This copyright suit shows the difficulty in finding protected creative expression in a crowded field, in this case, architectural design of single-family homes. The case also shows the challenge in administering intellectual property law to discourage so-called intellectual property “trolls” while protecting genuine creativity.
Plaintiffs Design Basics, LLC; Prime Designs, Inc.; and Plan Pro, Inc. (collectively, “Design Basics”) and their affiliates claim rights to some 2700 home designs. They sued defendants Lexington Homes, Inc. and related parties (collectively, “Lexington”) for copyright infringement, contending that Lexington built homes that infringed four of Design Basics’ designs.
The district court granted summary judgment to Lexington, finding no evidence that Lexington ever had access to Design Basics’ home plans. Without access, the court reasoned, there could be no copying and no copyright infringement. We affirm. We agree with the district court that Design Basics has no evidence of access. We also conclude that no reasonable jury could find that Lexington’s accused plans bear substantial similarities to any original material in Design Basics’ plans.
I. Factual and Procedural Background
A. Design Basics and the Art of the Intellectual Property Shakedown
Design Basics has been in the business of producing market-ready designs for modest single-family homes for several decades. In 2009, Patrick Carmichael and Myles Sherman purchased Design Basics as an investment opportunity. Carmichael acknowledged in his deposition that “potential copyright infringement cases influence[d his] decision to become an owner of *1097Design Basics.” He testified that proceeds from litigation have become a principal revenue stream for Design Basics.
A search of the Public Access to Court Electronic Records (PACER) system reveals that Design Basics has been party to over 100 federal lawsuits, the vast majority of which have been filed since the 2009 change in ownership. Nearly all involve copyright claims asserted by Design Basics. Design Basics offers its employees incentives to scout out potential copyright infringement cases, paying its employees a finder’s fee in the form of a percentage of the net recovery relating to any home plans that they located. Design Basics filed this lawsuit after employee Carl Cuozzo discovered Lexington’s website and its supposedly infringing plans while investigating other Design Basics cases on the Internet.
Design Basics’ business model of trawling the Internet for intellectual property treasures is not unique. In recent years, opportunistic holders of copyrights, patents, and other intellectual property have developed unsavory reputations for “trolling,” bringing strategic infringement claims of dubious merit in the hope of arranging prompt settlements with defendants who would prefer to pay modest or nuisance settlements rather than be tied up in expensive litigation. Like the proverbial troll under the bridge, these firms try to extract rents from market participants who must choose between the cost of settlement and the costs and risks of litigation.
This business strategy is far removed from the goals of the Constitution’s intellectual property clause to “promote the Progress of Science and useful Arts.” U.S. Const, art. I, § 8, cl. 8. Intellectual property trolls have attracted much attention and derision in scholarly literature and the mainstream media. See, e.g., Matthew Sag, Copyright Trolling, An Empirical Study, 100 Iowa L. Rev. 1105, 1108 (2015) (“The essence of trolling is that the plaintiff is more focused on the business of litigation than on selling a product or service or licensing their IP to third parties to sell a product or a service. The paradigmatic troll plays a numbers game in which it targets hundreds or thousands of defendants, seeking quick settlements priced just low enough that it is less expensive for the defendant to pay the troll rather than defend the claim.”); see also David Segal, Has Patent, Will Sue: An Alert to Corporate America, N.Y. Times (July 13, 2013), http://www.pytimes.com/2013/07/14/ business/has-patent-will-sue-an-alert-to-corporate-america.html; HI: When Patents Attack!, This American Life (July 22, 2011), https://www.thisamericanlife.org/ radio-archives/episode/441/when-patents-attack.
Courts are aware of these developments. See eBay Inc. v. MercExchange, LLC, 547 U.S. 388, 396, 126 S.Ct. 1837, 164 L.Ed.2d 641 (2006) (Kennedy, J., concurring) (“An industry has developed in which firms use patents not as a basis for producing and selling goods but, instead, primarily for obtaining licensing fees.”); Malibu Media, LLC v. Doe, No. 15 Civ. 4369(AKH), 2015 WL 4092417, at *2 (S.D.N.Y. July 6, 2015) (“Recent empirical studies show that the field of copyright litigation is increasingly being overtaken by ‘copyright trolls’.... ”); cf. Renaissance Learning, Inc. v. Doe No. 1, No. 11-cv-166-slc, 2011 WL 5983299, at *4 (W.D. Wis. Nov. 29, 2011) (“Remember Mark Antony’s funeral oration in Julius Caesarl That’s how an experienced business executive or lawyer would view [a patent holder’s] assertions that ‘we are focused on addressing ... issues without the need for costly and protracted litigation’ and ‘our client’s preferred approach is to conclude licensing discussions without *1098resorting to litigation. We hope you share this objective.’ The implied ‘or else!’ oozes from this letter like lye from lutefisk.”).1
B. The Claims Against Lexington
According to its federal complaint, Design Basics first learned of a possible copyright infringement claim against Lexington on September 11, 2011. Design Basics waited until September 10, 2014—one day before the three-year limitations period would certainly have expired—to bring suit. See 17 U.S.C. § 507(b); Caiman v. McFarlane, 360 F.3d 644, 653 (7th Cir. 2004) (copyright statute of limitations starts to run “when the plaintiff learns, or should as a reasonable person have learned, that the defendant was violating his rights”).
Design Basics filed a complaint alleging that four Lexington home plans—the Carlisle, Oakridge, Ashwood, and Ea-ston—infringed four Design Basics plans: the Aspen, Kendrick, Taylor, and Wom-ack, respectively. Lexington chose not to settle. After discovery, Lexington moved for summary judgment. In response, Design Basics offered little evidence to try to show that Lexington’s agents and employees had ever seen its plans, let alone copied them. Instead, Design Basics offered a declaration by its draftsman, Carl Cuozzo. He opined in conclusory terms that Lexington’s four accused plans are “substantially similar” to the four Design Basics plans “in too many ways to have been the product of independent creation.” Cuozzo added that Design Basics’ plans have been “widely disseminated” via the firm’s website, www.designbasics. com, since 1996, though he admitted in his deposition that he had no idea when the particular plans at issue in this case were first uploaded to the website.
Lexington, by contrast, offered a detailed report by Dr. Robert Greenstreet, Dean of the School of Architecture and Urban Planning at the University of Wisconsin-Milwaukee. Dr. Greenstreet compared the accused plans with Design Basics’ plans, identifying dozens of differences between the paired designs and ultimately finding “no evidence ... of substantial similarity between the designs claimed by Design Basics and those created by Lexington.” Lexington also relied on its shareholders’ and employees’ unre-butted declarations that they had never seen the plans at issue prior to this litigation.
In granting summary judgment for Lexington, the district court focused on the question of access. The court observed that there was no evidence Lexington or its agents received or reviewed any of the plans at issue. See Design Basics, LLC v. Lexington Homes, Inc., No. 14-CV-1102, 2016 WL 8116897, at *2 (E.D. Wis. Sept. 30, 2016). Also, the accused plans and Design Basics’ plans were not so strikingly similar as to permit an inference of copying without separate proof of access. Id. at *4. There was evidence that Lexington’s employees were generally familiar with Design Basics and may have seen some of its other designs. But the court found the evidence of access to other designs was not sufficient to infer access to the designs claimed in this case. Id. at *5. The court entered judgment for Lexington. Design Basics appeals.
*1099II. Analysis
We review de novo the district court’s grant of summary judgment, viewing the evidence in the light most favorable to Design Basics. See Arlington Specialties, Inc. v. Urban Aid, Inc., 847 F.3d 415, 418 (7th Cir. 2017). A non-moving party is entitled to all reasonable inferences in its favor, but “inferences that are supported by only speculation or conjecture will not defeat a summary judgment motion.” Herzog v. Graphic Packaging Int'l Inc., 742 F.3d 802, 806 (7th Cir. 2014), quoting Tubergen v. St. Vincent Hospital & Health Care Center, Inc., 517 F.3d 470, 473 (7th Cir. 2008).
A. Copyright Protection and Infringement
The Copyright Act of 1976, as amended by the Architectural Works Copyright Protection Act, extends copyright protection to “architectural works,” defined as building designs “embodied in any tangible medium of expression, including a building, architectural plans, or drawings.” 17 U.S.C. § 101. A protected architectural work “includes the overall form as well as the arrangement and composition of spaces and elements in the design, but does not include individual standard features.” Id.; see also 37 C.F.R. § 202.11(d) (“The following structures, features, or works cannot be registered: ... (2) Standard features. Standard configurations of spaces, and individual standard features, such as windows, doors, and other staple building components.”).2
To establish copyright infringement, a plaintiff must prove two elements: “(1) ownership of a valid copyright, and (2) copying of constituent elements of the work that are original.” JCW Investments, Inc. v. Novelty, Inc., 482 F.3d 910, 914 (7th Cir. 2007), quoting Feist Publications, Inc. v. Rural Telephone Serv. Co., 499 U.S. 340, 361, 111 S.Ct. 1282, 113 L.Ed.2d 358 (1991). Design Basics’ plans are registered with the United States Copyright Office. We assume without deciding that Design Basics owns the plans and that the plans are entitled to some copyright protection, i.e., they were created independently and possess a modicum of creativity. See Feist Publications, 499 U.S. at 346, 111 S.Ct. 1282. We focus on the second element of Design Basics’ infringement claim: proof of copying.
The Copyright Act affords no protection against the independent creation of a work that happens to resemble some prior creation. A plaintiff must prove that the defendant actually copied its original work. See Selle v. Gibb, 741 F.2d 896, 901 (7th Cir. 1984) (“Proof of copying is crucial to any claim of copyright infringement because no matter how similar the two works may be (even to the point of •identity), if the defendant did not copy the accused work, there is no infringement.”). Direct evidence of copying is only rarely available. A plaintiff may also “prove copying by showing that the defendant had the opportunity to copy the original (often called ‘access’) and that the two works are ‘substantially similar,’ thus permitting an inference that the defendant actually did copy the original.” Peters v. West, 692 F.3d 629, 633 (7th Cir. 2012).
The access requirement is not onerous. For example, a plaintiff may sat*1100isfy the requirement by showing that the copyrighted work was “sent directly to the defendant ... or a close associate of the defendant.” Selle, 741 F.2d at 901; see also, e.g., Peters, 692 F.3d at 634 (defendant closely collaborated with music producer who had copy of plaintiffs recording); JCW Investments, 482 F.3d at 913, 915 (officer of corporate defendant admitted that he saw and perhaps photographed plaintiffs doll and that doll gave him idea for accused work). A plaintiff may also satisfy the requirement by showing that the copyrighted work was so widely disseminated that the defendant can be presumed to have seen or heard it. E.g., Cholvin v. B. & F. Music Co., 253 F.2d 102, 103-04 (7th Cir. 1958) (access shown where plaintiffs distributed 2000 copies of sheet music and sold more than 200,000 records, and song was broadcast nationwide for years).
“Once a plaintiff establishes that a defendant could have copied her work, she must separately prove ... that the allegedly infringing work is indeed a copy of her original,” by “pointing to similarities between the two works.” Peters, 692 F.3d at 635. In other words, access cannot serve as a proxy for substantial similarity. “[W]e have never endorsed ... the idea that a ‘high degree of access’ justifies a ‘lower standard of proof for similarity.” Id. (citation omitted). In unusual cases, however, evidence of truly striking similarity may function as a proxy for access. As we observed in Selle, an inference of access may arise from “proof of similarity which is so striking that the possibilities of independent creation, coincidence and prior common source are, as a practical matter, precluded.” 741 F.2d at 901; see also Ty, Inc. v. GMA Accessories, Inc., 132 F.3d 1167, 1171 (7th Cir. 1997) (“Access (and copying) may be inferred when two works are so similar to each other and not to anything in the public domain that it is likely that the creator of the second work copied the first, but the inference can be rebutted by disproving' access or otherwise showing independent creation....”). In Selle and Ty, Inc., we took care to point out that a prior common source for both plaintiffs and defendant’s work could undermine an inference of copying. The possibility of a prior common source is especially important in crowded fields such as popular music in Selle and designs for single-family homes in this case.3
So in the rare case involving an uncanny resemblance for which copying the plaintiffs work is the only plausible explanation, further proof of access may not be required. But in general, to prove a circumstantial case of copyright infringement, the plaintiff must separately prove both access and similarity. In this case, Design Basics’ proof falls short in both ways.
B. Design Basics’ Infringement Claims
1. Similarity
The market for affordable home designs is crowded because opportunities for origi*1101nality are tightly constrained by functional requirements, consumer demands, and the vast body of similar designs already available. In this field, the substantial similarity requirement is particularly hard to satisfy. We agree with Zalewski v. Cicero Builder Dev., Inc., 754 F.3d 95 (2d Cir. 2014), which described the designer’s or architect’s copyright as thin, so that only very close copying of protected elements is actionable.
Like most circuits, we gauge substantial similarity through a framework described by Judge Learned Hand in Peter Pan Fabrics, Inc. v. Martin Weiner Corp., 274 F.2d 487 (2d Cir. 1960). We consider “whether the accused work is so similar to the plaintiffs work that an ordinary reasonable person would conclude that the defendant unlawfully appropriated the plaintiffs protectible expression by taking material of substance and value.” Wildlife Express Corp. v. Carol Wright Sales, Inc., 18 F.3d 502, 509 (7th Cir. 1994) (citation omitted); see also Peter Pan Fabrics, 274 F.2d at 489 (two works are substantially similar where an “ordinary observer, unless he set out to detect the disparities, would be disposed to overlook them, and regard their aesthetic appeal as the same”). “The test for substantial similarity is an objective one.” JCW Investments, 482 F.3d at 916.
When considering substantial similarity, it is essential to focus on pro-tectable expression. Similarity by itself— even close similarity—should not automatically spark an inference of unlawful copying. Countless ideas, themes, images, and even home designs crowd the public domain. They are available to all: no one can claim a copyright in an expression in the public domain. As we explained in Ty, Inc., “two works may be strikingly similar— may in fact be identical—not because one is copied from the other but because both are copies of the same thing in the public domain.... A similarity may be striking without being suspicious.” 132 F.3d at 1170; see also Peters, 692 F.3d at 633-34 (“Fundamentally, proving the basic tort of infringement ... requires the plaintiff to show that the defendant had an actual opportunity to copy the original ... and that the two works share enough unique features to give rise to a breach of the duty not to copy another’s work.”) (emphasis added); Incredible Technologies, Inc. v. Virtual Technologies, Inc., 400 F.3d 1007, 1011 (7th Cir. 2005) (“The test for substantial similarity may itself be expressed in two parts: whether the defendant copied from the plaintiffs work and whether the ‘copying, if proven, went so far as to constitute an improper appropriation.’ ”) (citation omitted).
The challenge of proving substantial similarity is heightened where the field is crowded or where aesthetic choices may be secondary to consumer demands or functional requirements. The challenge is particularly acute in the market for affordable designs for single-family homes, where form follows function so closely. Blueprints drafted by companies like Design Basics and Lexington are meant to appeal to home builders and buyers who want to build, sell, and buy homes that will hold their value. While it is possible to design a home that is a one-of-a-kind work of art, the home designs here do not fit that description. They comprise familiar configurations of spaces and features, having less in common with the work of Frank Gehry or Frank Lloyd Wright and more with that of William Levitt.
We have not previously considered the intersection between our cases on substantial similarity and copyright claims involving home designs. Cf. Nova Design Build, Inc. v. Grace Hotels, LLC, 652 F.3d 814, 818 (7th Cir. 2011) (architect’s hotel de*1102sign, based largely on Holiday Inn Express prototype, was devoid of originality and not entitled to copyright protection). We find helpful guidance from other courts.
In Zalewski, the plaintiff was a designer of colonial-style homes. The Second Circuit acknowledged that the plaintiffs copyright was valid and that substantial evidence showed that the defendants copied his designs. 754 F.3d at 101. The court concluded, though, that the defendants “took only the unprotected elements” of the plaintiffs work and thus did not infringe the plaintiffs copyright. Id. at 102.
The Zalewski court relied on two related and important doctrines that help distinguish protected expression from material in the public domain. The doctrine of scenes a, faire “teaches that elements of a work that are ‘indispensable, or at least standard, in the treatment of a given topic’ ... get no protection.” Id. at 102 (citation omitted); see also Bucklew v. Hawkins, Ash, Baptie & Co., 329 F.3d 923, 929 (7th Cir. 2003) (“[A] copyright owner can’t prove infringement by pointing to features of his work that are ... so rudimentary, commonplace, standard, or unavoidable that they do not serve to distinguish one work within a class of works from another.”). The related doctrine of merger “instructs that some ideas can only be expressed in a limited number of ways” such that “idea and expression ‘merge,’ ” with the resulting amalgam receiving no copyright protection. Zalewski, 754 F.3d at 102-03; see also Bucklew, 329 F.3d at 928 (merger “refers to the situation in which there is only one feasible way of expressing an idea, so that if the expression were copyrightable it would mean that the idea was copyrightable, and ideas are not copyrightable”).
Zalewski recognized that there are scenes a, faire in architecture just as in literature or music. Colonial-style homes— or, as relevant here, a wider variety of suburban homes—share many design elements and characteristics to which no individual designer can lay claim. Design Basics’ designs also share many attributes driven by “consumer expectations and standard house design generally,” and designers can “get no credit for putting a closet in every bedroom, a fireplace in the middle of an exterior wall, [or] kitchen counters against the kitchen walls.” Zalewski, 754 F.3d at 106. These features may appear similar across plans and models, but “the term ‘substantial similarity’ is properly reserved for similarity that exists between the protected elements of a work and another work.” Id. at 101 (emphasis added). When an architect hews closely to existing convention, the architect’s “original contribution [is] slight—his copyright very thin,” so that only “very close copying” could take whatever truly belongs to the architect. Id. at 107.4
Design Basics’ infringement claim fails for essentially the same reason that the plaintiffs claim failed in Zalewski. Lexington’s accused plans resemble Design Basics’ plans, but only because both sets resemble common home designs one might observe throughout the suburbs of *1103Milwaukee, Chicago, Indianapolis, or many other communities. There are only so many ways to arrange a few bedrooms, a kitchen, some common areas, and an attached garage, so “not every nook and cranny of an architectural floor plan enjoys copyright protection.” See Home Design Services, Inc. v. Turner Heritage Homes Inc., 825 F.3d 1314, 1321 (11th Cir. 2016).
Design Basics advertises over 2700 home plans at www.designbasics.com. We wonder whether there is any blueprint for a single-family home anywhere in the country that Design Basics could not match to one of its own designs by applying the loose standard of similarity it relies upon here. An infringement doctrine that would enable an aggressive designer to sue all, of its competitors and bring those claims before a jury would turn the law of copyright on its head. Far from promoting the “Progress of Science and useful Arts,” U.S. Const, art. I, § 8, cl. 8, such a doctrine would chill the market.
As noted, a close study of Design Basics’ and Lexington’s plans reveals many aesthetic distinctions, further undercutting Design Basics’ claim that Lexington took anything that belonged to it. Dr. Greenstreet, Lexington’s expert, identified 46 differences between Design Basics’ Aspen and Lexington’s Carlisle; 54 differences between Design Basics’ Kendrick and Lexington’s Oakridge; 49 differences between Design Basics’ Taylor and Lexington’s Ashwood; and 62 differences between Design Basics’ Womack and Lexington’s Easton. Dr. Greenstreet spotted differences in dimensions and spatial relationships, in roofing configurations and building materials, and in carpentry and decor.5
Given the economic and functional constraints on the designers and the vast body of similar designs in the public domain, these many differences are not trivial. See Home Design Servs., 825 F.3d at 1324 (“[W]hen floor plans are drawn in a customary style and to industry standards, even ‘subtle differences’ ... can indicate that there is no copyright infringement.”).
In response to Dr. Greenstreet’s detailed report, Design Basics offered essentially nothing.6 On summary judgment and *1104on appeal, Design Basics relied on a con-clusory six-page declaration by its draftsman, Carl Cuozzo, that does not reflect any effort at side-by-side comparison of the plans. Cuozzo said only generally that the accused plans are “substantially similar to Plaintiffs’ Copyrighted Works in too many ways to have been the product of independent creation” and that it was “obvious” to him that the accused plans were copies of Design Basics’ plans.
Cuozzo has offered remarkably similar opinions in other Design Basics cases. E.g., Design Basics LLC v. J & V Roberts Investments, Inc., 130 F.Supp.3d 1266, 1279 (E.D. Wis. 2015) (“The plaintiffs ... submit evidence from an employee, Carl Cuozzo, opining that the works at issue are substantially similar in too many ways to have been the product of independent creation.”) (citation omitted); Design Basics, LLC v. DeShano Cos., No. 10-14419, 2012 WL 4321313, at *15 (E.D. Mich. Sept. 21, 2012) (“Carl Cuozzo ... testified that each of [defendant’s] plans were ‘substantially similar to the Design Basics plan.’ Further, Mr. Cuozzo testified that ‘[t]he overlays show enough similarities ... that it’s unimaginable that it’s a coincidence.’ ”) (citations omitted). Design Basics’ repeated reliance on its in-house draftsman’s conclusory testimony in lieu of third-party expert testimony, where the draftsman stands to profit directly if his firm prevails in litigation, strikes us as a questionable workaround to the general prohibition on contingency payments for expert witnesses. See United States v. Dawson, 425 F.3d 389, 394 (7th Cir. 2005) (“[A]n expert witness ... may not be paid more if the party for whom he is testifying wins the case.”), supplemented on reh’g, 434 F.3d 956 (7th Cir. 2006); Tagatz v. Marquette University, 861 F.2d 1040, 1042 (7th Cir. 1988) (“There is a rule against employing expert witnesses on a contingent-fee basis....”); Milfam II LP v. American Commercial Lines, LLC, No. 4:05-CV-0030-DFH-WGH, 2006 WL 3247149, at *2 (S.D. Ind. Mar. 30, 2006) (“[A] highly unusual contingent fee for an expert witness raises obvious questions of credibility.”); Model Rule of Profl Conduct 3.4 cmt. [3] (Am. Bar Ass’n 1983) (“The common law rule in most jurisdictions is that it is improper to ... pay an expert witness a contingent fee.”); cf. Accrued Financial Servs., Inc. v. Prime Retail, Inc., Civ. No. JFM-99-2573, 2000 WL 976800, at *3 (D. Md. June 19, 2000) (“Financial arrangements that provide incentives for the falsification or exaggeration of testimony threaten the very integrity of the judicial process which depends upon the truthfulness of the witnesses.”), aff'd, 298 F.3d 291 (4th Cir. 2002).
Aong with his conclusory statements, Cuozzo included a list of nine factors he deems important in comparing architectural designs, but he failed to apply, or at least failed to explain how he applied,. those factors to the plans in this case. Assuming for the sake of argument that this portion of Cuozzo’s declaration has any evidentiary value at all, his broad assertion does not present a genuine issue of fact with respect to Dr. Greenstreet’s detailed comparisons. See Humphreys & *1105Partners Architects, L.P. v. Lessard Design, Inc., 790 F.3d 532, 541 (4th Cir. 2015) (plaintiff failed to establish substantial similarity where plaintiffs witnesses made conclusory assertions that designs had a “very similar appearance” and that the “arrangement and composition of spaces and elements ... represented] substantially similar features”).
Design Basics argues that Dr. Green-street and the district court improperly “dissected” the plans instead of considering their overall concept and feel. As Design Basics points out, we noted in Atan, Inc. v. North American Philips Consumer Electronics Corp. that the substantial similarity test “does not involve ‘analytic dissection and expert testimony,’ but depends on whether the accused work has captured the ‘total concept and feel’ of the copyrighted work.” 672 F.2d 607, 614 (7th Cir. 1982) (citations omitted), superseded in part by statute on other grounds. But that is not all we said in Atari. We added: “While dissection is generally disfavored, the ordinary observer test, in application, must take into account that the copyright laws preclude appropriation of only those elements of the work that are protected by the copyright.” Id. (emphasis added); see also Peters, 692 F.3d at 632 (“If the copied parts [of a work] are not, on their own, protectable expression, then there can be no claim for infringement....”).
Because so much of the content of the house plans at issue is dictated by functional requirements and industry norms, the district court would have erred if it had surveyed the plans from 30,000 feet, or even 500 feet. The court had to take a close look, and it did so using Dr. Green-street’s report as an analytic tool. Design Basics chides the district court for failing to chronicle its side-by-side review of the plans to Design Basics’ satisfaction, but our focus is not on the brevity or length of district court opinions. The judge understood the standard for assessing an infringement claim at summary judgment and applied that standard correctly to the facts before him.
Design Basics also criticizes Dr. Green-street’s description of the law of architectural copyright. We need not decide whether Dr. Greenstreet’s legal discussion was accurate in all respects, as the district court did not rely on it. Rather, the district court used Dr. Greenstreet’s factual analysis to decide whether there was a genuine issue of fact on striking similarity. That was an appropriate use of this essentially unrebutted evidence.
Design Basics has not offered evidence from which a reasonable jury could conclude that Lexington’s accused plans are substantially similar to protectable expression in Design Basics’ plans, let alone strikingly similar. To whatever extent the parties’ plans resemble one another, they likewise resemble countless other home designs in a crowded market. See Zalewski, 754 F.3dat 107 (“Defendants’ houses shared Plaintiffs general style, but took nothing from his original expression.”). For that reason alone, Lexington was entitled to judgment as a matter of law.
2. Access
Design Basics’ claim fails for another reason. It has not offered evidence that Lexington or its agents had a reasonable possibility of access to the plans at issue. See Selle, 741 F.2d at 901 (“The plaintiff must always present sufficient evidence to support a reasonable possibility of access because the jury cannot draw an inference of access based upon speculation and conjecture alone.”). Without access, there can be no copying and no viable claim for copyright infringement.
To prove access, Design Basics relies on two categories of evidence: (1) weak cir-*1106eumstantial evidence that Lexington’s agents may have been aware of some of Design Basics’ other plans; and (2) the existence of www.designbasics.com, a website that Design Basics hails as prominent and user-friendly. This evidence does not raise a genuine issue of material fact as to access.
For the first category of evidence, Design Basics cites Cuozzo’s declaration that the firm disseminates its plans to homebuilders’ associations and home improvement stores, as well as Patrick Carmichael’s declaration that Design Basics “regularly and systematically” mailed publications to members of the Brown County Homebuilders Association in Wisconsin. Lexington was a member of that association from 2004 until around 2012 or 2013. But Design Basics’ own records show that it last shipped an order to the association in 1994—a decade before Lexington joined the association and years before three of Design Basics’ plans at issue were even created. The association’s executive officer swore in an affidavit that (1) the association does not disseminate plans or plan books to its members; and (2) if the association had received any plan materials from Design Basics, it would have discarded those materials at the time of receipt.
There is no evidence that Design Basics ever sent any plans or plan catalogs directly to Lexington. During discovery, Lexington tracked down four old plan books in its office, two of which contained plans created by Design Basics. But the plans at issue do not appear in these old books. Lexington’s president and part-owner swore that he never saw the plans at issue prior to this lawsuit and was unaware of the books in his office. A minority shareholder at Lexington, the corporate controller, and Lexington’s residential drafter gave similar testimony. These statements are unre-butted. Design Basics’ Cuozzo had no idea whether Lexington ever possessed any of the plans at issue. Design Basics employees Paul Foresman and Greg Dodge likewise did not know whether Lexington ever had access to the plans. Design Basics has no direct proof of access and no probative circumstantial evidence that Lexington may have viewed the plans at issue.7
Design Basics also relies on its website, www.designbasics.com. Cuozzo praised the website as “user-friendly and easily accessible,” and he said that Design Basics “advertises and markets all of its copyrighted works” on the website. Simply having posted its designs on its website cannot spare Design Basics from summary judgment.
Cuozzo did not know when any of the plans at issue were uploaded to the site, nor did he know whether Lexington or its agents ever accessed the site. Design Basics has introduced no evidence concerning its web traffic, its web search rankings, or the number of times (if any) that the plans at issue have been viewed or downloaded. Design Basics argues in essence that the mere existence of the website establishes a reasonable possibility that Lexington accessed its plans.
We have not previously considered whether an Internet presence, without *1107more, can satisfy the access element of a copyright infringement claim. The Fourth Circuit rejected a similar argument in Building Graphics, Inc. v. Lennar Corp., 708 F.3d 573 (4th Cir. 2013). Piaintiff Building Graphics argued that because two of its home plans were available on the Internet, “a reasonable possibility of access can be inferred.” Id. at 580. The defendant admitted that it surveyed online plans as “part of due diligence,” but it limited its diligence to its competitors’ current offerings, and Building Graphics’ plans were no longer being built in the region. Id. In affirming summary judgment for the defendant, the Fourth Circuit said that it was “at best a mere possibility” that Building Graphics’ plans “would be considered as part of the due diligence process.” Id. The court added that Building Graphics made no showing that the plans were “prominently featured” on its website such that the plaintiff would “easily find them or stumble across them.” Id.
The Fourth Circuit’s analysis is persuasive. Design Basics attempts to distinguish Building Graphics on the ground that Lexington was ostensibly familiar with Design Basics, whereas there was apparently no evidence in Building Graphics that defendant was familiar with plaintiff or its work. But merely because Lexington knew there is such a firm as Design Basics, a jury could not reasonably infer that Lexington accessed its website and discovered—among the many hundreds of plans catalogued—the four plans at issue here.
Design Basics cites a handful of district court opinions suggesting that an Internet presence may. satisfy the access requirement for an infringement claim.8 The weight of authority, and in our view the more persuasive analysis, rejects this argument. See, e.g., Cain v. Hallmark Cards, Inc., Civ. No. 3:15-00351-JWD-EWD, 2016 WL 3189231, at *5 (M.D. La. June 6, 2016) (“[E]stablishing a bare possibility of access is not enough, and a plaintiff must prove that ... a defendant had a reasonable possibility of viewing the work. Applying this doctrine, courts have consistently refused to treat internet publication alone as sufficient to engender this requisite possibility.”) (alterations, internal quotation marks, and citations omitted); Stabile v. Paul Smith Ltd., .137 F.Supp.3d 1173, 1187 (C.D. Cal. 2015) (“Even if an inference can be drawn that [Plaintiffs painting] appeared at one time on Plaintiffs website, simply displaying an image on a website for an undeterminable period of time is insufficient to demonstrate that it was widely disseminated.”); Loomis v. Cornish, No. CV 12-5525 RSWL (JEMx), *11082013 WL 6044345, at *12 (C.D. Cal. Nov. 13, 2013) (“The availability of a copyrighted work on the Internet, in and of itself, is insufficient to show access through widespread dissemination.”), aff'd, 836 F.3d 991 (9th Cir. 2016); Hayes v. Minaj, No. 2:12-cv-07972-SVW-SH, 2012 WL 12887393, at *3 (C.D. Cal. Dec. 18, 2012) (same); O’Keefe v. Ogilvy & Mather Worldwide, Inc., 590 F.Supp.2d 500, 515 (S.D.N.Y. 2008) (same). On this issue, we agree with the reasoning in these decisions and with Building Graphics.
We do not draw here a bright line as to the quantity or quality of evidence, in addition to a web presence, a plaintiff must offer to raise a genuine issue of fact concerning access. We decide only that the existence of the plaintiffs copyrighted materials on the Internet, even on a public and “user-friendly” site, cannot by itself justify an inference that the defendant accessed those materials. It follows that a plaintiff who cannot show striking similarity and whose evidence of access reduces to the mere existence of a website cannot survive summary judgment on a copyright infringement claim. The district court correctly granted summary judgment to Lexington.
The judgment of the district court is AFFIRMED.
Appendix
*1109[[Image here]]
Design Basics' "Aspen" Lexington's "Carlisle"
[[Image here]]
Design Basics' "Taylor" Lexington's (first floor) (first floor) "Ashwood"
*1110[[Image here]]
Design Basics' "Womack' (first floor) Lexington's "Easton' (first floor)
[[Image here]]
Design Basics' "Kendrick" Lexington's "Oakridge'
. Counsel for Design Basics asserted in oral argument that, in addition to generating substantial revenue through litigation, the firm licenses home designs for consumer use. By counsel’s estimate, the firm sold around 600 licenses in the past year. It is unclear how much revenue these licenses generated, though for the four plans at issue, construction licenses start at around $450.
. The Architectural Works Copyright Protection Act was enacted to bring the United States into compliance with the Berne Convention for the Protection of Literary and Artistic Works. See H.R. Rep. No. 101-735, reprinted in 1990 U.S.C.C.A.N. 6935, 6936-37; see also 7 Patry on Copyright § 23:45 (discussing history of United States participation in Berne Convention).
. In Ty, Inc., we found striking similarity between the parties' beanbag animal dolls. They bore a remarkable resemblance to one another yet bore no particular resemblance either to real animals or to fictional animals in the public domain. 132 F.3d at 1170-71. In Bucklew v. Hawkins, Ash, Baptie & Co., we found striking similarity between plaintiffs and defendant’s electronic forms, which contained identical patterns of bolding, identical headings, the same font, and (most damning) an identical “output range” that served ho purpose. 329 F.3d 923, 926-27 (7th Cir. 2003). We observed that mapmakers sometimes include a fictitious geographical feature—a "copyright trap”—in their maps. Id. at 926. If that trap is duplicated in another map, "the inference of copying is compelling.” Id.
. Courts have applied similar reasoning in other contexts where opportunities for creative expression are narrow. See Feist Publications, 499 U.S. at 349, 111 S.Ct. 1282 (explaining that the "copyright in a factual compilation is thin”); Schrock v. Learning Curve Int’l, Inc., 586 F.3d 513, 521 (7th Cir. 2009) (explaining that the "copyright in a derivative work is thin”); Satava v. Lowry, 323 F.3d 805, 812 (9th Cir. 2003) (plaintiff who created glass jellyfish sculpture enjoyed only thin copyright protection to the extent his artistic choices were "not governed by jellyfish physiology or the glass-in-glass medium”).
. For example, among the 62 differences between Design Basics’ Womack and Lexington’s Easton, Dr. Greenstreet observed: the Easton had a double gable on its left roof while the Womack had a single gable; the Easton had a low stone wall along the front elevation while the Womack had none; and the Easton had no window feature in its garage gable while the Womack’s garage gable had a shuttered window feature. The Easton's front door was tri-paneled with no glass, while the Womack’s front door had six panels and a vertical glass feature. The Easton’s garage was almost square, while the Womack’s garage was oblong, and the internal garage doors were configured differently. The Ea-ston’s pantry had double cabinet doors while the Womack's pantry had a single walk-in door, and the dining areas differed with respect to shape, dimension, size, and use of doors. The Easton's master bedroom had two windows, while the Womack had a single window. The Easton had a tray ceiling and a walk-in closet, but the Womack had neither. The toilet and washbasin in the Easton's master bathroom were positioned differently than in the Womack. The Easton bathroom also featured a closet. Perhaps most striking, the Easton had a full basement accessible by a stairwell to the right of the family room, while the Womack was built on a slab. We could go on. The point simply is that these two designs, like the other pairs at issue in this case, differ in many respects, casting further doubt on Design Basics’ already tenuous claim of copyright infringement.
. Design Basics included in its appellate brief images that it calls "demonstrative aids.” These are side-by-side renderings of each accused plan and its Design Basics counterpart. But Design Basics modified these images by, in its words, "re-sizing, flipping, and juxtaposing” them. We have not used these manipulated images to assess the claim of substantial similarity. Design Basics did not even *1104present its manipulations to the district court. See Fed. R. App. P. 10(a) ("The following items constitute the record on appeal: (1) the original papers and exhibits filed in the district court....”); Midwest Fence Corp. v. U.S. Dep't of Transportation, 840 F.3d 932, 946 (7th Cir. 2016) (“As a general rule, we will not consider evidence on appeal that was not before the district court when it rendered its decision. Adding new evidence would essentially convert an appeal into a collateral attack on the district court's decision.”). The manipulated images are also too small and grainy to be of any use. We rely instead on the images filed in the district court, which are an appendix to this opinion.
. Lexington’s draftsman, Shane DuPuis, worked at Hoida Lumber from 1995 to 1997. The record reflects that Hoida received many Design Basics catalogs between 1992 and 2001. But three of the four plans at issue—the Kendrick, Taylor, and Womack plans—were not even created until the early-to-mid 2000s, years after DuPuis left Hoida. Even if Hoida did receive one or more plan books containing the fourth plan—the Aspen—there is no evidence that DuPuis reviewed this plan. He testified that he could not remember whether he ever viewed any plan books while employed at Hoida and that he had never seen any of the plans at issue until this litigation commenced.
. See Conference Archives, Inc. v. Sound Images, Inc., Civ. No. 3:2006-76, 2010 WL 1626072, at *10 (W.D. Pa. Mar. 31, 2010); Green Bullion Financial Servs., LLC v. Money4Gold Holdings, Inc., 639 F.Supp.2d 1356, 1361 (S.D. Fla. 2009); Allen v. Ghoulish Gallery, Civ. No. 06cv371 NLS, 2007 WL 4207923, at *20 (S.D. Cal. Nov. 20, 2007). However, several of the cases that Design Basics relies upon are readily distinguishable either factually or procedurally and are thus of limited persuasive value in this case. See Live Face on Web, LLC v. AZ Metroway, Inc., No. 5:15-cv-01701-CAS(KKx), 2016 WL 4402796, at *1 (C.D. Cal. Aug. 15, 2016) (defendant failed to appear or defend in suit, and plaintiff moved for default judgment); Schenck v. Orosz, No. 3:13-cv-00294, 2016 WL 912281, at *6 (M.D. Tenn. Mar. 10, 2016) (plaintiffs and defendants both marketed goods via Amazon.com); Nat'l Board of Medical Examiners v. Optima University LLC, No. 1:09-cv-01043-JDB-cgc, 2011 WL 7615071, at *6 (W.D. Tenn. Sept. 29, 2011) (defendants sat for exams to gain access to copyrighted questions and were found in possession of notes and screenshots); Frank Betz Assocs. v. J.O. Clark Construction, LLC. Civ. No. 3:08-cv-00159, 2010 WL 2253541, at *14 (M.D. Tenn. May 30, 2010) (defendants did not dispute that they actually received copies of magazine containing home plans at issue, though they denied reading magazine).