OPINION & ORDER
JAMES D. PETERSON, District JudgePlaintiff Epic Systems Corporation is suing defendants YourCareUniverse, Inc., MEDHOST of Tennessee, Inc., and MEDHOST Direct, Inc. under both state and federal law on the grounds that defendants’ YOURCAREEVERYWHERE mark infringes Epic’s trademark for CARE EVERYWHERE and results in unfair competition. Defendants have moved for summary judgment, Dkt. 113, arguing that no reasonable jury could find that their mark is likely to confuse potential customers, which is an element of all of Epic’s claims. In the alternative, defendants seek summary judgment on Epic’s request under 15 U.S.C. § 1117(a) for attorney fees and enhanced damages.
The primary fact in Epic’s favor is that both marks include the same words. But nearly all the other factors governing a likelihood of confusion analysis favor defendants or are neutral. Although Epic and MEDHOST offer some of the same products, the marked products (an “interoperability” application and a health and wellness website) are very different. There is no evidence that anyone has been confused by defendants’ mark and there is strong evidence that defendants did not intend to copy Epic’s mark or use YOURCAREEV-ERYWHERE to deceive potential custom*881ers. Several other factors favor defendants as well: the marks use commonplace words; the context of defendants’ mark generally makes its source clear; prospective customers generally use a great deal of care in choosing the relevant products and services; and the parties have little overlap in their customers. Even if it is assumed that potential customers associate CARE EVERYWHERE with Epic’s product, the possibility of customer confusion is simply too remote and speculative to require a trial. Accordingly, the court will grant defendants motion for summary judgment, which makes it unnecessary to consider defendants’ alternative argument regarding Epic’s request for relief under § 1117(a).
Several other motions are before the court as well: (1) defendants’ motion to exclude the testimony of one of Epic’s experts, Michael Cohen, Dkt. 118; (2) defendants’ motion to strike Epic’s jury demand, Dkt. 120; (3) Epic’s motion to amend its damages report, Dkt. 194; and (4) defendants’ motion for leave to amend their answer to assert the affirmative defense of abandonment, Dkt. 202; (5) defendants’ motion to compel discovery, Dkt. 224; (6) Epic’s motion for an extension of time to respond to defendants’ motion to compel, Dkt. 226; and (7) Epic’s motion to compel discovery, Dkt. 228.
The court will deny all of these motions as moot. Epic did not rely on Cohen’s testimony or opinions in its own proposed findings of fact. Although Epic cited Cohen’s testimony a few times in responses to defendants’ proposed facts, none of that testimony makes any difference to the outcome of defendants’ summary judgment motion. The remaining motions have no bearing on summary judgment, so it is unnecessary to consider them.
UNDISPUTED FACTS
The undisputed facts are taken from the parties’ proposed findings of fact and the record. The court has disregarded proposed facts that did not comply with the court’s procedures. For example, when responding to the other side’s proposed findings of fact, the parties often included additional facts that were not responsive to the original fact.1 Dkt. 45, at 14 (“When a responding party disputes a proposed finding of fact, the response must be limited to those facts necessary to raise a dispute. The court will disregard any new facts that are not directly responsive to the proposed fact.”). The court also excluded vague or conclusory proposed facts and responses that did not comply with the requirement to provide “specific facts” at summary judgment.2 Drake v. Minn. Mining & Mfg. *882Co., 134 F.3d 878, 887 (7th Cir. 1998) (“Rule 56 demands something more specific than the .bald assertion of the general truth of a particular matter[;] rather.it requires affidavits that cite specific com crete facts establishing the existence of the' truth of the matter asserted.”).
A. Epic Systems Corporation
1. General background
Plaintiff Epic Systems Corporation is a healthcare information technology vendor with about 10,000 employees. Epic’s primary product is “EpicCare,” an electronic health records (EHR) system. Epic'provides one version of its system to hospitals and another version to clinics.
Among other things, an EHR system such as EpicCare may include medical information about a patient, allow access to evidence-based tools that healthcare providers can use to make decisions about a patient’s care, and automate and streamline scheduling and billing. One purpose o'f an EHR system is to allow providers across more than one healthcare organization to share and exchange patient information. This sharing of digitized records across providers is sometimes called “interoperability.”
2. CARE EVERYWHERE
CARE EVERYWHERE is Epic’s interoperability “application” used to exchange patient data between and among healthcare institutions.3 It allows patient data to be exchanged between various end points, which include Epic facilities, non-Epic facilities, personal health records, health information exchanges, and various governmental entities.
.In May 2004, Epic applied for federal registration of CARÍE EVERYWHERE. In June 2005, the United States Patent and Trademark Office registered the mark. . In June 2011, the USPTO accepted Epic’s Section 8 and 15 application, rendering Epic’s mark “incontestable.” Epic has’ not registered CARE EVERYWHERE in Wisconsin..
From 2009 to 2014, Epic also used CARE EPIC and CARE ELSEWHERE in conjunction with CARE EVERYWHERE. Epic stopped using these other two marks because some had expressed confusion about the three' marks. Some people referred to one or more of the marks incorrectly as “Care Anywhere”' and others simply’had difficulty remembering the CARE EVERYWHERE name.
The purpose of CARE EVERYWHERE is to “help traveling patients get care regardless of where they are.” Dkt. 169, ¶ 28. The name “Care Everywhere” is meant to evoke “something to both patients and providers,' which is anywhere you get your care, there can be [a] composite record of all of that care in one place. So anywhere you go, we can pull information from CARE EVERYWHERE.” Id, ¶29,
CARE EVERYWHERE is part of EpicCare, which hospitals and clinics must license to have access to CARE EVERYWHERE. Along the same lines, a healthcare provider. cannot gain access to CARE EVERYWHERE unless he or she works at a facility that has a license to use EpicCare. A provider using an EHR system other than Epic’s must gain access to CARE EVERYWHERE through another Epic product. Epic uses the *883CARE EVERYWHERE mark in the context of training, installation, implementation, and maintenance related to the, interoperability application.
3.Other Epic products used with CARE EVERYWHERE
CARE EVERYWHERE’s interoperability application may be used “in connection with” individual modules that are also part of EpicCare, including a personal health record management tool, clinical and administrative modules used by healthcare professionals, and a population health management system. Dkt. 170, ¶ 23. One of the modules is an application called MyChart, which is a “shared patient record” or “patient portal.” MyChart “offers patients personalized and secure online access to portions of their medical records and permits patients to include and share with providers disease/health and wellness information.” Dkt. 170, ¶ 32. MyChart permits patients to review then-test results and visit summaries, schedule appointments, download records, and communicate directly with their physicians. MyChart is available only to patients' of healthcare providers who are affiliated with Epic. Within the past three years, CARE EVERYWHERE ^functionality” was added to MyChart. This allows for the exchange of records with a different provider. A -patient may see the CARE EVERYWHERE logo in MyChart when using this functionality.
Another module is called “Lucy,” which allows patients to collect their records in one place. CARE EVERYWHERE allows patients using Lucy to transfer records to different providers or to a MyChart account.
4. Customer use of CARE EVERYWHERE
Epic allows its customers to use its marks on their consent forms and websites so- long as they properly attribute the mark to Epic and'otherwise use the mark appropriately. Out of 18 customers for which Epic provided defendants relevant documents, seven of those customers had used CARE EVERYWHERE incorrectly, either not attributing the mark to Epic or attributing the mark to: the wrong products and services.4 Epic relies on its employees to detect and correct customer noncompliance.
5. Direct customers
Epic has approximately 400 customers that obtain a license directly through Epic. The vast majority of these customers are large healthcare organizations. Other customers, include “community” hospitals, retail clinics, and independent practices. Most of the smaller organizations share patients, with Epic’s larger customers. The smallest facility has 100 beds. Epic’s direct customers are “sophisticated” and they put “a lot of care and resources” into choosing an EHR system. Dkt. 96 (Faulkner Dep., at 270); Dkt. 91 (DeVault Dep., at 222).
The sales process often begins with a prospective customer issuing a “request for proposal” to various health IT vendors. After the customer narrows the list of vendors, those on the short list will give product demonstrations. The process for purchasing an EHR system usually lasts six months to one year. During' the sales process, Epic typically' interacts with high level executives and representatives of practitioners. Many of Epic’s potential'customers also use consultants who are familiar with and experienced in the healthcare *884IT industry and the request-for-proposal process. The chief information officer and other high level executives are involved in the final decision to purchase an EHR system from Epic because of the significant cost of the system, the necessary hardware to implement the system, and the training needed to use the system. The licensing fee can range from $1.5 million for a small customer to $20 million for a large customer. Because CARE EVERYWHERE is included in EpicCare, there is no separate fee for the application.
No organization that has purchased Epic’s EHR system has elected to switch to a different vendor later.
6.Community Connect program
For approximately 10 years, Epic has had the Community Connect program, which allows smaller hospitals and physician practices to use the Epic EHR System. Typically, Connect participants are small independent practices with one to five providers or small community hospitals with about one to 200 beds. At least 1,500 organizations participate in the Connect program. Under the program, a Connect participant is able to use the Epic software through Epic’s direct customers in the same ways that Epic’s direct customers use the software. Connect participants can receive training and support from Epic.
Generally, Connect participants have a contract with one of Epic’s larger customers rather than with Epic itself. However, Epic may receive a financial benefit from Connect participants because of increased usage by the direct customer. Connect participants are ordinarily geographically near to a direct customer and share patients with that customer. Prospective Connect participants generally hear about Epic through interactions with Epic’s direct customers.
7. Epic’s marketing and sales
Epic and its CARE EVERYWHERE products and services are “widely known” in the healthcare marketplace. Dkt. 170, ¶46. Prospective customers “often” mention CARE EVERYWHERE as an aspect of the Epic EHR system that they have heard about and are interested in. Most people in the industry, including prospective customers, already know what CARE EVERYWHERE is when they contact Epic. However, Epic has never conducted any survey, poll, search, study, or other investigation to determine whether CARE EVERYWHERE has acquired secondary meaning in the minds of consumers.
Epic does not have a marketing department. It does not buy advertising space or cold call potential customers. Most customers contact Epic first, often because of a recommendation from another customer. Epic promotes its products and services primarily through word of mouth and the results from rating agencies and trade shows. Epic’s “Events and Design Team” promotes Epic’s products and services at user group meetings and industry conferences. Since 2014, Epic has spent approximately $100 million on marketing and promotion of its products.
Epic does have a sales team. Generally, the team responds to inquiries made from prospective customers rather than the other way around. Initial discussions are held by telephone and then may lead to face-to-face meetings. When Epic meets with a prospective customer, Epic may recommend the Community Connect program if the customer is not the appropriate size or does not have a large IT department.
8. Patients’ exposure to CARE EVERYWHERE
Epic does not market or sell the CARE EVERYWHERE product or any other product or service to healthcare patients. *885Patients might see the mark when completing consent forms regarding the sharing of data, when using the MyChart patient portal, or on an Epic customer’s website. When the mark appears on the MyChart portal, it appears along with the CARE EVERYWHERE logo:
[[Image here]]
B. Defendants YourCareUniverse, Inc., MEDHOST of Tennessee, and MEDHOST Direct
1. General background
Defendant YourCareUniverse, Inc. is a subsidiary of MEDHOST Solutions Corp., which is a subsidiary of MEDHOST, Inc. YourCareUniverse “offers a range of software tools and services that assist healthcare providers in managing the business of healthcare.” Dkt. 169, ¶ 115.
Defendant MEDHOST of Tennessee, Inc. sells an EHR system called Enterprise. Defendant MEDHOST Direct is a subsidiary that provides hosting and managed services both internally to MED-HOST affiliates and externally to customers of MEDHOST. All of these companies are “related or affiliated corporate entities.” Dkt. 160, ¶ 5.
2. YOURCAREEVERYWHERE
In 2013, defendants began planning for products using different marks beginning with ‘YourCare.” The first product, launched in February 2014, was a patient portal called YOURCARECOMMUNITY (since renamed YOURCAREHEALTH). This was followed by YOURCARELINK, which is an “interface engine” that allows the transfer of information to a state health reporting agency. By October 2014, YOURCAREEVERYWHERE was the proposed named for a health and wellness website and related services.
After learning that yourcareevery-where.com was an available web domain, defendants conducted a search for similar marks. Defendants discovered Epic’s CARE EVERYWHERE mark at that time.
Defendants hired an advertising agency to perform a survey to test the name YOURCAREEVERYWHERE, along with two other names. Sixty five percent of the respondents chose YOURCAREEVERY-WHERE as their preferred name.
Defendants launched the youreareevery-where.com website in March 2015. At the same time, they executed a marketing campaign for the website, spending approximately $2,500,000. When advertising products and services for YOURCAR-EEVERYWHERE, defendants use the YOURCAREEVERYWHERE mark. In most of the advertisements, the mark appears alongside the YOURCAREUNIV-ERSE mark, the MEDHOST hark, or other ‘YourCare” marks.
Epic has identified two instances in which the YOURCAREEVERYWHERE mark has appeared without being clearly unaccompanied by one of defendants’ other marks. Dkt. 145-65 and 145-66. (Epic cites two other documents, Dkt. 145-28 and Dkt. 145-29, but either the MEDHOST mark or numerous YOURCAREUNIV-ERSE marks are prominent.) One is a brochure for YOURCAREEVERY-WHERE “engagement services.” (The *886MEDHOST mark appears at the bottom of the brochure, but it is not prominent.) The other is a press release announcing the YOURCAREEVERYWHERE website. (The YOURCAREUNIVERSE mark appears in the document, but again, it is not prominently displayed.) In both documents, the YOURCAREEVERYWHERE mark appears with a YOURCAREEVER-YWHERE logo, Epic has identified three documents (two press releases and a brochure) in which the YOURCAREEVERY-WHERE mark did not appear with the logo. Dkt. 169, ¶ 144. In each document without the logo, the MEDHOST mark, the YOURCAREUNIVERSE mark or both are prominent.
In April 2016, the USPTO allowed defendants’ YOURCAREEVERYWHERE mark over Epic’s CARE EVERYWHERE mark,
3. Other “YourCare” marks
Eventually, defendants chose to brand a number of other products as part of YOURCAREUNIVERSE and identify all of them with a different ‘YourCare” mark. Defendants now use many other marks with the ‘YourCare” prefix, including the following:
. YOURCAREANALYTICS
• YOURCARELINK.
. YOURCAREREFERRAL
• YOURCARETRANSFER
* YOURCAREHEALTH
• YOURGAREPROVIDER
• YOURCARESUCCESS
. .YOURCAREDATA
. YOURCAREEXCHANGE
• YOURCAREINTERACT
. YOURCARENAVIGATION
• YOURCAREWELLNESS
. YOURCARECOMMUNITY
• yourcareknowledge
• YOURCAREMESSENGER
• YOURCAREVISUALIZE
The Trademark Trial and Appeal Board rejected defendant YourCareUniverse’s application to register ‘YourCare” on the ground ■ that it is descriptive. The board also rejected YourCareUniverse’s contention that it owned a “family of marks.”
Each of the ‘YourCare” products and services can be purchased separately and function independently, both as to each other and as to MEDHOST’s EHR system, For example, defendants’ YOUR-CAREEVERYWHERE website can link with any health portal, regardless whether the portal is affiliated with defendants. Defendants’ YOURCAREHEALTH . patient portal can be used without MED-HOST’s EHR system. However, when cus» tomers of YourCareUniverse purchase products and services related to YOUR-CAREEVERYWHERE, they always purchase- at least one of defendants’ other products as well, though they are not required to do so.
4. Products and services associated with YOURCAREEVERY-WHERE
YOURCAREEVERYWHERE is used on a free, publicly available website that contains general information related to health and wellness, such as-diseases and conditions, healthcare news, pregnancy and childbirth, child and teen care, heart care, mental health, and exercise and nutrition. The mark is displayed with this logo:
*887[[Image here]]
There is a YOURCAREEVERY-WHERE mobile app as well. On the app, the mark is displayed with this logo:
[[Image here]]
Users of the website or app may register for an account, which allows them to track certain health and wellness data, such as data from a fitness band.
Healthcare providers can license a co-branded landing page that includes the YOURCAREEVERYWHERE mark and logo, along with information from the provider. If a patient of that provider has an account with YOURCAREEVERY-WHERE, the patient can link the account with the provider’s website and access the provider’s YOURCAREUNIVERSE patient portal.
Defendants offer marketing consultation services to healthcare providers under the YOURCAREEVERYWHERE mark.
Defendants’ revenue for YOURCAR-EEVERYWHERE comes from a combination of third-party advertisements on the website, co-branded landing pages, and marketing consultation services.
5. Defendants’ customers
Approximately half of the customers for Enterprise (MEDHOST’s EHR system) are specialty hospitals or special providers and the other half are short-term acute care facilities. The “bulk” of MEDHOST’s 700 to 800 Enterprise customers are “middle tier” hospitals. Approximately 10 to 15 percent .are “small” hospitals. MEDHOST does not “really compete ... very much” for “large tertiary care” hospitals, Dkt. 88 (Anderson Dep., at 168-69), but it does have some large healthcare facility customers. The average Enterprise license costs $400,000.
6. Defendants’ marketing and sales
MEDHOST promotes YOURCAREU-NIVERSE products and services to hospitals in conjunction with its promotion of Enterprise as a way to engage patients and potential patients, promote patient loyalty, expand the company’s brand presence and encourage patients to use the healthcare facilities’ portal. In accordance *888with this purpose, MEDHOST markets YOURCAREUNIVERSE products and services as part of a broader “patient engagement solution” to “open the door” to sell other MEDHOST products, including its EHR system. Dkt. 170, ¶73. MED-HOST receives a subscription fee from any customer who buys the ‘YOURCAREU-NIVERSE suite,” which includes the YOURCAREEVERYWHERE “consumer engagement solution.”
MEDHOST promotes its products and services at national conferences, some of which Epic attends as well.
C. Third party uses of “care” and “everywhere”
In the healthcare industry, the word “care” is commonly used and “everywhere” is “a fairly common term,” though not as common as “care.” Dkt. 105 (Kiesau Dep., at 137-38). Since 2004, the careevery-where.com domain name has been owned by Care Everywhere, LLC, a healthcare IT vendor that has no affiliation with Epic. The company identifies itself as “a certified software medical device company focused on integrating point-of-care medical devices such as infusion pumps and vital sign monitors with hospital information systems to improve patient safety, nursing productivity, and documentation.” Care Everywhere, “About,” available at http:// www.careeverywhere.com/ce/about/ (last visited March 14, 2017).
The following healthcare organizations use the phrase “care everywhere” as well:
• Primary Care Medical Center has a website called primarycareevery-where.com;
• a public charity called Kids Care Everywhere has a website called Md-scareeverywhere.org;
• a company called CommuniTake uses the phrase “Mobile Care Everywhere”;
• a home healthcare provider uses the phrase “Quality Care Everywhere”;
• United Healthcare uses the phrase “Health Care Everywhere” on their website and marketing materials;
• The Washington State Hospital Association uses the phrase “Essential Care, Everywhere” on their website;
• The U.S. National Institute for Health uses the phrase “Remote Access to Care Everywhere”;
• Axxess Technology Solutions uses the phrase “Powering Care Everywhere.”
Other registered marks in healthcare industry include EXPERT CARE EVERYWHERE, EVERYWHERE CARE, and CARE. VIRTUALLY EVERYWHERE.
D. Potential competition between Epic and defendants
Epic does not offer a website that includes general health and wellness content, under its CARE EVERYWHERE mark or otherwise.
According to Epic’s CEO, companies called Cemer, ALLscripts, MEDITECH, and McKesson are Epic’s “direct” competitors, meaning that they appear on shortlists with Epic after a prospective customer issues a request for proposal. Ninety percent of the time, Epic competes with Cerner or Allscripts. Epic has identified two instances in which it competed with MEDHOST for a contract ■. regarding an EHR system. In both instances, the company was a “long-term acute care” and “rehab” business. Dkt. 10SS (Hutchinson Dep., at 56).5
*889As of March 2016, Epic had 38 percent of the market share for EHR systems in all physician practices, rehab and long-term care facilities, and hospitals of all bed sizes. MEDHOST’s market share was approximately one percent. As to hospitals having more than 200 beds, Epic had 32 percent of the market and MEDHOST had approximately one percent.
E. Confusion about the parties’ marks
Epic has not identified any instances of actual confusion between the CARE EVERYWHERE and YOURCAREEVERY-WHERE marks, products, or services. Epic’s CEO believes that defendants’ YOURCAREEVERYWHERE mark is likely to be confusing for Epic’s Community Connect participants, but not its direct customers. Dkt. 96 (Faulkner Dep., at 31-32, 270-71). Epic has not conducted a survey to determine whether any of its customers are likely to be confused.
ANALYSIS
A. Legal standard
Epic is asserting claims for trademark infringement and unfair competition under both state and federal law. The parties agree that each of these claims requires proof of a likelihood of confusion and that the standard under both state and federal law is the same. Fortres Grand Corp. v. Warner Bros. Entm’t Inc., 763 F.3d 696, 700 (7th Cir. 2014) (likelihood of confusion is element of both trademark infringement and unfair competition claims). The parties do not cite any controlling authority for the view that Wisconsin and federal law have an identical standard for likelihood of confusion, but neither side identifies any differences and both sides cite federal law only, so the parties have forfeited any arguments related to any difference that may exist.
“Likelihood of confusion” in the context of trademark law has several components. First, the question is not whether anyone who might view the marks would be confused; rather, the relevant class consists of consumers who might purchase either the plaintiffs or the defendant’s products or services. Bd. of Regents of Univ. of Wis. Sys. v. Phx. Int’l Software, Inc., 653 F.3d 448, 455 (7th Cir. 2011). See also Dorr-Oliver, Inc. v. Fluid-Quip, Inc., 94 F.3d 376, 382 (7th Cir. 1996) (“[T]he proper inquiry centers on the confusion of consumers in the market for the particular products at issue.”).
Second, as to the type of confusion, the relevant question is whether a prospective customer is likely to believe that the plaintiff is the source of or is otherwise affiliated with the defendant’s marked products or services. Sorensen v. WD-40 Co., 792 F.3d 712, 726 (7th Cir. 2015) (“The keystone of trademark infringement is likelihood of confusion as to source, affiliation, connection, or sponsorship of goods or services.”). (There are other types of confusion in the context of a “dilution” or “reverse confusion” theory, e.g. Fortres Grand Corp., 763 F.3d at 701; Eli Lilly & Co. v. Nat’l Answers, Inc., 233 F.3d 456, 466 (7th Cir. 2000), but Epic in not asserting those theories.)
Third, as to the risk of confusion, “[possible confusion is not enough; rather, confusion must be probable.” Sorensen, 792 F.3d at 726 (internal quotations omitted). Finally, as to the extent of confusion required, it must be more than “de minimis.” Packman v. Chi. Tribune Co., 267 F.3d 628, 645 (7th Cir. 2001) (evidence of four *890confused customers not sufficient to show likelihood of confusion); Libman Co. v. Vining Indus., Inc., 69 F.3d 1360, 1363-64 (7th Cir. 1996) (plaintiffs claim'failed without evidence that “significant fraction” of consumers likely to be confused); Henri’s Food Prod. Co. v. Kraft, Inc., 717 F.2d 352, 358 (7th Cir. 1983) (survey showing 7.6% confusion rate weighed against finding of infringement).
In assessing likelihood of confusion, the Court of Appeals for the Seventh Circuit considers seven factors:
(1) the similarity between the marks in appearance and suggestion;
(2) the similarity of the products;
(3) the area and manner of concurrent use;
(4) the degree of care likely to be exercised by consumers;
(6)the strength of the plaintiffs mark;
(6) any evidence of actual confusion; and
(7), the intent of the defendant to “palm off’ his product as that of another.
Sorensen, 792 F.3d at 726.
Because these are factors not elements, no one factor is controlling and the relative importance of the factors may depend on the facts of a particular case. AutoZone, Inc. v. Strick, 543 F.3d 923, 929 (7th. Cir. 2008). See also Au-Tomotive Gold, Inc. v. Volkswagen of Am., Inc., 457 F.3d 1062, 1076 (9th Cir. 2006) (“The list of factors is not a score-card—whether a party wins a majority of the factors is riot the point. Nor shorild the factors be rigidly weighed; we do not count beans.”) (internal quotations omitted). However, the Seventh Circuit' has stated that “the similarity of the marks, the defendant’s intent,' and actual confusion aré particularly important.” AutoZone, 543 F.3d at 929. (The Seventh Circuit has said that “the list is not exclusive,” Nike, Inc. v. Just Did It Enterprises, 6 F.3d 1225, 1228 (7th Cir. 1993), but the parties do. not discuss other factors in their briefs, so the court will limit its analysis to the above list.)
Likelihood of confusion is a question of fact, which means that summary judgment is appropriate only if no reasonable factfinder could find in favor of the nonmoving party on that issue. Sorensen, 792 F.3d at 726. Even if there are genuine issues of fact as to some of the seven factors, a court may grant summary judgment if “no reasonable jury, looking at the seven factors as a whole, could conclude that there is a likelihood' of confusion.” Id,
Epic cites several _ alternative descriptions of the summary judgment standard in. trademark cases predating Sorensen. E.g., Door Sys., Inc. v. Pro-Line Door Sys., Inc., 83 F.3d 169, 171 (7th Cir. 1996) (likelihood of confusion can be resolved on summary judgment “if the evidence is so one-sided that there can be no doubt about how the question should be answered”); McGraw-Edison Co. v. Walt Disney Prods., 787 F.2d 1163, 1167 (7th Cir. 1986) (moving party in trademark case has “strict burden”"on summary judgment; “all doubts must be resolved against” moving party). See also AHP Subsidiary Holding Co. v. Stuart Hale Co. 1 F.3d 611, 616 (7th Cir. 1993) (“[A] motion for summary judgment in 'trademark infringement cases must be approached with great caution.”); Some of the statements are reminiscent of old discrimination cases in which courts stated that summary judgment “shorild be approached with special caution” in those cases, Courtney v. Biosound, Inc., 42 F.3d 414, 423 (7th Cir. 1994), and that courts should apply the summary judgment standard with “added rigor” in that context. Sarsha v. Sears, Roebuck & Co., 3 F.3d 1035, 1038 (7th Cir. 1993).
*891In more recent years, the court of appeals has clarified that the standard for summary judgment is the same in all cases. Bagley v. Blagojevich, 646 F.3d 378, 389 (7th Cir. 2011). In particular, “[i]f a genuine dispute as to a material fact exists ... summary judgment is inappropriate. But that genuine dispute must be supported by sufficient evidence to permit a jury, to return a verdict for” the non-moving party. Id.. (internal quotations and alterations omitted). In applying that standard, courts must draw all reasonable inferences in favor of the nonmoving party, but a party cannot avoid summary judgment simply by raising “some metaphysical doubt as to the material facts.” Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 586, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986).
These principles apply to trademark law as well. The Seventh Circuit has not hesitated to decide likelihood confusion as a matter of law on summary judgment when the plaintiff has not met its burden. E.g., Sorensen, 792 F.3d at 732; Sullivan v. CBS Corp., 385 F.3d 772, 774 (7th Cir. 2004); Packman, 267 F.3d at 647; Door Sys., 83 F.3d at 173; Libman, 69 F.3d at 1362; Spraying Sys. Co. v. Delavan, Inc., 975 F.2d 387, 389 (7th Cir. 1992). See also Fortres Grand, 763 F.3d at 705-06 (affirming dismissal of trademark claim for failure state a claim upon which relief may be granted on ground that plaintiff did not plausibly allege likelihood of confusion); Eastland Music Group, LLC v. Lionsgate Entm't, Inc., 707 F.3d 869, 871-72 (7th Cir. 2013) (same).
With this standard in mind, the court will consider each of the factors relevant to determining whether there is a likelihood of confusion in this.case.
B. Actual confusion
It is undisputed that Epic has no evidence that anyone has been confused by defendants’ YOURCAREEVERYWHERE mark. Epic has neither pointed to customers expressing confusion nor conducted a survey to ■ determine whether confusion is likely.
Epic is correct that actual confusion is not required to show a likelihood of confusion, Sands, Taylor & Wood Co. v. Quaker Oats Co., 978 F.2d 947, 960 (7th Cir. 1992), but as noted above, it is one of the more “important” factors. AutoZone, 543 F.3d at 929. Obviously, evidence showing whether customers have .already been confused is highly probative of the question whether customers are likely to be confused in the future. See Eastland Music Group, 707 F.3d at 871-72 (relying on absence of actual confusion to affirm dismissal of trademark infringement claim); Libman, 69 F.3d at 1363-64 (same). In the absence of any showing of actual confusion, Epic has a heavier burden to show that other factors would allow a reasonable factfinder to render a verdict in Epic’s favor. Libman, 69 F.3d at 1363-64 (evidence of actual confusion may not be needed when “it is obvious just from comparing the products that consumers are likely to be confused as to their source”).
C. Intent
A defendant’s intent behind choosing its mark is another “particularly important” factor. AutoZone, 543 F.3d at 929. “This factor looks primarily for evidence that the defendants are attempting to ‘pass off their products as having come from the plaintiff.” Packman, 267 F.3d at 644.
Epic has adduced no evidence that de-fendánts chose their YOURCAREEVER-YWHERE mark out of a hope to confuse customers into believing that the mark is associated with Epic’s CARE EVERYWHERE mark. The evidence to the contrary is strong. First, it is undisputed that YOURCAREEVERYWHERE did not *892come out of thin air. Rather, at the time defendants chose the mark, they already had two other “YourCare” marks: YOURCARECOMMUNITY and YOURCARELINK. Abercrombie & Fitch Co. v. Moose Creek, Inc., 486 F.3d 629, 637 (9th Cir. 2007) (defendant’s previous use of similar marks supports finding of good faith); King Research, Inc. v. Shulton, Inc., 454 F.2d 66, 69 (2d Cir. 1972) (same). Second, when defendants tested the mark against other potential marks, approximately two thirds of the respondents chose YOURCAREEVERYWHERE. Universal Money Centers, Inc. v. Am. Tel. & Tel. Co., 22 F.3d 1527, 1531 (10th Cir. 1994) (defendant’s use of consumer survey to choose name supports finding of good faith). Third, defendants have spent $2,500,000 marketing products and services related to the mark. Procter & Gamble Co. v. Johnson & Johnson Inc., 485 F.Supp. 1185, 1201 (S.D.N.Y. 1979) (good faith supported by defendant’s “plan[ ] [to make] enormous expenditures in publicizing and promoting the [marked] products”). Cf. Gen. Mills, Inc. v. Kellogg Co., 824 F.2d 622, 627 (8th Cir. 1987) (intent to confuse “not likely” when defendant “conducted extensive research and development” in choosing mark). Finally, since adopting the YOURCAREEVERYWHERE mark, defendants have adopted more than a dozen additional “YourCare” marks that are part of defendants’ YOURCAREUNIVERSE products and services. This is compelling evidence that defendants’ mark had nothing to do with Epic’s CARE EVERYWHERE brand, but was simply one part of defendants’ own “YourCare” brand.
Epic argues that defendants cannot rely on its other marks because defendants have not established that they have a “family of marks,” which is “a group of marks having a recognizable common characteristic, wherein the marks are composed and used in such a way that the public associates not only the individual marks, but the common characteristic of the family, with the trademark owner.” J & J Snack Foods Corp. v. McDonald’s Corp., 932 F.2d 1460, 1462 (Fed. Cir. 1991). Epic does not develop its argument and it makes little sense. The question whether defendants’ marks are recognized as related by the public has nothing to do with whether defendants intended to copy Epic’s mark. The chronology of events surrounding defendants’ adoption of YOUR-CAREEVERYWHERE shows not that defendants’ marks are famous but that defendants had a good reason unrelated to Epic for choosing the mark.
The only evidence Epic cites in support of a finding that defendants intended to confuse customers is the fact that defendants were aware of Epic’s CARE EVERYWHERE at the time they chose to adopt YOURCAREEVERY-WHERE. However, “[p]rior knowledge of a senior user’s mark does not, without more, create an inference of bad faith.” Corbond Corp. v. Core Foam, Inc., 356 F.Supp.2d 910, 919 (W.D. Wis. 2005) (quoting Playtex Prods., Inc. v. Georgia-Pacific Corp., 390 F.3d 158, 166 (2d Cir. 2004)); Bishops Bay Founders Grp., Inc. v. Bishops Bay Apartments, LLC, 301 F.Supp.2d 901, 913 (W.D. Wis. 2003) (“[T]he intent factor requires a showing of defendant’s ‘intent to confuse customers, not merely the intent to use a mark that is already in use somewhere else.’ ”) (quoting Meridian Mut. Ins. Co. v. Meridian Ins. Grp., Inc., 128 F.3d 1111, 1120 (7th Cir. 1997)). Under the circumstances of this case, when all other facts point strongly to other justifications for choosing the mark, the court sees little probative value in defendants’ refusal to choose a different name despite its knowledge of Epic’s mark.
D. Similarity of marks
For obvious reasons, the similarity of the parties’ marks is a key part of the *893likelihood of confusion analysis. At first look, this factor appears to support Epic because defendants’ mark includes the same words as Epic’s. However, a review of the words that make up the mark is simply the beginning of the analysis, not the end.
“Trademark law protects the source-denoting function of words used in conjunction with goods and services in the marketplace, not the words themselves.” Fortres Grand Corp., 763 F.3d at 705-06. See also James Burrough Ltd. v. Sign of Beefeater, Inc., 540 F.2d 266, 275 (7th Cir. 1976) (“[T]he test is not whether the public would confuse the marks, but whether the viewer of an accused mark would be likely to associate the product or service with which it is connected with the source of products or services with which an earlier mark is connected.”). Therefore, the court “must compare the marks in light of what happens in the marketplace and not merely by looking at the two marks side-by-side.” AutoZone, 543 F.3d at 930 (internal quotations omitted). In other words, context is key. Savin Corp. v. Savin Grp., 391 F.3d 439, 458 (2d Cir. 2004) (“[T]he impression conveyed by the setting in which the mark is used is often of critical importance.”) (internal quotations omitted). Even if one mark incorporates the other, as in this ease, or even if the marks are identical, the context in which the marks appear may show that confusion is not likely. E.g., Fortres Grand, 763 F.3d at 705-06 (“[B]oth marks are merely ‘clean slate’ or ‘the- clean slate.’ But juxtaposed against the weakness of all the other factors, this similarity is not enough.”); Lindy Pen Co. v. Bic Pen Corp., 725 F.2d 1240, 1245 (9th Cir. 1984) (“The two marks viewed in isolation are indeed identical, but their similarity must be considered in light of the way the marks are encountered in the marketplace and the circumstances surrounding the purchase of the [parties’ products].”).
In this case, defendants’ fnarks most often appear with other visual cues indicating that they do not come from Epic. In particular, it is undisputed that the logos for CARE EVERYWHERE and YOUR-CAREEVERYWHERE bear no resemblance to each other and that defendants’ YOURCAREEVERYWHERE mark generally appears with other marks related to YOURCAREUNIVERSE or MEDHOST or both. These are factors that tend to reduce the likelihood of confusion. Sullivan, 385 F.3d at 777-78 (two “Survivor” marks not likely to cause confusion because consumers in marketplace never see defendant’s mark alone but always surrounded by words, “Outplay, Outlast, Outwit”); Packman, 267 F.3d at 645-46 (although words in marks were identical, no likelihood of confusion when “the appearance and placement of the words are distinct” and “[a]ll of defendants’ products— and their tags—prominently display the [defendant’s name]”); G. Heileman Brewing Co. v. Anheuser-Busch, Inc., 873 F.2d 985, 999 (7th Cir. 1989) (upholding district court’s conclusion that “prominent house mark ... [and] use of a prominent brand name ... dispelled any significant likelihood of confusion”). See also Groeneveld Transp. Efficiency, Inc. v. Lubecore Intl., Inc.; 730 F.3d 494, 515 (6th Cir. 2013) (defendant “scrupulously avoided ... confusion by choosing a starkly different logo that it prominently displays on its [products] and on all its sales and marketing literature”); Playboy Enters., Inc. v. Netscape Commc’ns Corp., 354 F.3d 1020, 1030 (9th Cir. 2004) (likelihood of confusion diminished “when mark clearly identifies its source with its sponsor’s name”); Kwik-Site Corp. v. Clear View Mfg. Co., 758 F.2d 167, 178 (6th Cir. 1985) (“The most common and effective means of apprising intending purchasers of the source of goods is a prominent disclosure on the container, package, wrapper, or label of *894the manufacturer’s or trader’s name and when that is done, there is no basis for a charge of unfair competition.”) (internal quotations . omitted), Cf. AutoZone, 543 F.3d at 930-31 (likelihood of confusion between “AutoZone” and “OilZo.ne” supported by similar logos).
Epic offers two reasons for resisting the conclusion that logos and the use of other marks obviates any possible confusion caused by defendants’ mark. First, Epic points out that the YOURCAREEVERY-WHERE mark does not always appear with a logo and it does not always appear with defendants’ other marks. Second, Epic says that a prominent logo or accompanying mark does not dispel a belief that Epic is sponsoring defendants’ products and services, even if it is clear that Epic is not the source of the mark.
As to Epic’s first contention, Epic identifies only a tiny number of examples in which YOURCAREEVERYWHERE does not appear with a logo and one or more of defendants’ marks and no instances in which YOURCAREEVERYWHERE does not appear with either the logo .or other marks. Such a small number of instances of potential confusion is not sufficient to sustain a claim. Packman, 267 F.3d at 645; Libman, 69 F.3d at 1363-64.
It is true that a logo cannot limit confusion when the mark is communicated orally rather. than visually. Meridian. Mut. Ins., 128 F.3d at 1115-16; Unity Health Plans Ins. Co. v. Iowa Health Sys., 995 F.Supp.2d 874, 889 (W.D. Wis. 2014); Corbond, 356 F.Supp.2d at 916. Although defendants give prospective customers presentations about YOURCAREEVERY-WHERE products and services, Epic does not identify any instances in which visual materials are not provided as well.
Even if the court assumes that visual cues are not always present, by Epic’s own assertion, defendants market their products and services related to YOUR-CAREEVERYWHERE along with other products from MEDHOST and YourCa-reUniverse. Dkt. 170, ¶73. Epic fails to explain how a prospective customer could believe that Epic -is the source of YOUR-CAREEVERYWHERE- if defendants are pitching potential customers an array of MEDHOST products and services at the same time, particularly those related to YourCareUniverse, which would make it clear that the “common thread” among the products was ‘YourCare” rather than “CareEverywhere.” Ty, Inc. v. Jones Grp., Inc., 237 F.3d 891, 899 (7th Cir. 2001) (customers likely to believe that “common thread" in different marks was “salient” portion of marks).
Epic again observes that defendants have not established that they have a “family of marks,” and again that observation has little relevance in this context. Regardless whether a prospective customer would know when viewing YOURCAR-EEVERYWHERE in isolation that it is part of a family of marks, he or she does not need to know this if those marks are presented together. Epic says in its brief that defendants market YOURCAREEV-ERYWHERE without other, “YourCare” products and services, but the proposed findings of fact it cites actually support the opposite conclusion. E.g., Dkt. 170, ¶ 73 (“MEDHOST markets YOURCAREEV-ERYWHERE to hospitals as part of an even broader suite of YourCareUniverse EHR products and services.”); id., ¶74 (‘YourCareUniverse- does' not' have any customers who have independently purchased YOURCAREEVERYWHERE products and services, without purchasing other YourCareUniverse and/or MED-HOST products and services.”).
Epic’s alternative contention is that accompanying . YOURCAREEVERY-WHERE with a logo or one or more of defendants’ marks does not reduce the po*895tential for a customer to believe that Epic is sponsoring defendants’ products or services. Epic cites A. T. Cross Co. v. Jonathan Bradley Pens, Inc., 470 F.2d 689 (2d Cir. 1972), in which the court stated that a pen manufacturer could not dispel confusion between the marks “Cross” and “La Crosse” by adding the words “by Bradley” to its pen because “a purchaser could well think plaintiff had licensed defendant as a second user.” Id. at 692. However, a pen cannot be compared to the parties’ products and services at issue; the potential customers in this case are not simply viewing a box of software at Best Buy with the name YOURCAREEVERYWHERE on it. Again, when YOURCAREEVERY-WHERE is being offered as a “suite” along with YourCareUniverse and MED-HOST products, it is unclear how customers could believe that defendants are simply acting as Epic’s licensees.
In sum, the words in the parties’ marks overlap, but the context in which those marks appear does not support a finding that customers are likely to be confused.
E. Customer degree of care in choosing a product or service
“A consumer exercising a high degree of care in selecting a product reduces the likelihood of confusing similar trade names.” Heartstrings, Inc. v. Heartspring, Inc., 143 F.3d 550, 557 (10th Cir. 1998). But before the court can determine the level of care that a prospective customer is likely to exercise, it is important to first define who that customer is. As noted above, the relevant class of customers consists of those who could be making a choice between a purchase from Epic or one of the defendants. Phx. Intl. Software, 653 F.3d at 455. Because Epic does not sell products or services to patients, the court need not consider whether patients who see Epic’s CARE EVERYWHERE in the MyChart patient portal might be confused by seeing defendants’ YOURCAREEVER-YWHERE on their health and wellness website or. mobile app.
The customers Epic discusses in its summary judgment materials are potential purchasers of EpicCare (Epic’s EHR system) and Community Connect participants, who use EpicCare by purchasing a license from one of Epic’s direct customers. But Epic has all but conceded that its direct customers are not likely to be confused by defendants’ YOURCAREEVERYWHERE mark in light of the high degree of care they exercise when making a purchasing decision.- When asked about whether customers would be confused by defendants’ mark, plaintiffs CEO initially testified that she believed that the mark “could be very confusing, not to our direct customer, but to the connect customer.” Dkt. 96 (Faulkner Dep., at 31-32) (emphasis added). She later qualified this statement slightly by saying that it is “not impossible” that Epic’s direct customers would be confused. Id. at 270-71. “Not impossible” is a far cry from “likely” or “probable,” which is what Epic must show. Sorensen, 792 F.3d at 726.
Even if the CEO’s testimony is not treated as a concession on, this issue, the undisputed facts show that Epic’s direct customers are not.likely to be confused because they exercise a high degree of care in choosing whether to purchase from Epic or defendants. It is undisputed that the licensing fees for an EHR system are hundreds of thousands of dollars at a minimum, that the process for purchasing an EHR system can take six months to a year, that Epic’s direct customers are sophisticated healthcare organizations, and. that many of them hire consultants to assist them in making a purchase. Dkt, 146, at 22 (“Epic does not dispute that most healthcare providers that end up buying MEDHOST’s and Epic’s EHR-products and services exercise a substantial degree *896of care in making purchasing decisions. Epic admits that the offerings are relatively expensive and the ultimate purchasers tend to be sophisticated in their purchasing decisions”) (internal citations omitted).
In other words, these are not ordinary consumers who are making a split-second decision and can be easily misled. AutoZone, 543 F.3d at 933 (confusion less likely when customers “are particularly sophisticated or deliberative”); AM Gen. Corp. v. DaimlerChrysler Corp., 311 F.3d 796, 827-28 (7th Cir. 2002) (“None of the relevant motor vehicles is the sort of low-priced item to which a consumer might devote little careful thought.”); Rust Envt. & Infrastructure, Inc. v. Teunissen, 131 F.3d 1210, 1217 (7th Cir. 1997) (upholding district court's conclusion that “the consumers’ sophistication ... and high cost of [services] indicate that such confusion is unlikely”); Dorr-Oliver, 94 F.3d at 383 (“It is not possible that any of these companies, all of which own and operate Dorr-Oliver clamshells, will be confused by Fluid-Quip’s introduction of a competitive line of clamshells.”). See also Jet, Inc. v. Sewage Aeration Sys., 165 F.3d 419, 423 (6th Cir. 1999) (“[E]xperts in their fields [are] less likely to be confused than lay people”).
The process for direct customers to make a purchase is similar to the one at issue in Washington National Insurance Co. v. Blue Cross & Blue Shield United of Wisconsin, 727 F.Supp. 472, 473 (N.D. Ill. 1990), which involved a dispute between two insurance companies that offered managed care plans. Although the words in the marks were similar (“AdvantEDGE” versus “The Advantage Program”), the court held that no confusion was likely, in part because of the degree of care used by customers:
Since “managed care” is essentially a group insurance product, the market for such programs is composed largely of well-informed and sophisticated purchasers. A company seeking an insurance plan for its employees, for example, will typically solicit proposals, carefully review and compare the benefits offered by each insurance carrier, and then make a selection with the advice of a brokerage firm or consultant. Such a detailed selection process, which could last several months, presents little opportunity for confusion.
Id. at 475 (internal citations omitted). The court agrees with this analysis and concludes that it applies at least as strongly to the sales process for Epic’s prospective direct customers.
As to participants in the Community Connect program, there is some debate between the parties regarding whether those participants should “count” as Epic’s customers at all because Connect participants generally have a contract with one of Epic’s direct customers rather than with Epic itself. Defendants cite multiple documents from Epic indicating that it does not consider Connect participants to be customers. Dkt. 169, ¶¶ 14 and 61 (Connect participants not identified as customers on Epic’s website and published customer data). In fact, Epic cites no evidence that it makes any money from the Connect participants. However, defendants propose a fact that Epic may receive an economic benefit from Connect participants as a result of greater usage of the EHR system by direct customers. Id., ¶ 58. It is surprising that Epic does not present more evidence on such a key question, but the court will assume for the purpose of defendants’ motion that the indirect benefit that Epic receives is sufficient to place Connect participants in the class of relevant customers who may be confused.
This assumption does not get Epic very far because Connect participants cannot be *897accurately described as an ordinary consumer either. They tend to be small hospital administrators or healthcare providers, so they are still relatively sophisticated, and they are still purchasing an expensive license that should heighten their degree of care. The parties do not say a great deal about the process a Connect participant goes through to purchase a license, but it is undisputed that Connect participants generally share patients with one of Epic’s direct customers and learn about the Connect program through them. This suggests that participation in the Connect program is determined more through relationships than through marketing, so if Connect participants are not exercising a high degree of care (and simply becoming a participant out of convenience or a direct customer’s recommendation), that lack of care is bene-fitting Epic, not harming it.
This factor clearly favors defendants. Perhaps realizing this, Epic says little about the factor, devoting much of this section in its brief to the doctrine of “initial interest confusion,” which can apply even in some circumstances in which customers exercise a high degree of care. The court will discuss that theory in a separate discussion at the end of the opinion.
F. Strength of the mark
“The ‘strength’ of a trademark refers to the mark’s distinctiveness, meaning its propensity to identify the products or services sold as emanating from a particular source.” Sorensen, 792 F.3d at 731 (internal quotations omitted). Marks that are descriptive of the goods or services are not as strong as marks that are arbitrarily selected or consist of made up names. The stronger the mark, the more likely similar marks will cause confusion. Telemed Corp. v. Tel-Med, Inc., 588 F.2d 213, 219 (7th Cir. 1978) (“A mark that is strong because of its fame or its uniqueness is more likely to be associated in the public mind with a greater breadth of products or services than is a mark that is weak because it is very much like similar marks.”). See also Playtex Prod., 390 F.3d at 163 (“When there is widespread recognition of a mark among consumers, there is an increased likelihood that consumers will assume it identifies the previously familiar user, and therefore an increased likelihood of consumer confusion if the new user is in fact not related to the first.”) (internal quotations and alterations omitted).
A mark can be inherently distinctive and therefore strong or an otherwise weaker mark can acquire strength over time because the mark’s owner has taken steps to make the mark well known. Brennan’s, Inc. v. Brennan’s Rest., LLC, 360 F.3d 125, 130-31 (2d Cir. 2004). When a mark has acquired strength, it has developed a “secondary meaning.” Corbond, 356 F.Supp.2d at 918. In this case, Epic has made little showing that its mark is strong, either inherently or through development of secondary meaning.
1. Inherent strength
As to inherent strength, it is undisputed that the terms “care” and “everywhere” are relatively commonplace in the healthcare industry and defendants have cited various instances in which the phrase “care everywhere” is used by others in the healthcare industry. When a party chooses more common words as its trademark, the owner must be prepared to “tolera[te] ... a certain degree of confusion on the part of consumers.” KP Permanent Make-Up, Inc. v. Lasting Impression I, Inc., 543 U.S. 111, 122, 125 S.Ct. 542, 160 L.Ed.2d 440 (2004).
Also, Epic’s and defendants’ marks are relatively descriptive or at most suggestive. Packman, 267 F.3d at 638 (“The law recognizes five categories of trademarks, in ascending order of distinctiveness: generic, descriptive, suggestive, arbitrary, *898and fanciful.”); Sorensen, 792 F.3d at 724-25 (“A .descriptive term ordinarily names a characteristic of a product or service..., [A] suggestive [term] “requires some operation of the imagination.”) (internal quotations omitted). Epic’s mark refers to a product that allows patients and providers to' access healthcare records “everywhere,” that is, across difference providers; defendants’ mark refers to products and services that allow patients to review healthcare information “everywhere” by using defendants’ website or mobile app. The descriptive nature of Epic’s mark makes it less distinctive and thus weaker. TCPIP Holding Co. v. Haar Commc’ns, Inc., 244 F.3d 88, 101 (2d Cir. 2001) (“If the similarity between the two marks would lead consumers to assume the similar marks were chosen because they describe similar, desirable attributes of the products, the similarity is less likely to invite a mistaken assumption that the products must come from the same source.”). In fact, Epic admits that there has been confusion surrounding its own mark in that some people referred to it as “Care Anywhere” and others simply had difficulty recalling the name.
, Epic observes repeatedly in its discussion of the strength of .its mark and throughout the rest of its brief that its mark is “incontestable,” but that status simply restricts the bases on which the registration can be challenged. It does riot control an analysis whether the mark is strong in the context of a likelihood of confusion analysis. Restatement (Third) of Unfair Competition § 21, cmt. i (1995) (“A few cases suggest that an incontestable mark is ‘strong’ for purposes of the likelihood of confusion analysis, or at least that incontestability is a factor to be considered in assessing the strength of a mark. However, the test for likelihood of confusion is based on the perceptions of consumers in the marketplace, which are ordinarily unaffected by the status of a mark’s registration.”) (citing Munters Corp. v. Matsui Am., Inc., 909 F.2d 250 (7th Cir. 1990)). Thus, Epic cannot rely on its mark’s status to escape the conclusion that the mark has little inherent strength.
2. Acquired strength
Acquired strength can be determined in a number of different ways: direct consumer testimony; consumer surveys; the exclusivity, length, and manner of use of the mark; amount and manner of advertising for the mark; the owner’s amount of sales and number of customers; how established the owner’s place in the market is; and intentional copying. Echo Travel, Inc. v. Travel Assocs., Inc., 870 F.2d 1264, 1267 (7th Cir. 1989). See also AutoZone, 643 F.3d at 933 (“The AutoZone mark is displayed prominently on more than 3,000 stores nationwide, and it has been the subject of hundreds of millions of dollars’ worth of advertising since 1987.”).
• Epic has not conducted a survey to determine the strength of its mark. Such evidence can be “particularly persuasive,” Restatement (Third) of Unfair Competition § 13, cmt, e (1995), but it is also rare. Bose Corp. v. QSC Audio Prod., Inc., 293 F.3d 1367, 1371 (Fed. Cir. 2002). Epic says that it has spent approximately $100 million on marketing and promotion since 2014, but that figure is unhelpful because Epic does not identify what portion of that figure relates to CARE EVERYWHERE.
On the other hand, Epic has been using its mark since at least 2004 and it is undisputed that Epic and its CARE EVERYWHERE products and services are “widely known” in the “industry.” Dkt. 170, ¶ 46 and Dkt. 169, ¶ 90. The parties provide little other information on this issue and they do not say specifically whether CARE EVERYWHERE is well known among the class of prospective Connect participants, which, as the court discussed above, are *899the only relevant consumers who could be confused. The only fact related to this issue proposed by the parties is that Connect participants generally hear about Epic through interactions with Epic’s direct customers, Dkt. 169, ¶ 60, which could limit awareness of the mark to those who are most likely to be predisposed to seeking a license from Epic anyway.
Although the record is unclear regarding the strength of Epic’s mark as to the most relevant customers, for the purpose of defendants’ motion, the court will assume that potential Connect participants are aware of the mark and associate it with Epic’s interoperability application.6 Making this assumption, however, does little to buttress Epic’s claim in light of the many other factors favoring defendants.
G. Similarity of products
Epic’s CARE EVERYWHERE is associated with an interoperability application; defendants’ YOURCAREEVERYWHERE is associated primarily with a health and wellness website and mobile app. It is undisputed that these are not competing products and that Epic does not offer any product that competes with the YOURCAREEVERYWHERE products and services. This counsels against a finding of likely confusion. Checkpoint Sys., Inc. v. Check Point Software Techs., Inc., 269 F.3d 270, 282 (3d Cir. 2001) (“Mark similarity is not necessarily determinative of likely confusion, particularly when the products do not directly compete.”).
As Epic points out, -the lack of directly competing products is not fatal to its claims if the products are similar enough that a customer could reasonably think that the products come from the same source. CAE, Inc. v. Clean Air Eng’g, Inc., 267 F.3d 660, 679 (7th Cir. 2001). For example, in AutoZone, 643 F.3d at 931, the plaintiff was a car parts store and the defendant provided oil changes and ear washes. Athough there was no direct competition, the court concluded that the defendant’s services were sufficiently related to the plaintiffs products that a consumer might believe that the plaintiff was “venturing into a related service industry,” particularly because the defendants’ marks were so similar to the plaintiffs. Id, Although a trademark owner may not lose sales immediately because of that type of confusion, it might still harm the owner because of a loss of good will that results from a negative experience the customer has with the other business. (It might also hinder the owner’s ability to expand its business into related areas, CAE, 267 F.3d at 681, but Epic does not suggest that it has any intent to provide any product or service that falls under the YOURCAR-EEVERYWHERE mark.)
Athough the parties debate whether Epic and defendants sell “related” products, they say little about the question raised in AutoZone, which- is whether a relevant customer could reasonably expect Epic to sponsor a general health and wellness website. Because defendants offer both their website and an EHR system, *900perhaps it “would not be that surprising,” id., if Epic did the same thing. However, cases like AutoZone and CAE are distinguishable because the defendants in those cases were calling their businesses by a name that was similar to the plaintiffs name. In this case, there is no suggestion that anyone would think that defendants are masquerading as an affiliate of Epic. Thus, the relevant question is whether a customer would be likely to think that Epic would either (1) use a mark that is related to an interoperability application on products and services related to a health and wellness website or (2) allow a third party to do so. Because Epic fails even to develop an argument that it would be reasonable for a customer to connect those unrelated products with one mark, the court concludes that this factor favors defendants as well.
H. Area and manner of use
The Seventh Circuit has summarized this factor as follows: “[CJourts look at whether there is a relationship in use, promotion, distribution or sales between the goods or services of the parties. We also look to whether the parties use the same channels of commerce, target the same general audience, or use similar marketing procedures.” Sorensen, 792 F.3d at 730 (internal citations and quotations omitted). The key question is whether the plaintiff and defendant are competing for the same customers. Kraft Foods Grp. Brands LLC v. Cracker Barrel Old Country Store, Inc., 735 F.3d 735, 739 (7th Cir. 2013); Packman, 267 F.3d at 646; Smith Fiberglass Prod., Inc. v. Ameron, Inc., 7 F.3d 1327, 1330 (7th Cir. 1993). “[L]ikelihood of confusion [is] diminished where there [is] minimal, if any, overlap in customers.” Packman, 267 F.3d at 646.
As noted above, Epic has no products or services that compete directly with defendants’ YOURCAREEVERYWHERE products and services. However, both Epic and MEDHOST sell an EHR system. (Both Epic and defendants have a patient portal as well, but Epic’s portal is part of its EHR system and Epic does not does not develop a separate argument about the portal.) Defendants argue that their EHR system does not compete with Epic’s because MEDHOST’s EHR system is significantly cheaper than EpicCare and-'directed at smaller healthcare organizations. United States v. Torkington, 812 F.2d 1347, 1353 (11th Cir. 1987) (“[P]rice differential may be relevant to the determination of whether the mark is likely to confuse.”); Farberware, Inc. v. Mr. Coffee, Inc., 740 F.Supp. 291, 302 (D. Del. 1990) (“Although the fact that Mr. Coffee’s product is meant to appeal to modest consumers, while Far-berware’s is directed to the more upscale, does not negate that these products compete, it does lead the Court to consider another relevant factor, namely the price differential.”).
Epic denies that the parties target different customers, but the evidence it cites is limited. Epic has identified only two instances in which the companies competed for the same customer. This does not mean that competition will not increase in the future, but it limits the relevance of this factor considerably.
To sum up, the following factors diminish any likelihood of confusion in this case:
• no customers (or anyone else) have expressed confusion about defendants’ mark;
• Epic has not presented any survey evidence suggesting that confusion is likely;
• strong evidence supports a conclusion that defendants chose their mark as part of a group of ‘Your-Care” marks and did not intend to deceive customers;
• when prospective customers are presented with defendants’ YOURCAR-*901EEVERYWHERE mark, it is generally in the context of defendants’ other marks, particularly YOURCA-REUNIVERSE;
• Epic’s marked products and services have no logical relationship to defendants’ marked products and services;
• the parties have little overlap in the customers they serve;
• both marks are relatively descriptive and use commonplace words;
• many potential customers are sophisticated and deliberative;
• less sophisticated customers who are potential Connect participants are more likely to choose a product because of a relationship with an existing customer rather than a marketing pitch.
In light of these factors, which tip strongly in defendants’ favor, the similarity in the words used in the marks and the fact that Epic and its mark may have strong name recognition among some customers are not sufficient to show a likelihood of confusion in this case. Even if a customer who is familiar with Epic’s mark sees defendants’ mark, it should be clear from the context that Epic is neither the source nor the sponsor of defendants’ mark.
I. Initial interest confusion
Epic attempts to save its claims by relying on the doctrine of “initial interest confusion,” which occurs when the defendant “lur[es] potential customers away from [the plaintiff] by initially passing off its goods [or services] as those of the [plaintiffs] even if confusion as to the source of the goods is dispelled by the time any sales are consummated.” Dorr-Oliver, 94 F.3d at 382. The Seventh Circuit has summarized initial interest confusion as a “bait and switch” technique, in which a competitor will try to “get its foot in the door” and affect a purchasing decision by confusing the consumer. Syndicate Sales, Inc. v. Hampshire Paper Corp., 192 F.3d 633, 638 (7th Cir. 1999).
In Promatek Industries, Ltd. v. Equitrac Corp., 300 F.3d 808, 812 (7th Cir. 2002), the court gave the example of someone “posting a sign with another’s trademark in front of one’s store.” Even if the customer realizes after she enters that the store is not actually associated with the trademark owner, by then it may be too late. “Customers believing they are entering the first store rather than the second are.still likely to mill around before they leave,” id. at 812-13, and potentially make a purchase that should have gone to the trademark owner. The facts in Promatek involved the defendant’s use of the plaintiffs mark as a metatag, so that a consumer conducting an internet search for the plaintiffs site might find the defendant’s site instead. The court concluded that a consumer who clicked on a link for the wrong site might soon realize his error, but still would be exposed to the defendant’s products and give the defendant an opportunity to steal sales.7
The facts in this case bear no similarity to those in which a court accepted a theory of initial interest confusion as plausible. To begin with, as the court’s reference to a “bait and switch” suggests, cases involving initial interest confusion often include some indication that the defendant is trying to deceive customers who are looking *902for products or services from the plaintiff. E.g., Checkpoint Sys., 269 F.3d at 296 (“[Cjourts look at evidence that the defendant intentionally adopted the plaintiffs mark to create confusion among consumers making, purchasing decisions.”) (citing Dorr-Oliver, 94 F.3d at 382-83); Wolf Appliance v. Viking Range Corp., 686 F.Supp.2d 878, 891-92 (W.D. Wis. 2010) (reasonable jury could find that defendant copied plaintiffs trade dress' “because it wanted customers to' believe that plaintiff and' defendant are now affiliated”). As noted above, defendants have provided compelling evidence that their adoption of the YOURCAREEVERYWHERE mark had nothing to do with Epic and Epic has presented no evidence that defendants have attempted to use their mark to lure any customers away from Epic.
Courts are most likely to apply the doctrine of initial interest confusion doctrine in circumstances involving directly competing products, particularly when the potential purchasers are lay consumers making decisions in -a relatively short amount of time with limited information. Compare Promatek, 300 F.3d at 812 (initial interest confusion applied when parties were direct competitors and customers exercised low degree of care); Brookfield Commc’ns, 174 F.3d at 1057 (finding initial, interest confusion actionable because “in the Internet context, ... entering a web site takes little effort—usually one click from a linked site or a search engine’s list; thus, Web surfers are more likely to be confused as to the ownership of a web site than traditional patrons of a brick-and-mortar store would be of a store’s ownership.”), and Checkpoint Sys., 269 F.3d at 296-97 (“When products are similar, a firm is more likely to benefit from the goodwill of a firm with an established mark. And when consumers do not exercise á high level of care' in making their decisions, it is more likely that their initial confusion will result in a benefit to the alleged infringer from the use of the goodwill of the other firm.”), with Sensient Techs. Corp. v. SensoryEffects Flavor Co., 613 F.3d 754, 766 (8th Cir. 2010) (declining to apply doctrine of initial interest confusion; “although the products are similar, the parties agree the customers are sophisticated and exercise a relatively high degree of care, in making their purchasing decisions”), and AM Gen. Corp., 311 F.3d at 827-28 (“Any similarity in the grilles [of two competing car companies] does not present a risk of a ‘bait and switch’ effect” because “[n]oné of the relevant motor vehicles is the sort of low-priced item to which a consumer' might devote little careful thought.”), and Rust Envt. & Infrastructure, 131 F.3d at 1217 (no initial interest confusion in case in which “the relevant market of consumers is composed of sophisticated buyers of services in a highly specialized technical field”).
As discussed throughout this opinion, the purchasing decisions at issue in this case involve sophisticated consumers making expensive purchases often over a long period of time after acquiring much information. All of these factors counsel against the possibility that initial interest confusion is at all likely. ,
As Epic points out, the doctrine can apply to more expensive items and more sophisticated customers, e.g., Wolf Appliance, 686 F.Supp.2d at 891-92 (consumers buying expensive stoves), and the leading treatise cites one case before the Trademark Trial and Appeal Board in which the board applied the doctrine to an “extended purchasing process” involving expensive computer software. 4 McCarthy on Trademarks and Unfair Competition § 23:6 (4th ed. 2008) (citing Hrl Assocs. Inc. v. Weiss Assocs. Inc., 12 U.S.P.Q.2d 1819 (T.T.A.B. July 21, 1989)). However, in those cases, the confusion lasted long enough to give the defendant some com*903petitive advantage over the trademark owner, suggesting that there must be some basis from which to infer that the defendant has -influenced a customer’s decision to make a purchase. Wolf Appliance, 686 F.Supp.2d at 891-92 (consumer influenced to consider purchasing stove because it featured competitor’s red knobs); Hrl Assocs., 12 U.S.P.Q.2d 1819 (initial interest confusion applies if the trademark owner “may be precluded from further consideration by the'potential purchaser in reaching his or her buying decision (which may, in turn, prevent [the trademark owner] from making a sale)”). See also Gibson Guitar Corp. v. Paul Reed Smith Guitars, LP, 423 F.3d 539, 552 (6th Cir. 2005) (initial interest confusion does not apply if products appear similar from distance but are distinct upon closer inspection); Checkpoint Sys., 269 F.3d at 295-96 (“[I]f the initial interest confusion does not ultimately result in a purchasing decision, this factor counsels against finding the likelihood of confusion.”).
In this case, Epic has not articulated any plausible scenario under which a potential customer could be confused by defendants’ mark long enough to have any influence on a purchasing decision. In its brief, Epic offers the following hypothetical:
[A] hospital administrator that has received medical records through Epic’s CARE EVERYWHERE interoperability software (or is Googling to And other offerings related to CARE EVERYWHERE in order to expand) that is then pitched by.MEDHOST, using the YOURCAREEVERY-WHERE suite as an entrée for EHR or other products/services by MED-HOST, could mistakenly believe the two. offerings are associated and, therefore, will work as seamlessly as her interactions with Epic’s software. Thus, MEDHOST would be able to get in the door unfairly by trading on the goodwill associated with Epic’s mark.
Dkt. 146, at 23-24.
Epic has presented no evidence that internet searches will produce misleading results, so its reference to “Googling” is a nonstarter. But if one removes that reference from the hypothetical, all that remains is a vague and conclusory allegation, devoid of any context. Epic does not identify any particular scenario, likely or otherwise, in which a potential customer could be confused into believing that Epic is the source or sponsor of defendants’ YOÚR-CAREEVERYWHERE products and services. . . •
What would the “pitch” identified in Epic’s hypothetical look like? Under what circumstances could defendants make this pitch in a way that the administrator would be unable to discern from the context of the pitch that defendants and not Epic are the source and sponsor of defendants’ mark? Epic’s hypothetical rests on an assumption that the administrator is in a factual vacuum in which she knows nothing except that CARE EVERYWHERE is Epic’s mark and defendants are trying to use a similar mark to sell her a different product. But that assumption is inconsistent with the undisputed facts that defendants market YOURCAR-EEVERYWHERE along with other Your-CareUniverse products and services and that it is generally clear from the context of the ' YOURCAREEVERYWHERE mark that defendants are the source. 1-800 Contacts, Inc. v. Lens.com, Inc., 722 F.3d 1229, 1245 (10th Cir. 2013) (“Perhaps in the abstract, one who searches for a particular business with a strong mark and sees an entry on the results page will naturally. infer that the entry is for that business. But that inference is an unnatural one when the entry is clearly labeled as an advertisement and clearly identifies *904the source, which has a name quite different from the business being searched for.”); Network Automation, Inc. v. Advanced Sys. Concepts, Inc., 638 F.3d 1137, 1154 (9th Cir. 2011) (“[Cjlear labeling might eliminate the likelihood of initial interest confusion.”) (internal quotations omitted); Rust Envt. & Infrastructure, 131 F.3d at 1217 (upholding district court’s conclusion-that initial interest confusion unlikely when defendant’s “employees take precautions by introducing and advertising themselves as part of a new firm”).
“[Ijnitial interest confusion is not assumed and must be proven by the evidence.” Sensient Techs., 613 F.3d at 766 (quoting 4 McCarthy on Trademarks and Unfair Competition § 23:6 (4th ed. 2008)). Particularly when one considers all the factors that would diminish the likelihood of initial interest confusion in this case, Epic’s failure to clearly articulate any likely scenario under which a customer would be confused is fatal to its claims. Libman, 69 F.3d at 1363 (“[A] finding of likely confusion can no more be based on pure conjecture or a fetching narrative alone than any other finding on an issue on which the proponent bears the burden of proof.”).
CONCLUSION
In sum, the court agrees with defendants that Epic’s claims are fundamentally speculative. Although defendants’ mark includes the same words as Epic’s mark, under the circumstances of this case, that is not enough. Perhaps confusion is “not impossible,” Dkt. 96 (Faulkner Dep., at 27-71), but “no reasonable [factfinder], looking at the seven factors as a whole, could conclude that there is a likelihood of confusion.” Sorensen, 792 F.3d at 726.
ORDER
IT IS ORDERED that:
1. The motion for summary judgment filed by defendants YourCareUniv- ' erse, Inc., MEDHOST of Tennessee, Inc., and MEDHOST Direct, Inc., Dkt. 113, is GRANTED.
2. The following motions are DENIED as moot: defendants’ motion to exclude the testimony of Michael Cohen, Dkt. 118; defendants’ motion to strike Epic’s jury demand, Dkt. 120; Epic’s motion to amend its damages report, Dkt. 194; defendants’ motion for leave to amend their answer to assert the affirmative defense of abandonment, Dkt. 202; defendants’ motion to compel, Dkt. 224; Epic’s motion for an extension of time, Dkt. 226; and Epic’s motion to compel, Dkt. 228.
3. The clerk of court is directed to enter judgment in favor of defendants and close this case.
. E.g., Dkt. 170, ¶ 23, (disputing fact about programs that CARE EVERYWHERE application is used "in connection with" on ground that CARE EVERYWHERE mark is not used on those programs); Dkt. 169, ¶ 82 (not disputing fact but adding more information); id., ¶ 146, (disputing fact about products that include the YOURCAREEVERY-WHERE mark with information regarding how defendants use mark for marketing); id., ¶ 152 (disputing fact about registering an account on defendants’ website with information about how defendants "promoted and sell[ ]” their products and services); id., ¶ 162 (disputing fact about how products function with information about how products are marketed); id., ¶ 195 (disputing fact about Epic’s direct customers with information about customers who license Epic’s products from direct customers).
. E.g., Dkt. 170, ¶ 18 (“All Epic’s applications and modules are important for Epic’s customers to meet their patient care and patient engagement objectives.”); id., ¶ 28 ("Epic ... facilitates a network of providers participating in a health information exchange.”); id., ¶ 69 ("MEDHOST addressed the rising consumerism in the marketplace through marketing and sale of the YourCareUniverse products and services.”); id., ¶ 79 ("YOURCAREEVERYWHERE provides ‘front door’ access to the MEDHOST provider’s patient *882portal.”); id., ¶ 91 ("Epic .., has developed substantial good will in its CARE EVERYWHERE mark.”).
. The parties use terms inconsistently to describe the software at issue in this case, For example, they seem to use "platform” and "application” interchangeably. E.g., Dkt. 170, ¶ 20, ("CARE EVERYWHERE is Epic’s interoperability application ...”); id., ¶ 23 ("CARE EVERYWHERE's interoperability platform also supports ...”).
. In its reply materials, defendants identified additional instances of Epic’s customers allegedly using CARE EVERYWHERE incorrectly, Dkt. 169, ¶ 101, but the court has not included these because Epic has not had any opportunity to respond,
. Epic states generally in its proposed findings of fact that it has competed with MED-HOST on other occasions,-Dkt, 170, ¶ 86, but it does not identify any other examples. Instead, it cites generally to an exhibit that includes a list of dates and names of various companies, along with more than 20 pages of typed notes that are mostly sentence fragments. Dkt. 141-1. This citation with no cor*889responding explanation is insufficient to establish the proposed fact, particularly because the exhibit is far from self-explanatory.
. As evidence of the weakness of Epic’s mark, defendants cite several examples in which Epic’s customers used CARE EVERYWHERE incorrectly, either not attributing the mark to Epic or attributing the mark to the wrong products and services. However, the cases defendants cite involve an owner’s failure to use a mark consistently, not a third party’s failure to comply with a license, Sorensen, 792 F.3d at 731; Gimix, Inc. v. JS & A Grp., Inc., 699 F.2d 901, 907 (7th Cir. 1983). Defendants cite no authority for the view that an owner’s lax enforcement of the mark renders it weaker. Perhaps a customer’s failure to attribute the mark to the correct product could be circumstantial evidence that the customer is unfamiliar with the mark, but this is weak evidence at best, particularly because of the small sample size defendants provided.
. Using another party’s trademark to misdirect an internet search appears to be one of the more common situations in which the initial interest confusion doctrine has been applied E.g., Playboy Enters., Inc. v. Netscape Commc'ns Corp., 354 F.3d 1020, 1024-26 (9th Cir. 2004); Eli Lilly & Co. v. Nat. Answers, Inc., 233 F.3d 456, 464 (7th Cir. 2000); Brookfield Commc’ns, Inc. v. W. Coast Entm’t Corp., 174 F.3d 1036 (9th Cir. 1999).