, does not establish a defect in the catalyst or in the services provided by Cormetech; rather it is evidence that the cause of the phosphorus poisoning was poor combustion caused by faulty burners. Niksa’s testimony that the only thing Cormetech could have done differently was to throw up a flag that no catalyst would work ignores the fact that it was B&W and KCP&L who determined, months before B&W issued its Purchase Order to Cormetech, that an SCR system *1033(by definition a catalyst-based system) would be installed at the KCP&L power station. Cormetech was not asked to provide advice as to whether an SCR system would be appropriate for the La Cygne power station; Cormetech was asked to supply catalyst for an SCR system that B&W and KCP&L had already determined to build. Niksa’s report and deposition are evidence of a defect in the cyclone burners and/or an act or omission by KCP&L or B&W; they are not evidence of a defect in the catalyst or of an act or omission by Cormetech.
In sum, the evidence cited by Cormetech shows that its catalyst was not defective and B&W has offered no evidence that creates a genuine issue of material fact regarding a defect or an act of omission by Cormetech. Accordingly, Cormetech is entitled to summary judgment on B&W’s claim for indemnification. See Matsushita Elec. Indus. Co., 475 U.S. at 586, 106 S.Ct. 1348 (“When the moving party has carried its burden under Rule 56(c), its opponent must do more than simply show that there is some metaphysical doubt as to the material facts.”).
IV. Conclusion
For the reasons explained above, Cor-metech’s Motion for Summary Judgment (Doc. 98) is GRANTED because its breach of warranty claims are time-barred and its evidence does not show that there exists a genuine issue of material fact regarding its claim for indemnification.
IT IS SO ORDERED.