HOLMES, Circuit Judge.

Brian J. Connolly, (J. Thomas Macdonald with him on the briefs), Otten, Johnson, Robinson, Neff & Ragonetti, P.C., Denver, Colorado, for Defendant-Appellant.

Jeremy A. Fielding, Lynn Pinker Cox & Hurst, LLP, Dallas, Texas (David S. Coale, Jonathan D. Kelley and Paulette C. Miniter, Lynn Pinker Cox & Hurst, LLP, Dallas Texas; and Steven J. Perfrement, Bryan Cave Leighton Paisner LLP, Denver Colorado, with him on the brief), for Plaintiff-Appellee.

Laura K. Wendell and Susan L. Trevarthen, Weiss Serota Helfman Cole & Bierman, P.L., Coral Gables, Florida, filed an amicus curiae brief on behalf of International Municipal Lawyers Association supporting Defendant-Appellant.

Todd G. Messenger, Fairfield and Woods, P.C., Denver, Colorado, filed an amicus curiae brief on behalf of The Colorado Municipal League supporting Defendant-Appellant.

Before HARTZ, HOLMES, and CARSON, Circuit Judges.

HOLMES, Circuit Judge. The Town of Castle Rock, Colorado ("Castle Rock" or "Town") enacted a 7:00 p.m. curfew on commercial door-to-door solicitation (the "Curfew"). Aptive Environmental, LLC ("Aptive") sells pest-control services through door-to-door solicitation and encourages its salespeople to go door-to-door until dusk during the traditional business week. When Aptive came to Castle Rock in 2017, it struggled to sell its services as successfully as it had in other nearby markets. Blaming the Curfew, Aptive sued Castle Rock for violating its First Amendment rights and sought an injunction against the Curfew's enforcement. After a bench trial, the district court permanently enjoined Castle Rock from enforcing the Curfew. Castle Rock appealed. Exercising jurisdiction under 28 U.S.C. § 1291, we affirm the district court's judgment, concluding that Castle Rock has failed to demonstrate that the Curfew advances its substantial interests in a direct and material way. In the following discussion, we summarize the relevant factual and procedural background, assure ourselves that Aptive has standing to challenge the Curfew, and explain why Castle Rock has failed to carry its burden of demonstrating that its Curfew is constitutional.

I

A

On appeal, Aptive's dispute with Castle Rock is based solely on Castle Rock's decision to enact the Curfew on commercial door-to-door solicitation. Below, we discuss the purported impetus for the Curfew, the specific provisions of Castle Rock's laws that established the Curfew, and the Curfew's impact on Aptive.

1

Prior to 2008, Castle Rock did not have a curfew on door-to-door solicitation. However, in 2007, a door-to-door solicitor approached a member of Castle Rock's elected town council while the council member was working in his garage. The council member was startled by the interaction, and, at subsequent town council meetings, he suggested that Castle Rock enact further restrictions on door-to-door solicitation. At these meetings, the town council decided to specifically target commercial door-to-door solicitation because Castle Rock's attorney had explained that regulation of noncommercial speech would raise constitutional concerns. Meeting notes similarly reflect that council members discussed the "fact that even if sales could be limited, religious and other groups who went door to door to educate and inform citizens were protected by the First Amendment." Aplt.’s App., Vol. V, at A1321 (Town Council Study Session Notes, dated Aug. 21, 2007).

With the focus on commercial door-to-door solicitors, Castle Rock considered a monograph on door-to-door solicitation prepared by the Colorado Municipal League, which summarized various ordinances existing elsewhere. Id. , Vol. VI, at A1354 71 (Mem. re: Residential Door-to-Door Solicitation). Castle Rock's attorney also talked to other municipalities about ordinances that they had enacted. Based on this research, Castle Rock's attorney recommended adding a curfew barring commercial door-to-door solicitation from 7:00 p.m. to 9:00 a.m. because, in his view, such an ordinance "provide[d] a reasonable balance for residents and [commercial] solicitors." Id. at A1378 79 (Agenda Mem., dated Mar. 25, 2008).

The town council and Castle Rock employees discussed the purported need for such an ordinance with the Town's police chief. The police chief reported that the police had received eight "reports" concerning door-to-door solicitation and twenty to thirty more informal solicitation "complaints," some of which involved door-to-door solicitation, so far that year. Id. , Vol. V, at A1313 (Email re: Proposed Memo on Door-to-Door Solicitation, dated Aug. 15, 2007). He noted that several citizens reported feeling "harassed" or "intimidated" by solicitors. Id. Castle Rock's clerk circulated a memorandum that similarly stated "the Police Department often receives calls from citizens concerned about various salespersons wondering [sic] about their neighborhoods." Id. at A1305 (Agenda Mem., dated Aug. 21, 2007). And, more specifically, the clerk noted that the Town had received one complaint about an individual soliciting at 9:45 p.m. Id. at A1327 (Agenda Mem., dated Oct. 23, 2007).

Finally, in addition to this discussion with Castle Rock's attorney and the police department, a former council member and a former mayor later testified that the ordinance—including the Curfew—was enacted in response to citizens’ privacy and safety concerns. As to privacy, the former town council member testified that, before the ordinance's enactment, he had discussed the ordinance with his neighbors and that they had agreed that 7:00 p.m. was "a reasonable time" for a curfew. Id. at A1081-82 (Trial Test. of Mitch Dulleck, dated Feb. 20, 2018). Some neighbors told him that they had negative experiences with "aggressive" solicitors. Id. at A1084. Other council members had similar conversations with constituents. And as for public safety, the former mayor of Castle Rock testified that "when [the] Council was considering the Curfew, there ‘probably was [sic] some thoughts’ that ‘reasonable people would think that people walking around their neighborhood or up to their home could potentially be somebody that might create a crime in the town.’ " Id. , Vol. II, at A234 (Joint Stipulation as to Facts, filed Jan. 10, 2018). The former council member echoed these concerns, stating that he "definitely" thought that it "could be a possibility" that individuals posing as door-to-door solicitors were engaging in criminal activity. Id. , Vol. V, at A1085-86.

Nevertheless, the former council member said that he was unaware of any crimes that had been committed by commercial solicitors. And the former mayor testified that

to the best of his recollection ... prior to passing the ... Ordinance there was no discussion or analysis by the Town Council of (a) crime in Castle Rock; (b) solicitation-related crime in Castle Rock; (c) crime committed by commercial solicitors in Castle Rock; (d) crime committed by commercial solicitors in Castle

Rock after 7:00 p.m.; or (e) how a 7:00 p.m. curfew would protect public safety and privacy.

Id. , Vol. II, at A228; see also id. at A228-29 (the former mayor noting that "there was no discussion by the Town Council of specifically how a 7:00 p.m. curfew that applied only to for-profit commercial solicitors would protect public safety and privacy").

2

The Castle Rock town council enacted an ordinance to address their concerns regarding commercial door-to-door solicitation on April 8, 2008 (the "2008 ordinance"). The 2008 ordinance included certain prefatory clauses stating that the ordinance was being enacted to protect Castle Rock's citizens’ privacy and public safety. As to the latter, the prefatory clauses specifically stated that a significant percentage of criminal activity involved trespass, that individuals posed as door-to-door solicitors to commit crimes, that crime often occurs at night, and that unregulated door-to-door solicitation posed a risk to citizens. But, other than the anecdotal information summarized above, the town council did not have evidentiary support for these prefatory statements.

The 2008 ordinance required commercial solicitors to register with the town clerk, pay a fee, and follow various requirements, including—as most relevant here—the Curfew. Specifically, commercial solicitors were not to "[e]nter upon any private property within the Town after seven o'clock P.M. (7:00 P.M.) and before nine o'clock A.M. (9:00 A.M.)." Id. , Vol. VI, at A1450-51 (Ordinance No. 2008-15, dated Apr. 8, 2008). The 2008 ordinance also made it unlawful for any business entity to "instruct, direct, command, order, organize, or otherwise arrange for any person to engage in solicitation" in violation of the ordinance's provisions. Id. at A1451. The 2008 ordinance also included a criminal penalty: those violating it would "be punished by a fine of not more than one thousand dollars ($1,000.00) or by imprisonment not to exceed one (1) year, or by both such fine and imprisonment." Id. at A1452.

The 2008 ordinance expressly exempted noncommercial solicitors from its registration requirements and the Curfew. Individuals falling into the classification of noncommercial solicitors included those who engage in door-to-door solicitation "for the primary purpose of":

[a]ttempting to enlist support for or against a particular religion, philosophy, ideology, political party, issue or candidate, even if incidental to such purpose the canvasser accepts the donation or money for or against such cause; or

....

[a]ttempting to obtain a donation to a particular patriotic, philanthropic, social service, welfare, benevolent, educational, civic, fraternal, charitable, political or religious purpose, even if incidental to such purpose there is the sale of some good or service ....

Id. at A1447-48.

The Castle Rock town council considered a new ordinance addressing door-to-door solicitors in the latter part of 2013 and ultimately approved it on January 7, 2014 (the "2014 ordinance"). Castle Rock's mayor testified that "to the best of her recollection," she "was unaware of any research, studies, investigation or analysis performed by the Town prior to passing the 201[4] Ordinance," other than "non-specific complaints from residents and recommendations ... from the Town staff." Id. , Vol. II, at A231.

The 2014 ordinance primarily clarified and amplified certain definitions, and it also "expanded the nature of the background check, the grounds for denial, and required solicitors to wear identifying badges." Id. , Vol. IV, at A851 (Oral Ruling, filed Apr. 6, 2018). As relevant to our resolution of this appeal, the 2014 ordinance's changes to the 2008 ordinance are largely immaterial. Notably, the 2014 ordinance incorporated by reference the 2008 ordinance's prefatory statements—that is, "the findings and recitals"—"as support for continuation" of its regulation of commercial door-to-door solicitors and continued to impose registration requirements and the Curfew solely on commercial solicitors. Id. , Vol. V, at A1295 (Ordinance No. 2013-43, dated Jan. 7, 2014).

In all material respects, the Curfew remained the same. See id. , Vol. III, at A640 (stipulating that "[t]he 7:00 p.m. solicitation curfew ... has been in effect since the Town Council's approval of the 2008 Ordinance"). Like the 2008 ordinance, the 2014 ordinance provided that "[n]o [commercial] solicitor shall .... [e]nter upon any private property within the Town after 7:00 p.m. and before 9:00 a.m." Id. , Vol. V, at A1300-01. Furthermore, it continued to make it "unlawful for any person, organization, corporation, or business entity to instruct, direct, command, order, organize, or otherwise arrange for any person to engage in solicitation in violation" of the ordinance. Id. at A1301. Noncommercial solicitors were again expressly exempted from the ordinance's requirements; notably, they could solicit door-to-door unimpeded by the Curfew.

Because Aptive's conduct was regulated by Castle Rock's 2014 ordinance, we reference its terms below. We recognize, however, that we must carefully examine the circumstances leading up to the enactment of the 2008 ordinance in order to understand the legislative concerns that prompted Castle Rock to enact the Curfew—a curfew that carried over virtually verbatim from the 2008 ordinance to the 2014 ordinance.

3

Aptive engages in door-to-door sales of pest-control services. Aptive ordinarily encourages its solicitors, who are employed as independent contractors, to work during the traditional five-day business week "through dusk so as to contact people" that were not home earlier in the day. Id. , Vol. II, at A222. More specifically, the independent contractors "set their own daily schedule," but Aptive encourages them "to work between 10:00 and 11:00 a.m., take a lunch break between 2:30 and 4:00 p.m., and complete work at dark each day that they worked." Id. at A221-22. Aptive claims that "those [salespeople] that stay out" until dusk "do better" than those that stop earlier in the day. Id. at A224. And Aptive's data confirm that, in the neighboring Denver area, some of its highest rates of interaction with "decision makers" occur after 7:00 p.m. Id. , Vol. V, at A1293 (Company Stats).

On August 4, 2017, "Aptive applied for and received permits for its representatives to conduct door-to-door solicitation activities" in Castle Rock. Id. , Vol. I, at A87 (Scheduling Order, filed Aug. 29, 2017). Aptive complied with Castle Rock's Curfew and instructed its sales force to stop engaging in door-to-door solicitation at 7:00 p.m., even though it would normally encourage them to work through dusk. "In 2017, Aptive made 55 sales in [Castle Rock], which were generally evenly distributed throughout the day, though none were sold after 7:00 pm." Id. , Vol. II, at A225. Aptive claimed that it was "half as effective in Castle Rock as [it was] throughout the rest of the areas that didn't require the curfew." Id. , Vol. V, at A1197 (Trial Test. of Robert Hansen, dated Feb. 21, 2018). And Aptive's data confirm that it made fewer than half the average sales per hour in Castle Rock than it did in the neighboring Denver area. After only a brief period, Aptive ceased operations in Castle Rock.

B

Aptive filed suit alleging that the Curfew unconstitutionally burdened its First Amendment right to engage in door-to-door solicitation. The district court denied a pretrial motion to dismiss, in which Castle Rock had argued that Aptive lacked standing to challenge the Curfew. The court then held a bench trial in February 2018. In addition to presenting the above evidence about why Castle Rock had originally enacted the Curfew, Castle Rock put on further evidence about its present interests in maintaining the Curfew. Aptive also put on evidence about its experience with the Curfew. We summarize this evidence as well as the district court's order enjoining enforcement of the Curfew. 1

At trial, Castle Rock sought to demonstrate that commercial door-to-door solicitation after 7:00 p.m. was causing ongoing harm. But Castle Rock's statistics revealed that, while Castle Rock received twenty-five complaints relating generally to solicitation in 2016, there was no evidence that any of these concerned commercial solicitation after 7:00 p.m. Indeed, only a fraction of these complaints involved solicitation after the Curfew at all: one complaint concerned a noncommercial solicitor who was soliciting after 7:00 p.m.; there were no complaints about commercial solicitation after 7:00 p.m. Expanding the focus beyond the Curfew's time window and looking at total complaints in 2016, there was still no evidence of complaints about registered commercial solicitors like Aptive's. Two of the complaints involved noncommercial solicitors, six involved commercial solicitors that had failed to register with the Town, and seventeen involved individuals whose classification was unknown.

Likewise, Castle Rock's statistics showed that while Castle Rock received sixty-one solicitation-related complaints in 2017, ten of the complaints were about noncommercial solicitors after 7:00 p.m., and none of the complaints were about commercial solicitors after 7:00 p.m. And if we once again consider complaints throughout the entire day, we find only one related to a registered commercial solicitor. In contrast, there were fourteen complaints related to noncommercial solicitors, sixteen related to unregistered commercial solicitors, and twenty-seven related to individuals with an unknown classification.

Aside from these statistics, Castle Rock relied on testimony from individual town members about the scope of ongoing problems caused by commercial door-to-door solicitation. Castle Rock's current town manager characterized door-to-door solicitation as "a low-level, constant issue." Id. , Vol. IV, at A913 (Trial Test. of David Corliss, dated Feb. 20, 2018). Castle Rock introduced testimony that its clerk received an average of about one complaint per week about solicitors generally, i.e., without regard to whether they were commercial, noncommercial, or registered. Some indeterminate number of these complaints concerned solicitors that were coming after the Curfew and solicitors that were not displaying their badges.

Castle Rock also presented testimony about the ostensible threat to public safety posed by door-to-door solicitation. Castle Rock's police chief stated that door-to-door solicitation "gives somebody an opportunity to enter inside a home, or even if they're at the threshold of the doorway, to look inside the house, gather information by talking to the individual." Id. , Vol. V, at A1126 (Trial Test. of John Cauley, dated Feb. 21, 2018). He also stated "common sense tells us that as it gets later in the evening, people are a little bit more anxious." Id. at A1128, A1153. In response to this anxiety, "the 7:00 p.m. curfew provides clarity ... for the community members." Id. at A1125. The police chief, however, was unable to point to a specific instance where a commercial solicitor was accused of any solicitation-related crime or where an individual had posed as a door-to-door solicitor to hide criminal intent. Indeed, much of the police chief's testimony effectively undermined Castle Rock's safety rationale for the Curfew.

For example, he testified that "[m]ost crimes occur outside of the nighttime hours" which is at odds with one of the prefatory statements of the 2008 ordinance (incorporated into the 2014 ordinance). Id. at A1132; see id. at A1150 (answering "No" to the question, "You don't have any data that would reflect the fact that for-profit commercial solicitors tend to commit crimes more frequently between the hours of 7:00 p.m. [ ] and 10:00 in the evening, as opposed to some other time, do you?"). But cf. id. , Vol. VI, at A1445 (prefatory statement providing, "WHEREAS , criminal activity on private property often occurs during nighttime hours"). And he further testified that "a person is probably not more likely to commit a crime to persons or property arising out of door-to-door solicitation after 7:00 p.m. than before 7:00 p.m.," and that "moving the Town's Curfew from 7:00 p.m. to dusk would not materially hamper the Castle Rock Police Department's ability to protect its residents from crime." Id. , Vol. II, at A235. He admitted that he had seen no data suggesting that registered commercial solicitors had been a problem.

And the chief testified that the data actually suggested that Castle Rock's problem was with unregistered solicitors (a group to which Aptive's solicitors, of course, did not belong). Similarly, Castle Rock's current town manager testified that he was unaware of Castle Rock ever receiving any complaints about crimes committed by commercial solicitors who had registered with the town. Indeed, Castle Rock stipulated that it "has no record of ever charging or convicting a registered commercial solicitor for committing a crime to persons or property." Id. , Vol. III, at A643.

Castle Rock also sought to justify the Curfew as a necessary response to ongoing intrusions into privacy caused by commercial door-to-door solicitation. Castle Rock's current town manager testified that the Curfew "reflect[s] the common sense view that as the hours get—hours get later, past dinnertime, people are moving toward more private activities within their home, and they're not wanting to be inconvenienced or bothered by an uninvited stranger on [sic] their door." Id. , Vol. IV, at A940. He testified that some citizens had commented that they want him to "keep the curfew the way that it is," that they "value their time at home," and that "they have a problem with uninvited solicitations showing up on their doorstep after 7 o'clock." Id. at A920-21, A926.

Castle Rock's mayor testified generally from her personal experiences that commercial solicitors affected her family's evening routines through "persistent" and "aggressive" knocking. Id. , Vol. V, at A1177-78 (Trial Test. of Jennifer Green, dated Feb. 21, 2018). She testified about two particularly negative experiences that she had with solicitors: one entered her garage to talk to her without her permission, and the other rang her doorbell and opened her storm door, apparently while her front door was open. Id. at A1176. She clearly identified the one who entered the garage as a commercial solicitor but was not sure of the intentions of the one who rang the doorbell. Id. She described the experiences as "a little terrifying" and "very uncomfortable." Id. However, both of these specific experiences occurred during the daytime. See id. at A1182. In the mayor's experience, noncommercial solicitors were more respectful than commercial solicitors. She additionally believed that Castle Rock's residents supported the Curfew. Finally, a former council member expressed that he personally thought the Curfew helped give citizens privacy in their homes: "[b]y having a time when people could and could not come to your door, that gives you an at least a time period that you know you will you will have privacy." Id. at A1080-81.

2

Aptive called one witness, a regional operations manager, as part of its case. He testified as explained above that Aptive ordinarily encourages its salespeople to solicit until dusk during the business week, but that Aptive had instructed its salespeople to follow Castle Rock's Curfew and that this had caused the salespeople to be significantly less effective. Aptive's salespeople ceased soliciting in Castle Rock after only seven days. Aptive's regional operations manager testified that Aptive would "[p]robably not" return to Castle Rock if the Curfew was kept in place, but that "[p]resumably, if things [i.e., the Curfew] change, [Aptive will] be coming back to Castle Rock." Id. at A1202, A1205-06.

3

After the trial concluded, the district court gave an oral ruling from the bench. Id. , Vol. IV, at A841. As in its pretrial ruling, the court rejected the challenge to Aptive's standing. Turning to the merits, the court found the record to be "actually devoid of evidence showing that commercial solicitation after 7:00 p.m. causes a real harm in the form of increased crime or decreased public safety." Id. at A868. Likewise, the court concluded that "the curfew does not materially advance the Town's interest in protecting residents’ privacy after 7:00 p.m." Id. at A866-67. It thus concluded that "the Town has not established that the curfew materially impacts the legitimate objectives of reducing crime and promoting public safety," and that the Curfew thus unconstitutionally burdened Aptive's First Amendment rights. Id. at A870-71, A873. Consequently, the court permanently enjoined the operation of the Curfew. Castle Rock timely appealed.

II

Castle Rock first argues that Aptive lacks standing to challenge the Curfew. We briefly set out the three elements of constitutional standing before explaining why each element is satisfied here.

A

Article III of the United States Constitution restricts the jurisdiction of federal courts to the adjudication of "Cases" or "Controversies." U.S. CONST. art. III, § 2, cl. 1. To satisfy Article III's case-or-controversy requirement, a plaintiff must demonstrate standing by establishing "(1) an ‘injury-in-fact,’ (2) a sufficient ‘causal connection between the injury and the conduct complained of,’ and (3) a ‘likel[ihood]’ that the injury ‘will be redressed by a favorable decision.’ " Susan B. Anthony List v. Driehaus , 573 U.S. 149, 157-58, 134 S.Ct. 2334, 189 L.Ed.2d 246 (2014) (alteration in original) (quoting Lujan v. Defs. of Wildlife , 504 U.S. 555, 560-561, 112 S.Ct. 2130, 119 L.Ed.2d 351 (1992) ). "Put simply, a plaintiff must establish three elements: an injury-in-fact, causation, and redressability." Bronson v. Swensen , 500 F.3d 1099, 1106 (10th Cir. 2007).

"We ... review the district court's rulings on standing de novo." Niemi v. Lasshofer , 770 F.3d 1331, 1344 (10th Cir. 2014). "Because this case proceeded to trial, we ... look to the evidence presented there to determine whether the plaintiffs carried their burden of proving standing." Colo. Outfitters Ass'n v. Hickenlooper , 823 F.3d 537, 544 (10th Cir. 2016) ; accord Glover River Org. v. U.S. Dep't of Interior , 675 F.2d 251, 254 n.3 (10th Cir. 1982). B

Castle Rock argues that Aptive did not suffer an injury-in-fact and that Aptive cannot establish that the Curfew caused its injury. We address and reject each argument before additionally assuring ourselves that the injury-in-fact is redressable. See Frank v. Gaos , ––– U.S. ––––, 139 S. Ct. 1041, 1046, 203 L.Ed.2d 404 (2019) ("We have an obligation to assure ourselves of litigants’ standing under Article III." (quoting DaimlerChrysler Corp. v. Cuno , 547 U.S. 332, 340, 126 S.Ct. 1854, 164 L.Ed.2d 589 (2006) ).

1

"[A] plaintiff satisfies the injury-in-fact requirement where he alleges ‘an intention to engage in a course of conduct arguably affected with a constitutional interest, but proscribed by a statute, and there exists a credible threat of prosecution thereunder.’ " Susan B. Anthony List , 573 U.S. at 159, 134 S.Ct. 2334 (quoting Babbitt v. United Farm Workers Nat'l Union , 442 U.S. 289, 298, 99 S.Ct. 2301, 60 L.Ed.2d 895 (1979) ). For plaintiffs seeking "prospective relief based on a ‘chilling effect’ on speech," we have set forth a three-part test that, if satisfied, would establish the injury-in-fact requirement. Initiative & Referendum Inst. v. Walker , 450 F.3d 1082, 1089 (10th Cir. 2006) (en banc). Such a plaintiff satisfies the injury-in-fact requirement when they present:

(1) evidence that in the past they have engaged in the type of speech affected by the challenged government action; (2) affidavits or testimony stating a present desire, though no specific plans, to engage in such speech; and (3) a plausible

claim that they presently have no intention to do so because of a credible threat that the statute will be enforced.

Id. ; see Holder v. Humanitarian Law Project , 561 U.S. 1, 15-16, 130 S.Ct. 2705, 177 L.Ed.2d 355 (2010) (concluding plaintiffs had standing to bring pre-enforcement First Amendment challenge to the statute because (1) they had engaged in covered conduct before enactment of the statute, (2) they would engage in this conduct "again if the statute's allegedly unconstitutional bar were lifted," (3) the government had enforced the statute against other parties, and (4) the government had not argued it would not enforce the statute against plaintiffs); Winsness v. Yocom , 433 F.3d 727, 732 (10th Cir. 2006) ("[T]o satisfy Article III, the plaintiff's expressive activities must be inhibited by ‘an objectively justified fear of real consequences, which can be satisfied by showing a credible threat of prosecution or other consequences following from the statute's enforcement.’ " (quoting D.L.S. v. Utah , 374 F.3d 971, 975 (10th Cir. 2004) )); see also ACORN v. City of Tulsa , 835 F.2d 735, 739 (10th Cir. 1987) (applying Babbitt ’s standard when evaluating plaintiff's standing to challenge local ordinances under the First Amendment). Each of Initiative and Referendum Institute ’s three requirements is satisfied here.

First, Aptive has presented "evidence that in the past [it has] engaged in the type of speech affected by the challenged government action." Initiative & Referendum Inst. , 450 F.3d at 1089. Aptive established that it has engaged in door-to-door solicitation after 7:00 p.m. in locations near Castle Rock. Indeed, across its operating area, with the exception of Castle Rock, Aptive has encouraged its salespeople to solicit during the business week "through dusk so as to contact people who weren't home" earlier in the day. Aplt.’s App., Vol. II, at A221-22; see id. , Vol. V, at A1194. And Aptive's data show that, in the nearby Denver area, some of its highest rates of interaction with "decision makers" occur after 7:00 p.m. Id. at A1293; id. , Vol. II, at A224. We have previously considered similar evidence concerning conduct in "surrounding" localities in our standing analysis. See Initiative & Referendum Inst. , 450 F.3d at 1090 (considering initiatives prepared or supported by plaintiffs in "surrounding states" when determining whether plaintiffs had standing to challenge supermajority requirement). We conclude that this evidence of Aptive's solicitation after 7:00 p.m. in surrounding localities establishes that Aptive has a history of engaging in the type of speech affected by the challenged government action.

Second, Aptive has demonstrated "a present desire ... to engage in" solicitation in the hours prohibited by the Curfew. Id. at 1089. Aptive desires to solicit during the evening because for those salespeople "that stay out, they do better." Aplt.’s App., Vol. II, at A224. The Curfew had a "significant" effect on Aptive's business because it prevented solicitation during its "most effective time." Id. , Vol. V, at A1194; id. at A1195 ("That's the time when they normally sell the most."). As a result of the Curfew, Aptive's salespeople were "a lot less effective." Id. at A1203. Aptive's data show that it made less than half the average sales per hour in Castle Rock than it did elsewhere. Additional evidence that Aptive desired to solicit during these most profitable hours is intertwined with the third element and therefore discussed below.

The third element requires that Aptive maintain that it "presently ha[s] no intention to [engage in the speech affected by the challenged government action] because of a credible threat that the statute will be enforced." Initiative & Referendum Inst. , 450 F.3d at 1089. Although offering testimony concerning its desire to follow its ordinary practice of encouraging its representatives to work until dusk which they do in the Denver area near Castle Rock Aptive has counseled its representatives not to work until dusk in Castle Rock because of the threat of sanctions posed by the Curfew. And, after only a brief period of solicitation in Castle Rock, Aptive ceased doing business there.

Aptive's regional operations manager testified that Aptive will "[p]robably not" return to Castle Rock if the Curfew is kept in place, but that "[p]resumably, if things [i.e., the Curfew] change, [Aptive will] be coming back to Castle Rock." Aplt.’s App., Vol. V, at A1202, A1205 06. This testimony, when taken alongside the above evidence, demonstrates Aptive's desire to engage in speech affected by the challenged government action, i.e., the second element. And it also demonstrates that Aptive has no present intention to engage in that speech because of a threat of enforcement, i.e., the third element.

And we are able to conclude that the threat of enforcement here is credible . Castle Rock has not indicated that it would not enforce the Curfew against Aptive if Aptive solicited after 7:00 p.m., and Castle Rock has vigorously sought to uphold its Curfew in this litigation. Cf. Humanitarian Law Project , 561 U.S. at 15, 16, 130 S.Ct. 2705 (concluding plaintiffs had established standing when "[t]he Government [told the Court] that it ha[d] charged about 150 persons with violating [the relevant statute], and that several of those prosecutions involved the enforcement of the statutory terms at issue," and "[t]he Government ha[d] not argued to th[e] Court that [the] plaintiffs w[ould] not be prosecuted if they d[id] what they sa[id] they wish[ed] to do" (citation omitted)); Grant v. Meyer , 828 F.2d 1446, 1449 (10th Cir. 1987) (en banc) (finding dispute justiciable after noting that " ‘the State ha[d] not disavowed any intention of invoking the criminal penalty provision ...’ against the[ ] plaintiffs" and that "the State ... [wa]s vigorously upholding the statute in litigation with the[ ] plaintiffs" (quoting Babbitt , 442 U.S. at 302, 99 S.Ct. 2301 )), aff'd , 486 U.S. 414, 108 S.Ct. 1886, 100 L.Ed.2d 425 (1988). And so we conclude that there is a credible threat of enforcement. Aptive has demonstrated "an intention to engage in a course of conduct arguably affected with a constitutional interest, but proscribed by a statute, and there exists a credible threat of prosecution thereunder," i.e., an injury-in-fact. Susan B. Anthony List , 573 U.S. at 159, 134 S.Ct. 2334 (quoting Babbitt , 442 U.S. at 298, 99 S.Ct. 2301 ).

Castle Rock has two primary arguments to the contrary, but neither is persuasive. First, it argues that Aptive told its salespeople to follow the Curfew and that this acquiescence means that Aptive does not have standing. But while Aptive did not violate the law, it circumscribed its behavior to avoid criminal sanctions; this is the precise sort of "chilling effect," Initiative & Referendum Inst. , 450 F.3d at 1088, and "self-censorship," Virginia v. Am. Booksellers Ass'n , 484 U.S. 383, 393, 108 S.Ct. 636, 98 L.Ed.2d 782 (1988), certified questions answered sub nom. Commonwealth v. Am. Booksellers Ass'n , 236 Va. 168, 372 S.E.2d 618 (1988), that demonstrates why the Supreme Court has established the above tests for injury-in-fact in this context. See Susan B. Anthony List , 573 U.S. at 163, 134 S.Ct. 2334 ("Nothing in this Court's decisions requires a plaintiff who wishes to challenge the constitutionality of a law to confess that he will in fact violate that law."); cf. Am. Booksellers Ass'n , 484 U.S. at 392, 108 S.Ct. 636 ("[T]he law is aimed directly at plaintiffs, who, if their interpretation of the statute is correct, will have to take significant and costly compliance measures or risk criminal prosecution."). Aptive has sufficiently demonstrated that it suffered an injury-in-fact under those standards and need do no more.

Second, Castle Rock argues that because Aptive uses independent contractors as salespeople, it is those independent contractors and not Aptive itself that face any threat of enforcement. But the 2014 ordinance applies to those who "instruct, direct, command, order, organize or otherwise arrange for any person to engage in solicitation in violation" of the Curfew. Aplt.’s App., Vol. V, at 1301. Thus, even if we were to assume that Aptive could not rely on the Curfew's application to its independent contractors to establish standing for itself, Aptive's stated intention but for concerns regarding the Curfew's enforcement to encourage (i.e., "instruct" or "arrange for") its independent contractors to solicit until dusk directly implicates the Curfew's express prohibitions and thus makes the Curfew squarely applicable to Aptive.

Furthermore, in Pacific Frontier v. Pleasant Grove City , 414 F.3d 1221 (10th Cir. 2005), in the First Amendment context, we held that a company had standing to challenge a "solicitors licensing procedure" that was enforced against its independent contractors viz. , in determining that the company had standing for its First Amendment claim, we effectively determined that it could rely on a regulation's application to its independent contractors. Id. at 1226, 1228-29. While the injury-in-fact in Pacific Frontier was more concrete the independent contractors in that case were arrested that does nothing to diminish Pacific Frontier ’s determination that an employing company can suffer a First Amendment harm through the enforcement of an ordinance against its independent contractors. Thus, Castle Rock's two primary arguments fail, and we conclude that Aptive suffered an injury-in-fact.

2

In addition to its arguments that Aptive did not suffer an injury-in-fact, Castle Rock also argues that Aptive failed to demonstrate "a sufficient ‘causal connection between the injury and the conduct complained of.’ " Susan B. Anthony List , 573 U.S. at 157-58, 134 S.Ct. 2334 (quoting Lujan , 504 U.S. at 560-61, 112 S.Ct. 2130 ). This objection is based on the parties’ stipulations that the independent contractors "set their own daily schedule" but Aptive encourages them "to work between 10:00 and 11:00 a.m., take a lunch break between 2:30 and 4:00 p.m., and complete work at dark each day that they work[ ]." Aplt.’s App., Vol. II, at A221-22. Because the independent contractors are free to set their own hours, Castle Rock argues that any injury Aptive suffered was caused not by its Curfew but by the independent contractors’ decisions about when to work. See Aplt.’s Opening Br. at 15-16 ("Aptive cannot show that the Curfew—and not the individual decisions of the Contractors with respect to when they solicited—caused Aptive's alleged harm.").

"To satisfy the traceability requirement, the defendant's conduct must have caused the injury." Benham v. Ozark Materials River Rock, LLC , 885 F.3d 1267, 1273 (10th Cir. 2018) ; see Bronson , 500 F.3d at 1109-10 (explaining that Article III "require[s] proof of a substantial likelihood that the defendant's conduct caused plaintiff's injury in fact" (quoting Nova Health Sys. v. Gandy , 416 F.3d 1149, 1156 (10th Cir. 2005) )). We have already concluded above that Aptive itself has suffered an injury-in-fact because Aptive and its independent contractors are forced to limit their solicitation in accordance with the Curfew. And Aptive has demonstrated, at the very least, that there is a "substantial likelihood" that the Curfew is the reason why it and its independent contractors have not solicited until dusk in Castle Rock, i.e., the reason why it suffered the injury-in-fact. Bronson , 500 F.3d at 1109-10 (quoting Nova Health Sys. , 416 F.3d at 1156 ).

To be sure, the independent contractors set their own hours, but Aptive ordinarily would encourage them to stay out until dusk, and there is concrete evidence that the independent contractors have done so in neighboring jurisdictions that had no similar curfew. Thus, we have no difficulty concluding that Aptive's injury-in-fact is "fairly ... trace[able] to the challenged action of the defendant," i.e., Castle Rock's enactment of its curfew, "and not ... th[e] result [of] the independent action of some third party not before the court." Lujan , 504 U.S. at 560-61, 112 S.Ct. 2130 (alterations in original) (quoting Simon v. E. Ky. Welfare Rights Org. , 426 U.S. 26, 41-42, 96 S.Ct. 1917, 48 L.Ed.2d 450 (1976) ). Underscoring this point, Pacific Frontier recounted Lujan ’s causation requirement before concluding that "consequences following from the statute's enforcement" against the independent contractors were sufficient to establish standing and therefore the causation element for the employer. 414 F.3d at 1228 29. As in Pacific Frontier , we conclude that even though Aptive does not require its independent contractors to work certain hours but instead simply urges them to stay out until dusk the consequences emanating from the Curfew's enforcement against its independent contractors is sufficient under the circumstances here to establish a causal link between the Curfew's operation and Aptive's injury.

3

While uncontested by Castle Rock, we also note that the "plaintiff[’s] injury ... would be redressed by a judicial conclusion that the Ordinance[’s Curfew] is unconstitutional." Id. at 1229. We come to this conclusion because an order that enjoins Castle Rock from enforcing the Curfew would allow Aptive to solicit until dusk—its typical practice elsewhere—removing the Curfew's alleged violation of Aptive's First Amendment rights.

* * *

In sum, we conclude that Aptive has established an injury-in-fact, causation, and redressability, and thus standing.

III

Turning to the merits, we recite our standard of review, reject Castle Rock's threshold argument that the First Amendment does not apply to the Curfew at issue here, and explain why Castle Rock has failed to demonstrate that the Curfew directly and materially advances its substantial interests.

A

"In a First Amendment case, we have ‘an obligation to make an independent examination of the whole record in order to make sure that the judgment does not constitute a forbidden intrusion on the field of free expression.’ " Citizens for Peace in Space v. City of Colorado Springs , 477 F.3d 1212, 1219 (10th Cir. 2007) (quoting Bose Corp. v. Consumers Union, Inc. , 466 U.S. 485, 499, 104 S.Ct. 1949, 80 L.Ed.2d 502 (1984) ); accord Taylor v. Roswell Indep. Sch. Dist. , 713 F.3d 25, 34 (10th Cir. 2013). Thus, we have said that "[t]he factual findings, as well as the conclusions of law, are reviewed ‘without deference to the trial court.’ " Cressman v. Thompson , 798 F.3d 938, 946 (10th Cir. 2015) (quoting Citizens for Peace in Space , 477 F.3d at 1219 ).

However, the Supreme Court's and our own cases make clear that this de novo review of factual findings more precisely extends only to "crucial facts," Hurley v. Irish-Am. Gay, Lesbian & Bisexual Grp. of Bos. , 515 U.S. 557, 567, 115 S.Ct. 2338, 132 L.Ed.2d 487 (1995), or "constitutional facts," Green , 568 F.3d at 795-96. That is to say, "our searching review of the record with regard to ‘constitutional facts’ does not alter our ordinary clearly-erroneous review of the district court's other factual findings." Id. at 796 ; id. ("[T]he special Bose rule applies only to ‘constitutional facts’ and not to the basic historical facts upon which the claim is grounded, which are subject to the usual ‘clearly erroneous’ standard of review." (alteration in original) (quoting United States v. Friday , 525 F.3d 938, 950 (10th Cir. 2008) )); see also Bose Corp. , 466 U.S. at 514 n.31, 104 S.Ct. 1949 ("There are, of course, many findings of fact in a defamation case that are irrelevant to the constitutional standard of New York Times Co. v. Sullivan and to which the clearly-erroneous standard of Rule 52(a) is fully applicable.").

B

The First Amendment's Free Speech Clause provides that "Congress shall make no law ... abridging the freedom of speech." U.S. CONST. amend. I. By incorporation through the Fourteenth Amendment, this prohibition applies to states and their political subdivisions. Chaplinsky v. State of New Hampshire , 315 U.S. 568, 570-71, 62 S.Ct. 766, 86 L.Ed. 1031 (1942). It additionally "applies not only to legislative enactments, but also to less formal governmental acts, including city policies." Evans v. Sandy City , 944 F.3d 847, 852 (10th Cir. 2019) (quoting Hawkins v. City and Cty. of Denver , 170 F.3d 1281, 1286 (10th Cir. 1999), petition for cert. filed , (U.S. Mar. 5, 2020) (No. 19-1091)).

In a threshold challenge to the district court's judgment, Castle Rock argues that the Curfew at issue here does not "implicate[ ] protected speech" and thus does not violate the First Amendment's guarantee against abridgment of freedom of speech; instead, Castle Rock argues that its Curfew "simply establishes parameters on solicitors’ implied license to enter private property." Aplt.’s Opening Br. at 16. But this argument is at odds with our precedent. We explain why it is clear that the Curfew does regulate speech and thereby implicates the First Amendment, and then we reject Castle Rock's three counter-arguments.

1

Castle Rock's argument that the Curfew does not regulate speech is foreclosed by our decision in Pacific Frontier v. Pleasant Grove City, supra . There, the plaintiffs"individuals and entities engaged in selling Kirby vacuum cleaners through door-to-door solicitations"—claimed to be burdened by the locality's "ordinance establishing a solicitors licensing procedure." 414 F.3d at 1226. The ordinance "require[d] individuals to obtain a license before engaging in door-to-door solicitation." Id. The vacuum distributor declined to apply for licenses for its salespeople. Id. at 1226-27. Some of the salespeople were arrested for soliciting without a license. Id. The company sought a preliminary injunction against enforcement of the ordinance, which it alleged violated its First Amendment rights. Id. at 1230-31. The district court granted the preliminary injunction after finding that there was a substantial likelihood that the ordinance was unconstitutional. Id. On appeal, we applied the First Amendment test derived from Central Hudson Gas & Electric Corporation. v. Public Service Commission , 447 U.S. 557, 100 S.Ct. 2343, 65 L.Ed.2d 341 (1980), to evaluate the "First Amendment challenge" to the regulation. Pac. Frontier , 414 F.3d at 1231. In doing so, we noted that "[t]he Supreme Court has recognized that personal solicitation is imbued with important First Amendment interests." Id. at 1231 n.8 (citing Edenfield v. Fane , 507 U.S. 761, 766, 113 S.Ct. 1792, 123 L.Ed.2d 543 (1993) ). And, after applying the test, we concluded that there was a substantial likelihood that the ordinance violated the First Amendment and affirmed the preliminary injunction. Id. at 1235, 1238. Pacific Frontier and this case are alike in all the ways that matter here: the localities in both cases enacted ordinances restricting door-to-door commercial solicitation. And in Pacific Frontier we clearly held that the First Amendment was implicated by the ordinance at issue there; after all, we upheld the district court's decision to preliminarily enjoin aspects of the ordinance for being substantially likely to contravene the plaintiffs’ First Amendment rights. That holding has strong cogent force here and establishes that Castle Rock's Curfew on commercial door-to-door solicitation likewise implicates the First Amendment.

Were Pacific Frontier not enough, a multitude of other cases from the Supreme Court and other courts would militate in favor of our conclusion that Castle Rock's Curfew relating to door-to-door commercial solicitation implicates the First Amendment. See Ohio Citizen Action v. City of Englewood , 671 F.3d 564, 570-80 (6th Cir. 2012) (holding that 6:00 p.m. curfew on door-to-door solicitation violated First Amendment); N.J. Citizen Action v. Edison Township, 797 F.2d 1250, 1254, 1262 (3d Cir. 1986) (holding that 5:00 p.m., 6:00 p.m., and sunset curfews on door-to-door solicitation violated First Amendment); City of Watseka v. Ill. Pub. Action Council , 796 F.2d 1547, 1548 (7th Cir. 1986) (holding that 5:00 p.m. curfew on door-to-door solicitation violated First Amendment), aff'd , 479 U.S. 1048, 107 S.Ct. 919, 93 L.Ed.2d 972 (1987) ; Ass'n of Cmty. Orgs. for Reform Now v. City of Frontenac , 714 F.2d 813, 815 (8th Cir. 1983) (holding that 6:00 p.m. curfew on door-to-door solicitation violated the First Amendment); cf. Edenfield , 507 U.S. at 764-65, 113 S.Ct. 1792 (holding that rule banning "direct, in-person, uninvited solicitation" from certified public accountants violated the First Amendment and stating that "it is clear that this type of personal solicitation is commercial expression to which the protections of the First Amendment apply" (citation omitted)). We thus conclude that Castle Rock's Curfew implicates the First Amendment by regulating protected commercial speech.

2

Castle Rock tries to distinguish this body of authority in three ways. None is convincing.

a

First, Castle Rock attempts to distinguish these cases as "involv[ing] non commercial speech restrictions." Aplt.’s Reply Br. at 15-16 (emphasis added); see Aplt.’s Opening Br. at 19-20. But Pacific Frontier did involve commercial speech, see 414 F.3d at 1226, and so this proffered distinction falters at the starting blocks. Furthermore, while not involving commercial door-to-door solicitation, the Supreme Court has clearly held that in-person commercial solicitation is generally protected by the First Amendment. See Edenfield , 507 U.S. at 765-66, 113 S.Ct. 1792 ("In the commercial context, solicitation may have considerable value. Unlike many other forms of commercial expression, solicitation allows direct and spontaneous communication between buyer and seller."). In effect, "the city's argument attaches more importance to the distinction between commercial and noncommercial speech than [the Supreme Court's] cases warrant and seriously underestimates the value of commercial speech." City of Cincinnati v. Discovery Network, Inc. , 507 U.S. 410, 419, 113 S.Ct. 1505, 123 L.Ed.2d 99 (1993). While the Supreme Court has indicated that commercial speech is entitled to "lesser protection" than noncommercial speech, Cent. Hudson , 447 U.S. at 562-63, 100 S.Ct. 2343, this most certainly does not mean that commercial speech is entitled to no protection, see, e.g. , Discovery Network , 507 U.S. at 420-21, 113 S.Ct. 1505 ("Speech likewise is protected ... even though it may involve a solicitation to purchase or otherwise pay or contribute money." (citations omitted)); cf. Pac. Frontier , 414 F.3d at 1236 ("Commercial speech merits First Amendment protection not simply because it enables sellers to hawk their wares and gain a profit, but because it equips consumers with valuable information and because it contributes to the efficiency of a market economy."). The commercial nature of the speech is relevant to our analysis, but it does not remove the Curfew from First Amendment scrutiny.

Castle Rock relatedly argues that Aptive's challenge to its Curfew warrants only "rational basis review"—that is, review under a standard foreign to restrictions on First Amendment protected speech. Aplt.’s Opening Br. at 20; see id. at 20-21 ("While all commercial transactions involve the communication of some message, regulating the places and manner in which these activities occur does not implicate the First Amendment. ... [T]he Curfew does not regulate commercial actors’ speech; it regulates the act of stepping onto private property without an invitation." (citations omitted)). But the cases that Castle Rock cites either undermine this assertion or are inapposite.

The first, Expressions Hair Design v. Schneiderman , ––– U.S. ––––, 137 S. Ct. 1144, 197 L.Ed.2d 442 (2017), held that the law at issue did "regulate[ ] speech" under the First Amendment before remanding for the Second Circuit to analyze the law as a restriction on commercial speech under the same Central Hudson test we apply here. Expressions Hair Design , 137 S. Ct. at 1151. The second, Cash Inn of Dade, Inc. v. Metropolitan Dade County , 938 F.2d 1239 (11th Cir. 1991), reviewed an ordinance limiting pawnshop hours of operation under the rational-basis standard but, critically for our purposes, the ordinance had not been challenged under the First Amendment. Id. at 1241 (challenging the ordinance under Williamson v. Lee Optical of Oklahoma Inc. , 348 U.S. 483, 75 S.Ct. 461, 99 L.Ed. 563 (1955) ’s general due-process standard). The case thus is irrelevant to the First Amendment challenge before us. And the third, Wine and Spirits Retailers, Inc. v. Rhode Island , 418 F.3d 36 (1st Cir. 2005), did reject a First Amendment challenge to a law regulating the provision of business advice and advertising services. Id. at 47-50. But that case acknowledged that "restrictions imped[ing] [a party's] right to communicate with its potential customers" would be within the heart of the commercial-speech doctrine; such restrictions simply were not at issue there. Id. at 49. Thus, none of the cases that Castle Rock relies on provides any reason to question the authority squarely holding that commercial door-to-door solicitation is subject to First Amendment scrutiny.

b

Castle Rock's second proffered distinction is that the above cases that applied First Amendment scrutiny to door-to-door solicitation "involved regulations in which violations were premised upon the actual words used by a speaker," Aplt.’s Reply Br. at 15 16, i.e., they were "content based," Aplt.’s Opening Br. at 19. Castle Rock argues that, unlike those cases, its Curfew "regulat[es] the places and manner in which [commercial] activities occur," and thus "does not implicate the First Amendment." Aplt.’s Opening Br. at 20-21. Castle Rock thus appears to argue that its Curfew is immune from First Amendment scrutiny because the Curfew is a mere restriction on the time, place, and manner of door-to-door solicitation.

This attempted distinction fails first and foremost because even content-neutral ordinances e.g., ordinances that impose time, place, and manner restrictions remain subject to First Amendment scrutiny; the scrutiny is simply less rigorous than that traditionally attendant to regulations that are based on the content of the speech. See, e.g. , Ward v. Rock Against Racism , 491 U.S. 781, 790-91, 109 S.Ct. 2746, 105 L.Ed.2d 661 (1989) (analyzing a "reasonable restriction[ ] on the time, place, or manner of protected speech" as "subject to the protections of the First Amendment"); Evans , 944 F.3d at 856 ("Even though the [Ordinance] is content neutral, it still must be ‘narrowly tailored to serve a significant governmental interest.’ " (alteration in original) (quoting McCullen v. Coakley , 573 U.S. 464, 486, 134 S.Ct. 2518, 189 L.Ed.2d 502 (2014) )). Thus, even were the Curfew at issue here truly content-neutral, it still would not escape First Amendment scrutiny.

Furthermore, this argument additionally fails because the Curfew is content-based, at least insofar as the 2014 ordinance determines to whom the Curfew applies by distinguishing between the commercial and noncommercial content of the solicitors’ speech. Our conclusion in this regard is dictated by the Supreme Court's decision in Discovery Network . There, the Court struck down a law that banned commercial but not noncommercial newsracks. See 507 U.S. at 412-13, 113 S.Ct. 1505. In so doing, the Court concluded that the regulation was not content-neutral because "the very basis for the regulation is the difference in content between ordinary newspapers and commercial speech." Id. at 428-29, 113 S.Ct. 1505. Because the 2014 ordinance (like the 2008 ordinance before it) facially makes the application of its Curfew turn on whether the speech is commercial or not, the law is content-based. See id. at 429, 113 S.Ct. 1505.

Specifically, the ordinance treats civic, religious, philosophical, and ideological solicitors who incidentally sell a good or service differently from those who solicit with the "primary purpose" of selling a good or service. Compare Aplt.’s App., Vol. V, at A1296 (defining noncommercial solicitors, inter alia , as those "[a]ttempting to obtain a donation to a particular patriotic, philanthropic, social service, welfare, benevolent, educational, civic, fraternal, charitable, political or religious purpose, even if incidental to such purpose there is the sale of some good or service"), with id. (defining commercial solicitors as those who attempt to make personal contact at a resident's home "for the primary purpose of attempting to sell a good or service"). When an ordinance makes these sorts of facial distinctions, e.g., between those soliciting for religious purposes and those soliciting for commercial gain, not only the Supreme Court, but our court, has expressly held that it "contemplates a distinction based on content." Ass'n of Cmty. Orgs. for Reform Now v. Municipality of Golden , 744 F.2d 739, 749 (10th Cir. 1984) ; see id. at 750 ("[W]hether solicitation is for charitable, religious, patriotic or philanthropic purposes, or otherwise provides a service necessary to the general welfare of [the municipality]’s residents, clearly turns on the content of the message."). And so, because the 2014 ordinance creates a content-based distinction—which determines to which solicitors the Curfew applies—between commercial and noncommercial speech, we must reject any argument that the Curfew is either not subject to First Amendment scrutiny at all or can be analyzed merely as a content-neutral time, place, and manner restriction.

c

Third and finally, Castle Rock argues that its Curfew merely governs trespass and thus does not implicate the First Amendment. Specifically, it argues that "[w]hile solicitors may enjoy a license, the conduct of entering uninvited upon private property used for private purposes is not itself protected speech." Aplt.’s Opening Br. at 17 (citing Lloyd Corp. v. Tanner , 407 U.S. 551, 568, 92 S.Ct. 2219, 33 L.Ed.2d 131 (1972) ). But we rejected an analogous argument in Western Watersheds Project v. Michael , 869 F.3d 1189 (10th Cir. 2017). At issue in that case were statutes that imposed penalties on anyone who crossed private land to access adjacent or proximate land where that person would collect data about public lands. Id. at 1191. We said that "[t]he fact that one aspect of the challenged statutes concerns private property does not defeat the need for First Amendment scrutiny." Id. at 1195. While we acknowledged that "trespassing does not enjoy First Amendment protection," id. at 1192, we held "that the statutes regulate protected speech under the First Amendment and that they are not shielded from constitutional scrutiny merely because they touch upon access to private property," id. We were able to conclude that the statutes regulated protected speech because they "target[ed] the ‘creation’ of speech by imposing heightened penalties on those who collect resource data" about public lands. Id. (quoting Sorrell v. IMS Health Inc. , 564 U.S. 552, 570, 131 S.Ct. 2653, 180 L.Ed.2d 544 (2011) ).

In much the same way, the regulation at issue here also targets the creation of speech that the Supreme Court has long categorized as protected by the First Amendment. See Watchtower Bible & Tract Soc'y of N.Y., Inc. v. Vill. of Stratton , 536 U.S. 150, 160, 122 S.Ct. 2080, 153 L.Ed.2d 205 (2002) ("For over 50 years, the Court has invalidated restrictions on door-to-door canvassing and pamphleteering."); Ill. Pub. Action Council , 796 F.2d at 1550 ("The Supreme Court has recognized substantial First Amendment protection for door-to-door solicitors."). And, while we agree with the general and uncontroversial principle that private parties may choose what speech to allow on their property without violating the First Amendment, see, e.g. , Tanner , 407 U.S. at 552, 570, 92 S.Ct. 2219 (holding that a private shopping center did not violate the First Amendment by prohibiting the distribution of handbills on its property), Castle Rock fails to explain how that general principle interacts with the above cases squarely holding that First Amendment interests are implicated when municipalities attempt to restrict door-to-door solicitation. See, e.g. , Martin v. City of Struthers , 319 U.S. 141, 141, 63 S.Ct. 862, 87 L.Ed. 1313 (1943) ("For centuries it has been a common practice in this and other countries for persons not specifically invited to go from home to home and knock on doors or ring doorbells to communicate ideas to the occupants or to invite them to political, religious, or other kinds of public meetings. Whether such visiting shall be permitted has in general been deemed to depend upon the will of the individual master of each household, and not upon the determination of the community."); Ill. Pub. Action Council , 796 F.2d at 1556, 1558 (holding unconstitutional "5 p.m. to 9 p.m. ban [that was] essentially an attempt by [the city] to substitute its judgment for that of its citizens").

Moreover, we are unpersuaded by the cases that Castle Rock cites in support of its argument that its Curfew "does not regulate speech" but rather "only uninvited access onto private property." Aplt.’s Opening Br. at 12. For example, it is true that the Supreme Court in Cornelius v. NAACP Legal Defense and Educational Fund, Inc. , 473 U.S. 788, 105 S.Ct. 3439, 87 L.Ed.2d 567 (1985), generally observed that "a speaker must seek access to public property or to private property dedicated to public use to evoke First Amendment concerns." Id. at 801, 105 S.Ct. 3439. However, Cornelius is inapposite. Its general observation concerning what is necessary to invoke First Amendment protections was not made in the context of door-to-door solicitations, and the Court consequently had no occasion to discuss—much less undercut—its longstanding view that door-to-door solicitation is speech and is entitled to First Amendment protection. See, e.g. , Watchtower , 536 U.S. at 160, 122 S.Ct. 2080 ; Ill. Pub. Action Council , 796 F.2d at 1550.

Relatedly, there was never any question in Cornelius that private property was not involved there; the federal charitable program at issue—that plaintiffs wanted to access in order to solicit contributions—was a creature of the federal government and regulated by the federal government. See 473 U.S. at 795, 105 S.Ct. 3439 (noting that the program "was designed to lessen the Government's burden in meeting human health and welfare needs by providing a convenient, nondisruptive channel for federal employees to contribute to nonpartisan agencies that directly serve those needs"). Thus, the Court did not have occasion to address any argument remotely resembling the one that Castle Rock makes here, which turns on the government's authority to regulate uninvited access to private property. Cf. Ill. Pub. Action Council , 796 F.2d at 1552 n.12 ("[The city] spent most of its oral argument contending that this case is controlled by the public forum/private forum standards in Cornelius [ ]. This argument is meritless. The Supreme Court's analysis of government property turns upon the public forum/private forum distinction, but nothing in Cornelius suggests the Court intended to extend this distinction to time, place, and manner restrictions on private property." (citation omitted)). Indeed, the government argued unsuccessfully in Cornelius that the plaintiffs’ speech fell outside the bounds of the First Amendment—not because the charitable program was private, but rather because ostensibly "a First Amendment forum necessarily consists of tangible government property." 473 U.S. at 800, 105 S.Ct. 3439 ; see Koala v. Khosla , 931 F.3d 887, 901 (9th Cir. 2019) (noting that in Cornelius "[t]he government argued that the fundraising program was not a forum at all, because it was not a physical space"). Thus, we conclude that Cornelius is inapposite and does not provide a basis for departing from the controlling precedent explicated supra concerning the First Amendment protections accorded to door-to-door solicitors.

Castle Rock also cites Frisby v. Schultz , 487 U.S. 474, 108 S.Ct. 2495, 101 L.Ed.2d 420 (1988), where the Supreme Court stated that it has "repeatedly held that individuals are not required to welcome unwanted speech into their own homes and that the government may protect this freedom." Id. at 485, 108 S.Ct. 2495. It then upheld an ordinance that banned picketing outside residences. Id. at 476, 488, 108 S.Ct. 2495. However, while the Court upheld the ordinance, the Court clearly held that the picketing at issue was protected speech. See id. at 479, 108 S.Ct. 2495 ("The antipicketing ordinance operates at the core of the First Amendment by prohibiting appellees from engaging in picketing on an issue of public concern."). An ordinance can both implicate the First Amendment and ultimately survive the scrutiny mandated by its impingement on First Amendment rights. Therefore, the fact that the government can permissibly impose certain restrictions on the uninvited exercise of First Amendment rights on private property does not mean as Castle Rock suggests that those First Amendment rights somehow do not exist on private property. Castle Rock's citation to Frisby thus does not advance the argument it makes here. And other cases that Castle Rock cites run into the same problem. See, e.g. , Vill. of Schaumburg v. Citizens for a Better Env't , 444 U.S. 620, 633, 100 S.Ct. 826, 63 L.Ed.2d 73 (1980) ("The issue before us, then, is not whether charitable solicitations in residential neighborhoods are within the protections of the First Amendment. It is clear that they are."). None of these cases provides support for the idea that door-to-door solicitation does not implicate the First Amendment. For the first time in reply, Castle Rock has argued that the Curfew was analogous to zoning regulations because "it simply prohibits uninvited access by any commercial actor." Aplt.’s Reply Br. at 15-16. At the outset, we note that this late-blooming argument—appearing for the first time in Castle Rock's reply brief—is waived; therefore, we need not consider it. See, e.g. , In re: Motor Fuel Temperature Sales Practices Litig. , 872 F.3d 1094, 1112 n.5 (10th Cir. 2017) (noting that "arguments raised for the first time in a reply brief are waived"); Star Fuel Marts, LLC v. Sam's E., Inc. , 362 F.3d 639, 647 (10th Cir. 2004) ("Generally, arguments raised for the first time on appeal in an appellant's reply brief are waived."). However, even if we were to consider it, we would not be persuaded. In particular, neither of the cases Castle Rock cites for support are helpful to the argument that it makes here. It specifically contends that "under Aptive's theory, the front porch of every home in America would be transformed into a public forum." Aplt.’s Reply Br. at 15. But Castle Rock relies on Cornelius , 473 U.S. 788, 105 S.Ct. 3439, which we determined supra is inapposite. Castle Rock also argues that under Aptive's theory "any regulation of commercial activity, such as zoning laws prohibiting commercial activities in residential areas, would be subjected to intermediate scrutiny analysis." Id. The only case cited in support of this proposition is Expressions Hair Design , 137 S. Ct. at 1151. But, again, that case held that the law at issue, which regulated how stores communicated the prices of products, "regulates speech." Id. That case thus does not support Castle Rock's argument that laws regulating door-to-door solicitation escape First Amendment scrutiny. In sum, none of Castle Rock's arguments discussed here that essentially posit that the Curfew merely regulates trespass on private property give us any reason to second-guess the ample controlling, as well as persuasive, authority discussed supra holding that regulation of door-to-door solicitation implicates the First Amendment.

* * *

In sum, we conclude that Castle Rock's 2014 ordinance—and, more specifically, the Curfew—regulates commercial speech protected by the First Amendment, and we reject Castle Rock's arguments to the contrary.

C

Our prior cases and the parties agree that—assuming that the Curfew implicates the First Amendment, as we have just decided—our analysis is governed by Central Hudson Gas & Electric Corporation v. Public Service Commission , supra . We set out the substantive legal standards from Central Hudson and then address in detail whether Castle Rock's Curfew advances its substantial interests in a direct and material way— more specifically, whether "the harms [Castle Rock] recites are real and that its restriction will in fact alleviate them to a material degree." Edenfield , 507 U.S. at 771, 113 S.Ct. 1792. We conclude that Castle Rock has failed to carry its burden under Central Hudson .

1

The test enunciated in Central Hudson is a form of "intermediate standard of review," and, as well summarized in Edenfield , it provides that in determining whether commercial speech may be proscribed,

we must ask [1] whether the State's interests in proscribing it are substantial, [2] whether the challenged regulation advances these interests in a direct and material way, and [3] whether the extent of the restriction on protected speech is in reasonable proportion to the interests served.

Edenfield , 507 U.S. at 767, 113 S.Ct. 1792 ; accord Cent. Hudson , 447 U.S. at 566, 100 S.Ct. 2343 ; Pac. Frontier , 414 F.3d at 1231-32. Castle Rock bears the burden on all three prongs. See Edenfield , 507 U.S. at 770, 113 S.Ct. 1792 ("It is well established that ‘[t]he party seeking to uphold a restriction on commercial speech carries the burden of justifying it.’ " (quoting Bolger v. Youngs Drug Prods. Corp. , 463 U.S. 60, 71 n.20, 103 S.Ct. 2875, 77 L.Ed.2d 469 (1983) )); Pac. Frontier , 414 F.3d at 1231-35 ("A municipality has the burden of justifying its regulation even on a motion to enjoin enforcement of an ordinance."). We summarize each element, focusing on the second, which we find determinative.

First, "a municipality must assert ‘a substantial interest to be achieved by restrictions on commercial speech.’ " Id. at 1231 (quoting Cent. Hudson , 447 U.S. at 564, 100 S.Ct. 2343 ). "[W]e must identify with care the interests the State itself asserts. Unlike rational-basis review, the Central Hudson standard does not permit us to supplant the precise interests put forward by the State with other suppositions." Edenfield , 507 U.S. at 768, 113 S.Ct. 1792. Castle Rock asserts that the Curfew furthers its residents’ interests in privacy and safety from crime, and Aptive does not contest that these interests are substantial. Cf. Pac. Frontier , 414 F.3d at 1232 n.10 ("Their concession is appropriate given longstanding Supreme Court precedent recognizing a municipality's right to protect its residents’ peaceful enjoyment of their homes and to prevent crime."). Thus, we do not linger on this factor.

Second, "[t]hat [the government's] asserted interests are substantial in the abstract does not mean, however," that its restriction on commercial speech purporting to effectuate those interests is permissible under Central Hudson . Edenfield , 507 U.S. at 770, 113 S.Ct. 1792. "[T]he restriction must directly advance th[e] substantial interest[s]" asserted by the municipality. Pac. Frontier , 414 F.3d at 1231 ; accord Revo v. Disciplinary Bd. of the Supreme Court of N.M. , 106 F.3d 929, 933 (10th Cir. 1997) ("The second element asks whether the ban directly and materially advances the asserted state interest."). This means that "[i]f the regulation ‘provides only ineffective or remote support for the government's purpose,’ it will not be upheld." Pac. Frontier , 414 F.3d at 1231 (quoting Cent. Hudson , 447 U.S. at 564, 100 S.Ct. 2343 ); accord Mainstream Mktg. , 358 F.3d at 1237 ("[T]he regulation must directly advance that governmental interest, meaning that it must do more than provide ‘only ineffective or remote support for the government's purpose.’ " (quoting Cent. Hudson , 447 U.S. at 564, 100 S.Ct. 2343 )). "This burden is not satisfied by mere speculation or conjecture; rather, a governmental body seeking to sustain a restriction on commercial speech must demonstrate that the harms it recites are real and that its restriction will in fact alleviate them to a material degree." Edenfield , 507 U.S. at 770-71, 113 S.Ct. 1792 ; accord Utah Licensed Beverage Ass'n v. Leavitt , 256 F.3d 1061, 1071 (10th Cir. 2001). This requirement is "critical; otherwise, ‘a [governmental body] could with ease restrict commercial speech in the service of other objectives that could not themselves justify a burden on commercial expression.’ " Rubin v. Coors Brewing Co. , 514 U.S. 476, 487, 115 S.Ct. 1585, 131 L.Ed.2d 532 (1995) (quoting Edenfield , 507 U.S. at 771, 113 S.Ct. 1792 ). The Supreme Court does "not, however, require that ‘empirical data come ... accompanied by a surfeit of background information.’ " Lorillard Tobacco Co. v. Reilly , 533 U.S. 525, 555, 121 S.Ct. 2404, 150 L.Ed.2d 532 (2001) (quoting Fla. Bar v. Went For It, Inc. , 515 U.S. 618, 628, 115 S.Ct. 2371, 132 L.Ed.2d 541 (1995) ). Instead, the Court "ha[s] permitted litigants to justify speech restrictions by reference to studies and anecdotes pertaining to different locales altogether, or even, in a case applying strict scrutiny, to justify restrictions based solely on history, consensus, and ‘simple common sense.’ " Id. (quoting Went For It , 515 U.S. at 628, 115 S.Ct. 2371 ). When evaluating whether a municipality has put forward sufficient anecdotes, history, or common sense to demonstrate "that the harms it recites are real and that its restriction will in fact alleviate them to a material degree," Edenfield , 507 U.S. at 771, 113 S.Ct. 1792, we evaluate the evidence in light of the cases where those categories of evidence have previously been invoked, see Pac. Frontier , 414 F.3d at 1235 n.12 ("What constituted sufficient anecdotal evidence in [ Went For It ] provides helpful guidance as we evaluate [the municipality]’s [anecdotal and common-sense] evidence.").

As for the third element of the Central Hudson test, "the regulation is unconstitutional ‘if the governmental interest could be served as well by a more limited restriction on commercial speech.’ " Id. at 1231-32 (quoting Cent. Hudson , 447 U.S. at 564, 100 S.Ct. 2343 ); Mainstream Mktg. , 358 F.3d at 1238 (asking "if the government did not suppress an excessive amount of speech when substantially narrower restrictions would have worked just as well"). "[L]aws restricting commercial speech, unlike laws burdening other forms of protected expression, need only be tailored in a reasonable manner to serve a substantial state interest in order to survive First Amendment scrutiny." Edenfield , 507 U.S. at 767, 113 S.Ct. 1792 ; see Bd. of Trs. of State Univ. v. Fox , 492 U.S. 469, 477, 109 S.Ct. 3028, 106 L.Ed.2d 388 (1989) (requiring "something short of a least-restrictive-means standard"). However, because we conclude that Castle Rock fails at the second step in demonstrating that its Curfew "advances [its substantial] interests in a direct and material way," Edenfield , 507 U.S. at 767, 113 S.Ct. 1792, we do not reach this third element.

2

We now apply Central Hudson ’s second prong, i.e., whether Castle Rock's Curfew "directly advance[s]" in a material way its substantial interests. Pac. Frontier , 414 F.3d at 1231. Castle Rock asserts that the Curfew "directly advances" its interests in public safety and privacy. We conclude that for both asserted interests—Castle Rock has failed to carry its burden of demonstrating "that the harms it recites are real and that its restriction will in fact alleviate them to a material degree." Edenfield , 507 U.S. at 770-71. We thus conclude that the Curfew unconstitutionally burdens Aptive's First Amendment rights.

a

Castle Rock first asserts that the Curfew "[d]irectly [a]dvances" its interests in public safety. Aplt.’s Opening Br. at 26. While all agree this interest is "substantial in the abstract," Edenfield , 507 U.S. at 770, 113 S.Ct. 1792 ; see Pac. Frontier , 414 F.3d at 1232 n.10 (noting "longstanding Supreme Court precedent recognizing a municipality's right ... to prevent crime"), we conclude that Castle Rock has failed to carry its burden of demonstrating "that the harms it recites are real and that its restriction will in fact alleviate them to a material degree," Edenfield , 507 U.S. at 770-71, 113 S.Ct. 1792.

Castle Rock seeks to demonstrate the existence of a "real" threat to public safety through both data and the testimony of various council members and town employees. We start with the data. Castle Rock plainly concedes that it "retains no data predating the Curfew's enactment that correlates evening-hours solicitation with serious crime issues or rampant invasions of privacy." Aplt.’s Opening Br. at 24; see Aplt.’s App., Vol. III, at A639 (stipulating that "[t]he Town has no record of any crimes committed by registered commercial solicitors in Castle Rock prior to the passage of the 2008 Ordinance"); id. at A643 (stipulating that "[t]he Town has no record of ever charging or convicting a registered commercial solicitor for committing a crime to persons or property"). The record does contain evidence that the police department received eight "reports" concerning door-to-door solicitation and twenty to thirty more informal "complaints" concerning solicitation, some of which concerned door-to-door solicitation, in the year preceding the Curfew's enactment. Id. , Vol. V, at A1313. One of these complaints evidently involved a solicitor who was going door-to-door at 9:45 p.m. Id. at A1327. But Castle Rock is right to acknowledge that these complaints do not "correlate[ ] evening-hours solicitation with serious crime issues." Aplt.’s Opening Br. at 24. Indeed, only the one complaint concerning the 9:45 p.m. conduct related to evening hours solicitation and was thus relevant to the Curfew. But even the existence of this complaint does not demonstrate that any solicitors (let alone commercial solicitors) specifically threatened public safety, as the record provides no detail about the complaint and whether it was public-safety related. The record evidence thus does not demonstrate that a "concrete, nonspeculative harm" to public safety existed before the Curfew's enactment. Went For It , 515 U.S. at 628 29, 115 S.Ct. 2371.

The only remaining data are from 2016 and 2017, i.e., after the Curfew was enacted. These data demonstrate that in 2016 and 2017 combined there were only eleven complaints about doorbell interruptions after 7:00 p.m. and that none of those complaints involved commercial solicitors. All eleven of the complaints about interruptions after 7:00 p.m. concerned noncommercial solicitors. Under this evidence, the Curfew is not crafted well to materially alleviate Castle Rock's posited threat to public safety because those causing the evening interruptions are not covered by the Curfew noncommercial solicitors. Cf. Pac. Frontier , 414 F.3d at 1234-35 (noting that "[a]ny speculation that [the ordinance] would assist with crime prevention is further undercut by evidence that residential burglaries committed in [the city] by those posing as solicitors involved individuals who did not apply for a license" and so "the [restrictions imposed by the ordinance] would have been of no use in those investigations"). Furthermore, were we to look at all of the complaints across both 2016 and 2017, i.e., not just those occurring after 7:00 p.m., there still would be only one complaint about a registered commercial solicitor. And even this complaint tells us nothing about a purported threat to public safety . The data thus provide no support for Castle Rock's argument that commercial solicitation after 7:00 p.m. has presented a threat to public safety—either before or after the Curfew's enactment.

Castle Rock counters that the absence of post-enactment data showing that commercial solicitors are a threat to public safety demonstrates that the Curfew "is working." Aplt.’s Opening Br. at 24 (emphasis omitted). But this unadorned assertion is self-defeating. It is Castle Rock's burden to demonstrate that the Curfew directly advances its interest in public safety, Pac. Frontier , 414 F.3d at 1231, and, at least under these circumstances, the absence of data cannot carry that burden.

As for testimonial evidence, the majority of Castle Rock's remaining evidence of commercial solicitors’ purported threat to public safety comes from testimonial evidence from various council members and town employees pertaining to the post-enactment time period. This testimonial evidence—which purports to demonstrate that commercial door-to-door solicitation after 7:00 p.m. remains a threat to public safety—is thoroughly eviscerated by the data above to the contrary. Indeed, Castle Rock's current police chief was unable to point to a specific instance where a commercial solicitor was accused of any solicitation-related crime or where an individual had posed as a door-to-door solicitor for purposes of hiding criminal intent. He also stated that "[m]ost crimes occur outside of the nighttime hours," Aplt.’s App., Vol. V, at A1132, A1150—which not only directly undercuts the public-safety rationale for Castle Rock's Curfew but also is at odds with one of the Castle Rock town council's express prefatory statements in justifying the Curfew's original enactment—viz. , "criminal activity on private property often occurs during nighttime hours," id. , Vol. VI, at A1445. Indeed, the chief testified that "a person is probably not more likely to commit a crime to persons or property arising out of door-to-door solicitation after 7:00 p.m. than before 7:00 p.m.," and that "moving the Town's Curfew from 7:00 p.m. to dusk would not materially hamper the Castle Rock Police Department's ability to protect its citizens from crime." Id. , Vol. II, at A235. Additionally, he testified that the data suggest that Castle Rock's problem is with unregistered solicitors, not registered commercial solicitors like Aptive's. In sum, all of this testimony from Castle Rock's own police chief "contradicts, rather than strengthens," the proffered public safety interest in the Curfew. Edenfield , 507 U.S. at 772, 113 S.Ct. 1792.

The only additional pre-enactment indicators of the ostensible threat to public safety from commercial door-to-door solicitation come from the 2008 ordinance's prefatory clauses (which were incorporated into the 2014 ordinance) and the testimony of a former mayor and town council member. But, as our discussion of the police chief's testimony firmly underscores, there was no factual basis for the various prefatory statements to the ordinance, which purported to connect Castle Rock's public-safety interest to its Curfew restriction on commercial door-to-door solicitation. And Castle Rock's former mayor testified that "prior to passing the 2008 Ordinance[,] there was no discussion or analysis by the Town Council of ... solicitation-related crime in Castle Rock," "crime committed by commercial solicitors in Castle Rock," "crime committed by commercial solicitors in Castle Rock after 7:00 p.m.," or "how a 7:00 p.m. curfew would protect public safety." Aplt.’s App., Vol. II, at A228. The prefatory clauses thus cannot serve as evidence of a real harm. See Edenfield , 507 U.S. at 770-71, 113 S.Ct. 1792 ("This burden is not satisfied by mere speculation or conjecture; rather, a governmental body seeking to sustain a restriction on commercial speech must demonstrate that the harms it recites are real ...." (emphases added)).

To be sure, Castle Rock's former mayor stated that "when [the] Council was considering the Curfew, there ‘probably was [sic] some thoughts,’ that ‘reasonable people would think that people walking around their neighborhood or up to their home could potentially be somebody that might create a crime in the town.’ " Aplt.’s App., Vol. II, at A234. And a former council member echoed these concerns, stating that he "definitely" thought that it "could be a possibility " that individuals posing as door-to-door solicitors were engaging in criminal activity. Id. , Vol. V, at A1085-86 (emphasis added). But these sorts of equivocal and hypothetical statements about the possibility of a connection between crime and commercial door-to-door solicitation are insufficient to establish a "concrete, nonspeculative harm" to public safety. Went For It , 515 U.S. at 628-29, 115 S.Ct. 2371 ; cf. Ill. Pub. Action Council , 796 F.2d at 1555 n.15 ("When a city ... wants to pass an ordinance that will substantially limit First Amendment rights, the city must produce more than a few conclusory affidavits of city leaders which primarily contain unsubstantiated opinions and allegations.").

The insufficiency of Castle Rock's evidence purporting to establish that the Curfew "directly advances" Castle Rock's interest in public safety is underscored when we consider the analysis in Pacific Frontier . There, the city sought to justify a fingerprint requirement for commercial solicitors "by arguing that it further[ed] the city's legitimate interests in assuring peaceful use of private property and in protecting its citizens against crime." 414 F.3d at 1233. More specifically, the city argued that its fingerprint requirement "uniquely further[ed] [the city's] interest in crime prevention and investigation." Id. at 1234. A police officer and city attorney, respectively, testified that the fingerprinting requirement allowed the city to "potentially" identify culprits and that the requirement deterred crime. Id. But we determined that the city "provided no evidence other than conjecture to support its argument that having solicitors’ fingerprints on file would either deter crime or aid the investigation of a burglary." Id. at 1235.

In arguing to the contrary, the city said that its evidence was the kind of "anecdotes and common sense" that the Supreme Court had permitted to "justify" restrictions concerning First Amendment speech like its ordinance. Id. at 1235 n.12. We acknowledged that the Court in Went For It had relied on "anecdotal evidence" to uphold a lawyer "solicitation regulation" when applying the Central Hudson test. Id. But we also noted that "[w]hat constituted sufficient anecdotal evidence" in Went For It could provide "helpful guidance" in assessing the adequacy of the anecdotal and common-sense evidence that the city advanced. Id. We noted that, in Went For It , the defendant "Florida Bar conducted a two-year study of the impact of lawyer advertising and solicitation, and compiled a 106-page summary of its findings including survey results, newspaper editorials, and complaints filed by citizens." Id. Considering the Went For It guidepost, we concluded that the "anecdotes" and "common sense" that the city presented to justify its fingerprint requirement were "wholly insufficient." Id. In short, the city's evidence did not demonstrate "that the harms it recite[d] we[re] real," Edenfield , 507 U.S. at 770-71, 113 S.Ct. 1792.

To be clear, we do not read Pacific Frontier ’s focus on the evidentiary showing in Went For It as establishing a floor or threshold showing for the quantity or quality of the common-sense or anecdotal evidence that a municipality must necessarily marshal in demonstrating that the harms that support its restriction on commercial speech are based in reality. Pacific Frontier used the showing in Went For It only as "helpful guidance"—a useful example of a circumstance where the government's common-sense and anecdotal showing was held to be sufficient . 414 F.3d at 1235 n.12. We had no occasion in Pacific Frontier to opine on how much or what kind of common-sense or anecdotal evidence would be necessary to satisfy the reality-based "standard elaborated in Edenfield ." Id. (quoting Went For It , 515 U.S. at 628, 115 S.Ct. 2371 ); see also United States v. Burkholder , 816 F.3d 607, 620 n.10 (10th Cir. 2016) ("An event or condition is sufficient if its existence means that another event or condition will occur. An event or condition is necessary if, in its absence, another event or condition could not occur."). We are in a like situation here. Castle Rock's anecdotal and common-sense showing is woefully insufficient, when viewed through the "helpful" prism of the showing in Went For It . Pac. Frontier , 414 F.3d at 1235 n.12. In support of its public-safety justification, Castle Rock has provided us no studies, no supportive evidence-based findings, and no survey results. And, with respect to citizen complaints, there is no evidence that commercial solicitors are the source of any public-safety problems. Indeed, what common-sense and anecdotal evidence that Castle Rock presents is contradicted by the police chief's testimony that there was no evidence that commercial solicitation posed a threat to public safety and the accompanying data demonstrating that there have not been any complaints about commercial solicitation after 7:00 p.m. In short, using as we did in Pacific Frontier the "helpful guidance" of Went For It , id. , we are able to conclude that the common-sense and anecdotal evidence that Castle Rock has advanced in support of its interest in public safety is woefully insufficient.

Castle Rock resists, but presents no persuasive argument to the contrary. Castle Rock first argues that this result "would eviscerate a municipality's ability to address common sense concerns of residents," Aplt.’s Reply Br. at 8-9, by requiring it to perform "extensive, data-driven" research before legislating, Aplt.’s Opening Br. at 27. But Castle Rock attacks a straw man. We have held no such thing. The upshot of our decision here is only that, when burdening commercial speech, a municipality must be able to demonstrate "that the harms it recites are real and that its restriction will in fact alleviate them to a material degree." Edenfield , 507 U.S. at 770-71, 113 S.Ct. 1792. It theoretically could do so by presenting many different types of evidence: "[t]he government is not limited in the evidence it may use to meet its burden." Mainstream Mktg. , 358 F.3d at 1237. Castle Rock cites to Lorillard Tobacco Co. v. Reilly , supra , and Phillips v. Borough of Keyport , 107 F.3d 164 (3d Cir. 1997) (en banc), in further support of its argument on this point, but neither leads to a contrary conclusion. Castle Rock only cites to Lorillard ’s recitation of the standard enunciated in Went For It ; Lorillard neither adds to nor subtracts from that standard. See Lorillard , 533 U.S. at 555, 121 S.Ct. 2404. And we are bound to follow Pacific Frontier ’s interpretation of that standard, as we have done here. So, Castle Rock's reliance on Lorillard is unavailing.

In Phillips —which did not apply the Central Hudson framework at all—it is true that the Third Circuit stated that "[w]hatever level of scrutiny we have applied in a given case, we have always found it acceptable for individual legislators to base their judgments on their own study of the subject matter of the legislation, their communications with constituents, and their own life experience and common sense." 107 F.3d at 178. However, in making this statement, the Phillips court was only rejecting appellants’ suggestion that the "factual basis for a legislative judgment" regarding a First Amendment regulation had to "have been submitted to the legislative body prior to the enactment of the legislative measure." Id. (emphasis added); see id. ("[I]n appellants’ view, a governmental entity may successfully defend a First Amendment challenge of the kind here mounted only if it can show that it was exposed, before taking action, to evidence from which one could reasonably conclude that undesirable secondary effects would occur in the absence of legislative action and that the particular action taken was narrowly tailored to ameliorate those secondary effects."). Yet the Phillips court left no doubt that its position regarding pre-enactment evidence did not mean that there was no "requirement that there be a factual basis for a legislative judgment presented in court when that judgment is challenged." Id. In fact, the Phillips court required the municipality to put on a substantial showing indeed: "our First Amendment jurisprudence requires that the Borough identify the justifying secondary effects with some particularity, that they offer some record support for the existence of those effects and for the Ordinance's amelioration thereof ...." Id. at 175. Castle Rock's reliance on Phillips is thus unavailing. Even if we were to adopt that court's view concerning the lack of necessity for a sufficient pre-enactment evidentiary showing—a matter that is not before us and upon which we do not opine—that would not help Castle Rock because it still could not satisfy Phillips ’s "requirement that there be a factual basis for a legislative judgment presented in court when that judgment is challenged." Id. at 178. That is, as we have demonstrated supra , both the pre- and post-enactment evidence Castle Rock presented in district court in support of its public-safety justification was woefully inadequate.

Castle Rock also endeavors (ultimately unsuccessfully) to make a viable argument based on the factually dissimilar Supreme Court case City of Renton v. Playtime Theatres, Inc. , 475 U.S. 41, 106 S.Ct. 925, 89 L.Ed.2d 29 (1986). There, the Supreme Court upheld a zoning ordinance that applied to adult motion picture theaters. In doing so, it held that the city "was entitled to rely on the experiences of Seattle and other cities ... in enacting its adult theater zoning ordinance." Id. at 51, 106 S.Ct. 925. As most relevant here, the Court stated, "[t]he First Amendment does not require a city, before enacting such an ordinance, to conduct new studies or produce evidence independent of that already generated by other cities, so long as whatever evidence the city relies upon is reasonably believed to be relevant to the problem that the city addresses." Id. at 51-52, 106 S.Ct. 925.

Castle Rock argues that it similarly could rely on other cities’ experiences. Aplt.’s Reply Br. at 10-11. We have no quarrel with this unremarkable proposition. See Lorillard , 533 U.S. at 555, 121 S.Ct. 2404 ("[W]e have permitted litigants to justify speech restrictions by reference to studies and anecdotes pertaining to different locales altogether ...." (alteration in original) (quoting Went For It , 515 U.S. at 628, 115 S.Ct. 2371 )); Phillips , 107 F.3d at 175 (noting that "[i]t may well be that the defendants here, by pointing [inter alia ] to studies from other towns ... will be able to carry their burden of showing that the ordinance [which allegedly infringes on First Amendment interests] is reasonably designed to address the reasonably foreseeable secondary effect problems" of plaintiffs’ adult-entertainment business).

However, in Playtime Theatres , the Court allowed Renton to rely on studies that Seattle (a neighboring city) had commissioned that established the basis for "detailed findings" concerning the "harmful effect on th[at] area" and "neighborhood blight" caused by adult movie theaters. 475 U.S. at 50-51, 106 S.Ct. 925 (quoting Northend Cinema, Inc. v. City of Seattle , 90 Wash.2d 709, 585 P.2d 1153, 1156 (1978) (en banc)). Castle Rock has presented no similar evidence here from other cities. It does not even identify specific municipalities that have purportedly experienced similar problems with commercial door-to-door solicitors, let alone why the experience of those cities "is reasonably believed to be relevant to the problem that [Castle Rock] addresses." Id. at 51-52, 106 S.Ct. 925. It certainly has not presented any evidence containing detailed findings concerning these matters. Instead, Castle Rock only points to the fact that other municipalities have similar ordinances . But the fact that other cities have similar ordinances cannot, standing alone, give us any basis to infer that the public-safety "harms" that Castle Rock "recites are real," Edenfield , 507 U.S. at 770-71, 113 S.Ct. 1792. The ordinances of those cities could very well be grounded on the same sort of inadequate common-sense and anecdotal evidence as Castle Rock's. Or, conversely, their door-to-door solicitors may have in fact presented local officials with real public-safety concerns for reasons unique to those cities. Based on Castle Rock's showing there is no way to tell, and that is a problem for Castle Rock. That means that it has not carried its burden of proof to demonstrate the reality of its claimed public-safety interest by reference to evidence of such harms in other cities. Therefore, we reject Castle Rock's argument predicated on Playtime Theaters .

Castle Rock also notes that in Watchtower the Supreme Court "recognized the interests a town may have in some form of regulation, particularly when the solicitation of money is involved." 536 U.S. at 162, 122 S.Ct. 2080. Other cases note the potential for safety issues even more specifically. See, e.g. , Int'l Soc'y for Krishna Consciousness, Inc. v. Lee , 505 U.S. 672, 684, 112 S.Ct. 2701, 120 L.Ed.2d 541 (1992) ("[F]ace-to-face solicitation presents risks of duress that are an appropriate target of regulation."). But no one disputes that Castle Rock's interest in public safety is substantial in the abstract or that certain forms of regulation might be necessary to limit commercial solicitation. The problem here is a matter of proof or, in Pacific Frontier ’s language, of demonstrating that Castle Rock "faces real harms." 414 F.3d at 1235 ; see Edenfield , 507 U.S. at 770, 113 S.Ct. 1792 ("That the Board's asserted interests are substantial in the abstract does not mean, however, that its blanket prohibition on solicitation serves them."). Castle Rock has not presented evidence of any risk of crime associated with registered commercial solicitation after 7:00 p.m., and its police chief's testimony undermines any claim to the contrary. The fact that Castle Rock's interest in public safety is substantial in the abstract does not allow Castle Rock to justify any regulation simply through talismanic invocation of that interest.

Castle Rock makes the related argument that even if the regulations do not advance its interest in public safety, they advance its interest in "ensuring that Town residents feel safe in their community." Aplt.’s Opening Br. at 27 (emphasis added). But the primary connection between this interest and the Curfew that Castle Rock makes is a citation to the following statement from its police chief:

I think that, you know, common sense tells us that as it gets later in the evening, people are a little bit more anxious about [opening their door to a stranger]. One reason is because they're less likely to be somebody, just a neighbor or a friend, show up at your door unannounced at the later evenings, as opposed to earlier in the day.

Aplt.’s App., Vol. V, at A1128. The chief continued to explain that the Curfew helped to provide a "bright line" that has provided clarity to residents. Id. at A1125, A1128. But this "common sense" rationalization is "wholly insufficient." Pac. Frontier , 414 F.3d at 1235 n.12. Castle Rock cannot infringe on Aptive's First Amendment interests in its commercial speech based on conjectural harms and suppositions regarding how their citizens might feel about the removal of such harms. See Went For It , 515 U.S. at 628-29, 115 S.Ct. 2371 (noting that the regulation at issue was adequately shown to "target[ ] a concrete, nonspeculative harm"); cf. Ill. Pub. Action Council , 796 F.2d at 1553 (noting that "[t]he Supreme Court has ... emphasized the need for precision in the regulation of the exercise of First Amendment rights").

In particular, while the record does contain evidence about various citizen complaints, many of these complaints are not relevant to the specific Curfew at issue here—in that they do not focus on interactions with commercial solicitors or interactions after 7:00 p.m.—and the remainder are undermined both by the quantitative data that has been analyzed supra and by the police chief's own testimony. Without concrete evidence of relevant complaints, the fact that some of Caste Rock's citizens may feel "anxious" is woefully insufficient to demonstrate that Castle Rock "faces real harms, which are materially palliated by the [Curfew] requirements." Pac. Frontier , 414 F.3d at 1235.

In sum, we hold that Castle Rock has failed to demonstrate that the Curfew directly advances in a material way its substantial interest in public safety.

b

Castle Rock next argues that the Curfew directly advances its interest in its citizens’ privacy. See Aplt.’s Opening Br. at 23-26. Like its interest in public safety, Castle Rock's interest in protecting the privacy of its citizens is in the abstract an undisputedly substantial one. See Carey v. Brown , 447 U.S. 455, 471, 100 S.Ct. 2286, 65 L.Ed.2d 263 (1980) ("Preserving the sanctity of the home, the one retreat to which men and women can repair to escape from the tribulations of their daily pursuits, is surely an important value."); accord Pac. Frontier , 414 F.3d at 1232 n.10. However, Castle Rock must again demonstrate that the harm it recites, i.e., the invasion of its citizens’ privacy by commercial solicitors after 7:00 p.m., is "real" and that this Curfew alleviates that harm "to a material degree." Edenfield , 507 U.S. at 770-71, 113 S.Ct. 1792. It primarily attempts to carry this burden through a series of personal anecdotes. While as discussed above we agree that ordinances can be justified by anecdotes, we again conclude that Castle Rock's evidence is wholly insufficient.

We start by recounting Castle Rock's support for its assertion that the Curfew directly advances its interest in protecting its citizens’ privacy. This support comes primarily from the personal experiences of council members and town employees and their discussions with constituents. Castle Rock argues that—before enactment of the Curfew in 2008—"the Town received complaints regarding solicitors, and at least one related to evening-hours solicitation." Aplt.’s Opening Br. at 23-24. But the only pre-enactment evidentiary support relevant to the Curfew is testimony concerning the single 9:45 p.m. solicitor. See Aplt.’s App., Vol. V, at A1094 (Q: "[T]he only time-specific complaint that the Town Council considered was one at 9:45, when it was dark; right?" A: "Yes."). The former town clerk did testify that after the Curfew was enacted in 2008 "[c]itizens would call and complain" about, as relevant here, "unregistered solicitors at their door" and "solicitors at their door after the curfew time." Id. , Vol. IV, at A981-82.

In addition to this collection of anecdotes, Castle Rock also attempts to support its privacy rationale for its Curfew requirement by relying on common sense, which as noted is permissible as a general matter. A former town council member testified about his personal belief that the Curfew "gives you ... a time period that you know you will ... have privacy," id. , Vol. V, at A1080 81, and about his conversations with neighbors who thought 7:00 p.m. was "a reasonable time," id. at A1082. The former police chief stated that community members felt "uncomfortable" with having strangers come to their door. Id. at A1313. And the town manager testified, based on conversations with several citizens, that citizens wanted to "keep the curfew the way that it is" and had "a problem with uninvited solicitations [sic] showing up on their doorstep after 7 o'clock." Id. , Vol. IV, at A920, A926, A970. He thought the Curfew "reflect[ed] the common sense view that as the hours ... get later, past dinnertime, people are moving toward more private activities within their home, and they're not wanting to be inconvenienced or bothered by an uninvited stranger on [sic] their door." Id. at A940. The mayor also testified that she believed that community members supported the Curfew. Counsel asked the mayor whether, "based on the input [she] received from [her] constituents, [she thought] there [was] sort of a community-wide concern about commercial solicitors going door to door after 7:00 p.m. in the evening." Id. , Vol. V, at A1183. She answered as follows: "There is [sic] concerns from residents about that. Yes." Id.

Castle Rock also points to "other municipalities’ experiences" with evening-hours solicitation invading residents’ privacy. Aplt.’s Opening Br. at 24. However, as we explained in connection with Castle Rock's public-safety justification, supra , Castle Rock's citations on this point simply indicate that other municipalities have similar ordinances . That fact, standing alone, provides us with no basis to infer that the privacy "harms" that Castle Rock "recites are real," Edenfield , 507 U.S. at 770-71, 113 S.Ct. 1792, or that Castle Rock seeks to address a "concrete, nonspeculative harm" to its citizens’ privacy, Went For It , 515 U.S. at 628-29, 115 S.Ct. 2371. Cf. Playtime Theatres , 475 U.S. at 50-51, 106 S.Ct. 925 (allowing municipality to rely on studies commissioned by neighboring municipality that established the basis for "detailed findings" concerning the "harmful effect on th[at] area" and "neighborhood blight" caused by the regulated activity (quoting Northend Cinema , 585 P.2d at 1156 )).

Taken together, the record then shows a handful of complaints about evening solicitation spread over approximately ten years and the representations of Castle Rock officials and employees that at least some (unknown) number of citizens thought the Curfew made sense and was a good idea. This evidence is either irrelevant or undercut by contrary quantitative data.

As for lack of relevancy, the vast majority of this testimony tells us nothing about whether commercial solicitation after 7:00 p.m. is a "real" harm to privacy. As we have generally discussed supra , the evidence focuses on all i.e., not just commercial door-to-door solicitation and it also includes incidents that definitely or likely took place during the daytime, if the timing of the incidents may be discerned at all. See, e.g. , Aplt.’s App., Vol. IV, at A983, A1015 (describing daytime incident between council member and solicitor); id. , Vol. V, at A1082 83 (same); id. at A1084 (describing interactions with children selling candy and magazines, without reference to time of day); id. at A1305 (describing complaints about solicitors without reference to time of day); id. at A1313 (same); id. at A1318 19 (Email re: Council Legislative Items for Aug. 21 Study Session, dated Aug. 6, 2007) (same). Underscoring the lack of evidence relating to the period after 7:00 p.m., remember that the 2008 ordinance and, more specifically, the Curfew grew out of the daytime interaction between a solicitor and a council member, and Castle Rock's mayor described two daytime interactions she had with solicitors. And while Castle Rock claims that "[c]omplainants reported harassment, intimidation, and discomfort as a result of solicitation," Aplt.’s Opening Br. at 24, the only record support for this allegation makes no reference to evening solicitation.

More specifically, there is no support for a finding that commercial solicitation after 7:00 p.m. has intruded on the privacy of Castle Rock's citizens. At most, the general and undifferentiated evidence concerning complaints demonstrates that some Castle Rock residents did not like door-to-door solicitation in any form—including during the day or by noncommercial solicitors. Cf. Vill. of Schaumburg , 444 U.S. at 638-39, 100 S.Ct. 826 (noting that the local government's "requirement is related to the protection of privacy only in the most indirect of ways" given that the government "concedes[ that] householders are equally disturbed by solicitation on behalf of organizations satisfying the [requirement] as they are by solicitation on behalf of other organizations").

Lastly, we return to the data from 2016 and 2017 that established that there were only eleven complaints about doorbell interruptions after 7:00 p.m., and none of those complaints involved commercial solicitors. Instead, all eleven of the complaints about interruptions after 7:00 p.m. concerned noncommercial solicitors. But the Curfew, which only applies to commercial solicitors, does nothing to stop such interruptions. Thus, even if Castle Rock had demonstrated that door-to-door solicitation imposed a real harm to its citizens’ privacy, there is ample reason to doubt that this particular Curfew "w[ould] in fact alleviate" it, let alone "to a material degree," because noncommercial solicitors—the group of identified offenders—would be able to continue soliciting throughout the evening. Lorillard , 533 U.S. at 528, 121 S.Ct. 2404 ; cf. Pac. Frontier , 414 F.3d at 1234-35.

In sum, our cases and the record demonstrate that, as to the privacy interest it asserts, Castle Rock has not carried its burden of demonstrating "that the harms it recites are real and that its restriction will in fact alleviate them to a material degree." Edenfield , 507 U.S. at 770-71, 113 S.Ct. 1792.

* * *

Based on the record before us, we conclude that Castle Rock has not demonstrated that the Curfew directly advanced to a material degree its interests in public safety and privacy. Because Castle Rock has failed at this second step of the Central Hudson analysis, we need not and do not proceed to the final step. See Adolph Coors Co. v. Bentsen , 2 F.3d 355, 359 n.6 (10th Cir. 1993) ("Because we conclude that the Government has failed to satisfy its burden under the third part of the Central Hudson test, we need not proceed to the fourth part to determine whether there is a reasonable fit between the prohibition and the Government's interest." (framing Central Hudson as involving four prongs)), aff'd sub nom. Rubin v. Coors Brewing Co. , 514 U.S. 476, 115 S.Ct. 1585, 131 L.Ed.2d 532 (1995) ; see also People for Ethical Treatment of Prop. Owners v. U.S. Fish & Wildlife Serv. , 852 F.3d 990, 1008 (10th Cir. 2017) ("[I]f it is not necessary to decide more, it is necessary not to decide more." (alteration in original) (quoting PDK Labs. Inc. v. DEA , 362 F.3d 786, 799 (D.C. Cir. 2004) (Roberts, J., concurring in part and concurring in the judgment))). We thus conclude that Castle Rock's Curfew does not withstand First Amendment scrutiny.

IV

For the foregoing reasons, we conclude that Aptive has standing to sue and that Castle Rock's Curfew unconstitutionally burdens Aptive's First Amendment rights. Thus, we AFFIRM the district court's judgment.

HARTZ, J., Circuit Judge, concurring

I concur in the judgment and join all of the majority opinion except § III(C)(2)(b), which addresses Castle Rock's argument that the curfew can be justified by its interest in residents’ privacy. The privacy right at issue in the present context is the right to be left alone. As the Supreme Court has said: "The unwilling listener's interest in avoiding unwanted communication ... is an aspect of the broader ‘right to be let alone’ that one of our wisest Justices characterized as ‘the most comprehensive of rights and the right most valued by civilized men.’ " Hill v. Colorado , 530 U.S. 703, 716–17, 120 S.Ct. 2480, 147 L.Ed.2d 597 (2000) (quoting dissent of Justice Brandeis in Olmstead v. United States , 277 U.S. 438, 478, 48 S.Ct. 564, 72 L.Ed. 944 (1928) ). It added, "The right to avoid unwelcome speech has special force in the privacy of the home." Id . at 717, 120 S.Ct. 2480 ; see Frisby v. Schultz , 487 U.S. 474, 492, 108 S.Ct. 2495, 101 L.Ed.2d 420 (1988) (Brennan, J., dissenting) ("I also agree with the Court that the town has a substantial interest in protecting its residents’ right to be left alone in their homes."). Castle Rock asserts that its curfew ordinance is justified because residents do not want to be bothered by commercial solicitors after 7:00 p.m. As I understand the majority opinion, it states that the City has not shown that this justification—this harm to privacy interests—is "real."

I am not sure what is meant by "real." Perhaps if there were real doubt about whether commercial solicitors would bother to contact residents after 7:00 p.m., one could say there is insufficient evidence to show that the harm is "real." But there is no doubt that, absent the ordinance, Aptive's independent contractors would be soliciting residents after 7:00 p.m. So I assume that the majority opinion is not complaining about lack of evidence in that regard.

A second possibility is that there is no evidence that any resident would feel bothered—would believe his or her privacy interests were offended—by being contacted by a commercial solicitor after 7:00 p.m. But I do not think that the majority opinion is saying that such evidence is totally absent. After all, if one of the city counselors who voted for the ordinance did not want to be bothered by commercial solicitor after 7:00 p.m., that would be evidence regarding one resident. And conversations of counselors with their constituents who express support for the ordinance would multiply that evidence. I would also think, although the majority opinion rejects this, that the fact that other municipalities had tried to justify evening curfews at various hours on the ground that they protect the privacy of residents would serve as evidence that at least some city dwellers of this era do not like to be bothered by commercial solicitors late in the day.

I conclude that when the majority opinion states that Castle Rock "provides us with no basis to infer that the privacy harms that [it] recites are real, or that [it] seeks to address a concrete nonspeculative harm to its citizens’ privacy," Maj. Op. at 997-98 (citation and internal quotation marks omitted), it means that the above evidence mentioned by the majority does not show that the privacy harms have reached a minimal threshold.

It is at this stage of the analysis that I become perplexed. Two questions need to be answered. First, what is the standard that must be met by the City's evidence? That is, what is the threshold of harm? Is it enough that a few residents do not wish to be bothered? Is it enough that a substantial portion of the City's residents do not want to be bothered? Does a majority of the City suffice, or must it be almost everyone? Second, what kind of evidence will suffice to prove that the standard has been met? The extent to which residents believe that their privacy has been invaded cannot be measured by impersonal "hard" evidence, as when a law is justified as a protection of health or safety. An individual is likely to measure public sentiment on the issue simply by recalling how often such privacy concerns come up in casual conversation. Can city leaders rely on their sense of what the community wants? Are complaints necessary; and if so, how many? Or are vote tallies or survey results the only legally acceptable sources?

The majority opinion states that its analysis is consistent with the Second Circuit opinion in Vugo, Inc. v. City of New York , 931 F.3d 42 (2019). In that opinion the court said that the ban on Taxi TV in taxicabs and for-hire vehicles was justified by a survey that substantiated the harm it sought to prevent. The survey showed that "nearly one-third [33 1/3 %] of respondents indicated that Taxi TV is annoying." Id . at 52 (internal quotation marks omitted). Although an independent survey showed that 45% of respondents thought Taxi TV was a "pleasant diversion" while 41% thought it annoying, the court wrote that "[t]his single third-party survey does not provide a basis for us to second guess the City's conclusion that in-ride advertisements are annoying to its citizens—a conclusion it reached based on its own survey results and first-hand experience receiving complaints from customers." Id . This sentence may suggest that majority displeasure is necessary, but that a 41% survey result plus constituent complaints and the personal experience of city officials could suffice to establish that the threshold had been reached. A footnote, however, added: "Moreover, we see no reason why the City may not seek to alleviate a harm when the harm is experienced by forty-one percent of the population." Id . at n.9. An explicit endorsement of this analysis in the majority opinion would be helpful.

But it appears that the majority would not go so far. Applying Vugo to the present context, I would think that there is sufficient evidence of "real" harm. The record indicates that 30% of the households in Castle Rock had signed up for the do-not-knock list, which prohibits commercial solicitors from bothering them at any time of the day. See Aplt. App., Vol. 2 at 419 n.20 (list contains over 6300 addresses out of some 21,000 housing units in the town). Can we not infer that at least 30% of the households would not want to be bothered after 7:00 p.m.? Does a commercial solicitation after 7:00 p.m. not harm a resident who also would be upset by a solicitation at noon, or 6:00 p.m., or 6:30 p.m.?

I would hope that we could provide useful guidance to municipalities in this circuit that would like to consider solicitation curfews. To be safe, a city may wish to pay for a resident survey before considering whether to enact an ordinance. But unless the survey results are overwhelming in one direction or the other, city officials might well be at a loss regarding the likelihood that an ordinance would pass constitutional muster. City attorneys may find it difficult to provide advice based on the majority opinion.

I think it would have been preferable if this court had avoided those issues and resolved this case on another component of the Central Hudson analysis that could provide clear guidance. In my view, it is likely that no 7 p.m. curfew could survive Central Hudson ’s requirement that "it is not more extensive than is necessary to serve [the government's asserted] interest." Cent. Hudson Gas & Elec. Corp. v. Pub. Serv. Comm'n , 447 U.S. 557, 566, 100 S.Ct. 2343, 65 L.Ed.2d 341 (1980) ; see id. at 564, 100 S.Ct. 2343 ("[I]f the governmental interest could be served as well by a more limited restriction on commercial speech, the excessive restrictions cannot survive."); Edenfield v. Fane , 507 U.S. 761, 767, 113 S.Ct. 1792, 123 L.Ed.2d 543 (1993) (describing Central Hudson requirement as "whether the extent of the restriction on protected speech is in reasonable proportion to the interests served"). The record indicates that with relative ease Castle Rock could achieve its ends in a way that restricts commercial solicitors from intruding on only those residents who are offended by the intrusion. The Ordinance already contains a provision that prohibits solicitation of a home if the homeowner requests to be put on a do-not-knock list. Commercial solicitors are required to check the list and comply with it. The record indicates that the prohibition has been violated only once, and in that case apparently inadvertently by a sight-impaired solicitor. There is no reason to think that registered commercial solicitors would violate a provision stating that they should not contact homes that are listed as not wanting commercial solicitors after 7 p.m. I would infer that banning all commercial solicitation after 7 p.m., which burdens solicitation of those who do not object to the practice (a substantial additional constraint on the solicitors), is more burdensome than necessary to further the City's substantial interest in privacy. Thus, the reasoning that requires invalidating the 7 p.m. curfew in the present ordinance at the same time suggests a workable well-tailored solution.