ORDER DENYING MOTION FOR CLASS CERTIFICATION

HomeTeam Pest Defense developed a pest control system that involves building homes with tube systems in the walls. The benefit of the tube system is that pesticides can be sprayed into the tubes from a port on the outside of the house. After being sprayed into the port, the pesticides exit the tubes through various perforations, killing pests inside the walls. The pesticides apparently don’t find their way into the living area of the home.

HomeTeam enters into deals with home-builders for installation of the tube sys- *1157 terns as the homes are being built. Then HomeTeam enters into contracts with homeowners to “service” the systems (that is, to periodically spray pesticides into the ports).

The plaintiffs in this case are Jose Luis Garnica, who lives in Fresno, and Cora Potter, who lives in Bakersfield. Both their homes have tube systems. And both hired HomeTeam to service their systems. Gar-nica and Potter contend that HomeTeam has violated the antitrust laws by engaging in a variety of improper conduct to prevent competitors from entering the market for servicing the tube systems, which in turn has forced homeowners to pay HomeTeam supracompetitive prices for tube service. In a related case that is proceeding on the same track and in the same court, a company called Killian Pest Control, which has tried to compete with HomeTeam to service tube systems in Fresno and Bakersfield, has sued HomeTeam alleging the same antitrust violation.

The lawsuit by Garnica and Potter is a proposed class action, and this is a motion for class certification. Garnica and Potter contend there are 32 distinct geographic markets for servicing tube systems throughout the country. One of those markets, they contend, includes both Fresno and Bakersfield. 1 But Garnica and Potter don’t merely seek to represent a class of homeowners in that market; they seek to represent all homeowners with tube systems in all 32 alleged geographic markets.

The plaintiffs’ presentation in support of class certification is rife with problems, and it’s questionable whether any of their expert opinions would be admissible at trial. But in any event, their presentation fails at the threshold: The first element of the plaintiffs’ antitrust claim is monopoly power—that is, whether HomeTeam has the power to control prices or exclude competition. United States v. Grinnell Corp., 384 U.S. 563, 571 , 86 S.Ct. 1698 , 16 L.Ed.2d 778 (1966). To satisfy this element, they must: (1) define the relevant product market and geographic markets; (2) show that HomeTeam owns a dominant share in the relevant markets; and (3) show that there are “significant barriers to entry” in the markets such that “existing competitors lack the capacity to increase their output in the short run.” Rebel Oil Co., Inc. v. Atl. Richfield Co., 51 F.3d 1421, 1434 (9th Cir. 1995). Assuming for the sake of argument that the plaintiffs and their experts have defined the product market correctly (that is, as the market for tube servicing), they have made no meaningful attempt to show that competitive conditions with respect to tube servicing are similar in each of the 32 geographic markets. Instead, the plaintiffs rely almost exclusively on an assessment of market conditions in the aggregate. As a result, they have not met their burden to show that key questions about monopoly power in the 32 markets—for example, whether HomeTeam owns a dominant share of the tube servicing markets or whether there are significant barriers to entry into those markets—are subject to “common answers.” Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 350 , 131 S.Ct. 2541 , 180 L.Ed.2d 374 (2011); see also Heerwagen v. Clear Channel Cmmc’ns, 435 F.3d 219 , 228-29 (2d Cir. 2006); Rodney v. Northwest Airlines, Inc., 146 Fed.Appx. 783, 787-88 (6th Cir. 2005); Blades v. Monsanto Co., 400 F.3d 562, 570, 574 (8th Cir. 2005); In re Cox Enters., Inc. Set-Top Cable Television Box Antitrust Litig., No. 09-ml-2048-C, 2011 WL 6826813 , at *11-13 (W.D. Okla. Dec. 28, 2011).

*1158 For example, the plaintiffs’ experts assume, based on data obtained from Home-Team, that [redacted]% of tube systems are currently serviced by HomeTeam. 2 The plaintiffs’ experts call this the “capture rate,” and they rely on this figure to reach their ultimate conclusion that Home-Team has a monopoly throughout the country. But the experts don’t consider capture rates in the different geographic areas. For example, the “capture rates” in [redacted] and [redacted] are roughly [redacted^ each. The “capture rates” for [redacted] and [redacted] are roughly [redacted]%. The plaintiffs’ reliance on the aggregate figure prevents a proper comparison of conditions in the different markets. 3

The plaintiffs’ “switching rate” runs into the same problem. To come up with this rate, the plaintiffs’ experts examined HomeTeam customer surveys, including a “Cancel Customers Survey.” The survey identifies one of eight possible reasons given by a customer for cancelling tube service from HomeTeam. Among these reasons are: “service,” “financial,” “price-increase related,” and “competitor.” The plaintiffs’ expert report notes that [redacted]% of respondents throughout the country gave “competitor” as the reason for cancelling, which, when compared against the percentage of all “uncaptured” Home-Team customers ([redacted]%), allegedly translates into just [redacted]% of all tube-in-wall homeowners using competitor services. But the alleged [redacted]% aggregate “switching rate,” in addition to being based on the flawed assumptions discussed in footnote 4, reveals little about competitive conditions in individual geographic markets. In fact, the survey data upon which the plaintiffs rely vary significantly by HomeTeam branch. For example, in 2014, although fewer than [redacted]% of customers from the [redacted], [redacted], and [redacted] branches gave “competitor” as the reason for cancelling, more than [redacted]% of customers from the [redacted], [redacted], [redacted], and [redacted] branches gave “competitor” as the reason for cancelling. 4

In [redacted], [redaeted]% of customers reported canceling HomeTeam’s services due to a “competitor.” Apparently this is no accident. There is evidence that, in the geographic market identified by the plaintiffs that includes [redacted], several other pest control companies are in the business of servicing tube systems. For example, [redacted] in [redacted] advertises: “[re *1159 dacted].” But the plaintiffs made no effort to identify other companies that service tube systems in the [redacted] market, or to analyze the impact those other companies may have on the [redacted] market. Nor did they attempt to do so for any of the other geographic markets. Indeed, in discovery, HomeTeam gave the plaintiffs a list of 50 pest control companies that Hom-eTeam believed offered tube system servicing, but the plaintiffs’ experts conducted no real analysis of this, instead blithely waving it away.

The plaintiffs’ failure to conduct a market-by-market analysis is compounded by their failure to even delineate the individual geographic markets properly. A “geographic market” in an antitrust case is defined by the area of competition—the geographic area where customers can access alternative sources of supply. Saint Alphonsus Med. Ctr.-Nampa Inc. v. St. Luke’s Health Sys., Ltd., 778 F.3d 775, 784 (9th Cir. 2015). So, for example, if customers in City A can turn to companies W, X, and Y as alternatives, and customers from City B can turn only to company Z as an alternative, City A and City B will probably not be part of the same geographic market. But the plaintiffs’ experts, in defining their 32 geographic markets, did not use this legal standard. Instead, their markets are defined primarily by the areas serviced by different HomeTeam branches, without regard to the presence or absence of alternative sources of supply within different parts of those areas. For example, the plaintiffs’ experts consider Houston and New Orleans to be part of the same geographic market for tube servicing (presumably because the same HomeTeam branch provides service in those cities), even though the cities are over 300 miles apart. Indeed, in Garnica and Potter’s proposed home market, which includes both Fresno and Bakersfield, one of the key competitors to service tube systems (the aforementioned Killian Pest Control) responded to service calls in five zip codes. But there are at least 30 zip codes in Fresno and Bakersfield alone, not to mention the zip codes in other parts of this alleged geographic area, which means that competitive conditions are likely different in the other parts of this Bakersfield-Fresno geographic area. 5 The failure to properly delineate the individual geographic markets further precludes an assessment of whether the antitrust questions presented with respect to each individual market are susceptible to common answers. See Bailey v. Allgas, Inc., 284 F.3d 1237, 1249 (11th Cir. 2002) (“The law is clear ... that a geographic market cannot be drawn simply to coincide with the market area of a specific company.”).

Against all this, the plaintiffs contended at oral argument (for the first time) that the need for a market-by-market analysis to assess whether the monopolization question is susceptible to common answers is obviated by the fact that HomeTeam’s overall servicing business (including tube servicing and conventional pest control servicing) enjoys similarly high gross margins in all 32 markets compared to the gross margins for conventional pest control services provided by other companies. But inferring market power from gross margins is a dicey proposition, and high gross margins are generally not by themselves sufficient to prove such power. See Dial Corp. v. News Corp., 165 F.Supp.3d 25 , 42 n.6 (S.D.N.Y. 2016); see also Blue Cross & Blue Shield United of Wis. v. Marshfield Clinic, 65 F.3d 1406 , 1412 (7th Cir. 1995) (“[I]t is always treacherous to try to infer monopoly power from a high rate of return.”). If anything, the evidence *1160 of differing levels of competition in different markets suggests that HomeTeam’s allegedly high gross margins across geographic markets is attributable to something other than the amount of competition for tube servicing. For example, using the plaintiffs’ figures, HomeTeam enjoys a higher margin in the market that contains [redacted], where there appears to be tube servicing competition, than in other markets where there appears to be less tube servicing competition (at least according to the plaintiffs’ capture and switching rates). Moreover, the plaintiffs seem to be comparing apples and oranges when they present their “gross margin” analysis. Presumably, a true comparison of gross margins would ask whether HomeTeam’s tube servicing margins were significantly higher than other companies’ tube servicing margins (which might tend to show that there are barriers to entry to the servicing market). But the plaintiffs didn’t bother to look for other companies that offer tube servicing, so instead their experts compared HomeTeam’s overall pest control servicing margins (that is, for tube servicing and conventional pest control) to other companies’ margins for the provision of conventional pest control only.

In sum, the failure by the plaintiffs to consider variance among markets (not to mention within individual markets), combined with the evidence HomeTeam has submitted to suggest that such variance exists, has prevented the plaintiffs from meeting their burden to show that common issues predominate. It is far more likely, based on the evidence in the record (not to mention common sense), that the inclusion of 32 geographic markets in this proposed class action would require the jury to ask many questions that are not susceptible to common answers. See Heerwagen, 435 F.3d at 228-29; Rodney, 146 Fed.Appx. at 787-88 ; Blades, 400 F.3d at 570, 574 ; In re Cox Enters., Inc., 2011 WL 6826813 , at *12 (“Essentially, Plaintiffs ask the Court to ignore the economic reality that Cox customers face different options depending on where they live.... Cox’s national conduct regarding set-top boxes does not convert a local geographic market into a national one, and the determination of Cox’s market power should not be divorced from the appropriate geographic market.”).

Nor does In re Northwest Airlines Corp., the principal case on which the plaintiffs rely, change the analysis. 208 F.R.D. 174 (E.D. Mich. 2002). In Northwest Airlines, the court certified a class of plaintiffs asserting antitrust violations in 234 different “geographic markets,” where each market was defined as the flight route between an airline hub and a destination city. Id. at 179 & n.4. Although the Northwest Airlines decision may well have been erroneous for some of the reasons discussed above, there was greater justification for class certification in that case than this one. In Northwest Airlines, the plaintiffs’ experts at least attempted to conduct a true market-by-market analysis. In a separate order not cited by Garnica and Potter here, the court in Northwest Airlines explained that the testimony of the plaintiffs’ expert “refuted” the defendants’ argument that the expert did not perform “the necessary route-by-route analysis of the Airlines’ market power.” In re Northwest Airlines Corp. Antitrust Litig., 197 F.Supp.2d 908, 920 (E.D. Mich. 2002). The plaintiffs’ expert in Northwest Airlines testified that, for each of the 234 relevant markets he identified, he had “reviewed DOT origin/destination data to determine the identities and sizes of Defendants’ competitors, the lengths of time these competitors were present in the market, and the likelihood that they could sustain this presence and affect pricing.” Id. Here, the plaintiffs’ experts considered no tube servicing competitors other than Killian. Therefore, even if Northwest Airlines was correctly decided, it serves only *1161 to highlight the deficiencies in the plaintiffs’ presentation here. 6

The motion for class certification is therefore denied. At the upcoming case management conference, the Court will discuss with the parties whether it would be unfairly prejudicial to HomeTeam, at this late stage, to allow the plaintiffs to seek to certify a class of plaintiffs whose homes are in the geographic markets (as properly defined) in which the named plaintiffs reside.

IT IS SO ORDERED.

. Plus Ventura County, as well as portions of Madera County and the Visalia-Porterville-Hanford area.

. Portions of this Order are redacted to reflect information subject to a sealing request. An order regarding the pending motions to seal is forthcoming.

. There appears to be another major shortcoming in the plaintiffs' use of "capture rate,” although perhaps one that relates more to the merits than class certification. Although the plaintiffs’ experts assume that the data from HomeTeam reflects that [redacted]% of tube systems are currently served by HomeTeam, the data appear only to show that [redacted]% of tube systems have been served by Home-Team at least once at some point in time.

. As with the capture rate, there are other problems with the plaintiffs’ “switching rate.” First, the switching rate is based in part on the capture rate, and thus, as discussed in the preceding footnote, is built on the apparently incorrect assumption that [redacted]% of tube systems are currently serviced by HomeTeam. Second, with respect to the [redacted]% figure from the "Cancel Customer Survey,” many customers who were reported as can-celling for other reasons (such as "financial” and "service”) may naturally have gone on to use a competitor. Indeed, the HomeTeam report on cancellations that the plaintiffs rely on states: "Of the former customers that currently are using another pest control company for their needs ([redacted]%) only about [redacted] ([redacted]%) were coded in the database as leaving for a competitor. Another [redacted]% were coded as service related and [redacted]% as moving or for financial reasons.”

. The U.S. Postal Service (USPS.com) offers a list of the zip codes associated with Fresno and Bakersfield. See https://tools.usps.com/go/ ZipLookupActionünput.action.

. There also appear to have been fewer manageability concerns in that case, as the “Airline's purported monopoly power in a given city-pair market” might have been assessed "largely through consideration of market conditions at the hub point of origin or destination,” and there were only seven hubs. In re Northwest Airlines Corp., 208 F.R.D. at 217 . Here, the geographic markets don’t overlap at all.