Sabafon, a telephone company based in Yemen, wanted to buy cards that would provide its customers with prepaid minutes of phone use. To make these cards more *970attractive, Sabafon wanted to add a game of chance. Both the number that would supply phone time and the symbols representing prizes were to be covered by a scratch-off coating. (An opaque coating prevents the numbers and symbols from being discovered before the cards are sold, and it also prevents buyers from knowing in advance which parts of the card contain the winning boxes.) Emirat, a German firm in the risk-management business, promised to supply Sabafon with 25 million high-security scratch-off cards. Emirat contracted with High Point Printing LLC of Ohio to print the cards, and High Point in turn engaged WS Packaging of Wisconsin to do the work.
Emirat paid High Point about $700,000, and in this suit under Wisconsin law it demands that much in damages-not from High Point, which is defunct, but from WS Packaging. Emirat did not have a contract with WS Packaging, whose deal was with High Point. Still, Emirat contends that a settlement agreement it reached directly with WS Packaging after an initial run of cards was not shipped in numerical order, 500 pieces to a tray, subjects WS Packaging to all terms of the contract between Emirat and High Point. The district court thought otherwise and granted summary judgment to WS Packaging. 248 F.Supp.3d 911 (E.D. Wis. 2017). To simplify this decision, we assume that WS Packaging is indeed obliged to perform up to the standards that High Point promised Emirat. But what are those standards?
First we need to explain why Emirat refused to accept the cards. A scratch-off covering is supposed to prevent the reading of phone codes or game symbols through a process known as candling-shining a bright light behind the cards to reveal the printing under the opaque coating. With a goal of zero candling, WS Packaging tested its print runs using high-intensity light and did not find a problem. Emirat wanted extra assurance and sent samples of the cards to Force Technology in Denmark. Following protocols prescribed by Emirat, Force Technology tried to read the printing using light from the sun, fluorescent tubes, infrared (thermal), x-rays, and lasers. It reported that "it is not possible-with the techniques used-to disclose the symbols underneath the scratch layer."
That was June 2009. In October 2009 Emirat sent Force Technology more samples. It used the same tests and reached the same result: "based on the cards received and the tests done on the cards, ... the symbols can not be disclosed and the security of the scratch card ... is good." But Sabafon insisted that some cards could be candled by removing them from their plastic overwrap and bending them until the scratch-off layer cracked, then shining a bright light from behind. Emirat sent a third batch of cards to Force Technology with instructions to bend the cards around a pencil, crack the coating, and find out what could be seen. In December 2009 Force Technology reported that it was able to see some digits and game symbols using this technique. Emirat rejected the whole print run and demanded its money back, but by then High Point was out of business.
Emirat insists that zero candling is the only acceptable level. By that standard, however, no scratch-off card is secure. With a sufficiently high-tech approach (think CT scanners or muon tomography ), or sufficiently clever manipulation, any security can be compromised. But no one will spend $1,000 to break the security of a card promising $50 worth of phone time. Security suffices if it prevents practical kinds of cheating. A trade association called the World Lottery Association has defined what this means: a scratch-off card is a security risk only if it can be candled *971in five minutes or less. This is the norm from which contractual negotiations can begin. A buyer that wants more security-resisting candling for 10 minutes, 20 minutes, an hour, or forever-can contract for it. But Emirat did not do so. It promised "high level" security (an undefined phrase) to Sabafon, but its contracts with High Point and WS Packaging are silent on candling. Our assumption that WS Packaging assumed all of High Point's promises still leaves silence.
In this suit Emirat presented the evidence of two experts, James Blanco and Larry Stewart, who bent the cards, cracked the coating, and then tried to read the printing. Blanco reported that it took him 13 or 14 hours to learn how to bend a card without doing so much damage to the coating that the tampering could be detected. After acquiring this skill, he stated, he could read all digits of the phone code and some prize symbols on some cards. ("Some" prize symbols would not have been enough. To win a prize the customer had to scratch the coating off exactly six areas, which had to contain the same prize. Each card had six identical prize symbols and 18 different prize symbols, so unless close to all of the symbols could be read it would not be possible to choose the six winning areas to scratch off.) Stewart testified at a deposition that after bending and cracking cards he could decipher some of the digits in the phone code but that he did not see all of the digits in any code, because "[t]hat would take many more hours than I was provided here."
Both Blanco and Stewart conceded that the cards were not security risks under the World Lottery Association's five-minutes-per-card standard. That is dispositive against Emirat, for neither High Point nor WS Packaging promised any higher level of security. (WS Packaging's internal goal of zero candling was not a promise to Emirat.)
We might imagine an argument that the World Lottery Association is a tool of the printing industry and that its five-minute standard is not the norm from customers' perspectives. But Emirat does not make such an argument; the Association's principal members are lottery-running governments-volume buyers of scratch-off cards. Nor does Emirat contend that usages of trade supply a norm that cards must be more candling-resistant than the five-minute standard. It has presented evidence that participants in the industry expect to receive cards that do not candle. What it has not presented is evidence that "do not candle" has a shared meaning different from the trade association's standard. One of Emirat's witnesses testified that cards are candle-proof if a high-intensity light does not reveal what is under the coating. WS Packaging's cards passed that test repeatedly. This witness did not mention bending cards to crack the coating as a norm in the industry. Other witnesses, including Blanco and Stewart, testified to different approaches. Force Technology did not think to bend the cards until Emirat told it to do so. Multiple witnesses testified that cards' security-more valuable cards need extra security-normally is specified by contract. This implies that there is no off-the-rack standard, no usage of trade read into every contract, requiring more than the trade association's five-minute standard.
The district court treated the contract as too ambiguous to make the "how secure?" subject appropriate for summary judgment. But the contract is not ambiguous. It is silent . It does not promise Emirat any level of security, except through the possibility that usages of trade are read into every contract for scratch-off cards. And Emirat has not produced evidence-has not even argued-that there is any usage of trade, except perhaps the World Lottery Association's five-minute standard.
*972Because this record does not contain evidence from which a reasonable jury could conclude that WS Packaging promised to supply cards more secure than the World Lottery Association's norm, and because Emirat's own expert witnesses conceded that these cards could not be candled in less than five minutes, the judgment of the district court is
AFFIRMED .