dissenting in part.
Without justification, the majority reads two new limitations into claim 13 — one of which is not even urged by either party— and then holds that Mirror Worlds’ evidence does not show that those limitations are met. I cannot agree. I believe that under the correct reading of the claim Mirror Worlds adduced sufficient evidence to allow a reasonable jury to conclude that Apple’s customers infringed (and continue to infringe) claim 13 of the '227 patent. Thus, with respect, I dissent from the majority’s conclusion that the district court did not err in granting judgment as a matter of law that Apple did not induce infringement of the '227 method patent.
I
There is no dispute that the use of Spotlight would infringe claim 13 if four things occur: 1) a search, 2) receiving data units from other computer systems, 3) generating data units, and 4) generating a sub-stream. What the majority fails to note, however, is that for the most part, these four steps are performed automatically and without any need for user interaction. The first step (searching) occurs even when the user is not using Spotlight. Indeed, that is precisely what enables Spotlight to return search results quickly. Spotlight continuously combs through every file or document that exists on the computer and collects certain information from the file such as creation time, content, and owner. This information is then immediately incorporated into an all-inclusive index akin to a look up table. By having this comprehensive look up table ready at hand, Spotlight is able to return search results almost immediately upon request. As Mirror Worlds’ expert explained, this constant monitoring of computer data reduces the search operation to a simple filtering of the already organized data based on the user’s search query. In sum, Spotlight’s search engine never sleeps.
The next two steps (generating and receiving data units) are so common to any *1363computer operation that is hard to imagine one can do anything with a computer without also performing these steps. The user “generates data” every time she creates a file or document, such as a letter, a drawing, a photograph, a calendar entry, a shopping list, or anything of that sort. The computer “receives data” whenever the user receives an e-mail, an online instant message, or even a file while visiting a webpage. Simply put, chances are much better than 50/50 that every time an Apple computer has been used, data have been generated and received.
That leaves creating a substream. Under the district court’s undisputed claim construction, a substream is “a subset of data units, or documents, yielded by a filter on a stream.” J.A. 103. A “main stream” is on the other hand “inclusive of every data unit, or document, received by or generated by the computer system.” Id. at 102. In plain English, a substream is generated whenever the user runs a search query in Spotlight, which query is then used to filter the main stream (encompassing all the data) and generate a sub-stream (the search result).
Putting it all together, claim 13 simply requires that a search query be run in Spotlight on a computer (that at least contains one file and also that at some point has received one e-mail or other type of file from another computer), and that Spotlight filter its organized database based on the inputted query (return search results).
That’s it.
II
Mirror Worlds introduced ample evidence to allow a reasonable jury to determine that the required steps were performed. This court has consistently held that circumstantial evidence suffices to prove infringement. Lucent Techs., Inc. v. Gateway, Inc., 580 F.3d 1301, 1317 (Fed. Cir.2009); Moleculon Research Corp. v. CBS, Inc., 793 F.2d 1261, 1272 (Fed.Cir.1986). If that is true, then Mirror Worlds’ evidence should have sufficed. The evidence includes Apple brochures and manuals that encouraged users to use Spotlight to search through their files, including their e-mails. For example, one manual states that “Mail [ (Apple’s e-mail interface) ] uses the power of Spotlight for faster, more accurate searching.” J.A. 5449. Even more specifically mirroring the claim language, Apple taught its customers to use Spotlight to search through data that are both generated in and received by the computer. For example, one document explained that Spotlight could “search e-mail archives;” another taught users to use Spotlight to search their “entire system from anywhere to find documents, email, contacts, calendars, music, movies, photos, bookmarks, and applications.” J.A. 5461 (emphasis added). But the evidence of infringement is not limited to manuals and brochures. Mirror Worlds also introduced a macworld.com review article that states, “the improved speed of Spotlight, which makes searching for messages within Mail much less painful.” J.A. 6425. This evidence strongly suggests that at least one individual ran a search in Spotlight on a computer that had received data (e-mails). Surely, it would not have been unreasonable for a jury to also assume that the author of the article had at some point saved a single file on his computer too, so that the data generation requirement would be satisfied. Based on all this evidence, the jury could have reasonably found that at least one Apple customer, more likely than not, infringed claim 13— indeed, one would think it unlikely that the jury would arrive at the opposite conclusion.
But the majority is not satisfied. It casually brushes the manuals and the review article aside, reasoning that they do *1364not show that the Spotlight search was performed in a manner that would infringe claim 13. The majority’s objection to the evidence appears to be based on two erroneous assumptions about what claim 13 requires: 1) that the search results be displayed in chronological order, and 2) that data be generated and received either while the user performs the search or at some point in time sufficiently close to the search. Neither assumption finds any support in the claim language, the district court’s claim construction, or anywhere else in the record.
Take the macworld.com article as an example. The majority first complains that the article does not show that the search results were displayed chronologically. Majority Op. 1360-61. Where does that requirement come from? Citing the claim language, the majority notes that the claim requires “ ‘time-ordered’ search results.” Id. at 1358 (citing '227 patent col. 16 11.9-25). From that, the majority simply assumes that the search results should be displayed chronologically. But the claim language and the district court’s claim construction (or anything else in the record for that matter) do not justify this logical leap. To begin with, claim 13 does not even include the word “display.” That sets it apart from other claims involved in this litigation that do expressly address how search results should be displayed. See J.A. 116. And, although claim 13 requires that data units be both associated with a “chronological indicator” and accordingly included in the main stream, see '227 patent col. 16 11.19-23, it never recites a similar requirement for a substream. Moreover, there is no reason to believe that bearing a chronological indicator or being “time-ordered” is the same thing as being displayed chronologically. Nor is there anything in the district court’s claim construction that supports the majority’s theory. The only requirement set by the district court is that the data units within the main stream bear a time-stamp, not that they be ordered chronologically. Indeed, the district court did not even address whether the search results should be displayed chronologically; it only addressed (and rejected) Apple’s argument that the main stream data units should be chronologically stored. See J.A. 103. And, Apple does not even whole-heartedly press the chronological display argument on appeal, apart from making two passing remarks neither one of which bears a helpful citation to the record. See Appellee’s Br. 8, 30 (citing J.A. 101, 6420). In similar conclusory fashion, the majority has endorsed Apple’s unsupportable argument.
The majority’s second reason for disregarding the macworld.com review article is even more fragile than the first. The objection is that “the reviewer might have only searched his existing mail, without sending any new mail.” Majority Op. 1361. To try to understand this objection is to undermine it. It is not clear what the majority means by existing e-mail. If existing means “already searched,” then infringement has already occurred. If it means “unread,” then the majority’s reasoning has no basis in the claim language at all. At any rate, claim 13 does not have a temporal limitation; there is no requirement that the generated and received data be fresh. As long as the user at some point has created and received some data on her computer, the pertinent claim limitation is satisfied. And, as already explained, it goes without saying that any computer user has at some point generated and received data on her computer. After all, all computer data are either generated or received, and there would be nothing to search on a computer without data. Finally, and perhaps most remarkably, Apple has not even urged us to read a temporal requirement into claim 13.
*1365In sum, it does not take much user interaction to infringe claim 13 with Spotlight; Mirror Worlds has offered sufficient evidence to allow a reasonable jury to find that Apple’s customers used Spotlight to infringe claim 13; and the majority’s reasons for discounting Mirror Worlds’ evidence are wholly unconvincing. I respectfully dissent.