Data-as-a-Service Analytics Startup Kyper Taps WSH for Events Data

WSH is providing Kyper with forward-looking and historical datasets covering earnings dates, changes to earnings dates, earnings per share, dividend dates and investor conferences. Kyper chief executive Michael Chang says clients will use the WSH data to correlate with market data and news from different sources in its "data marketplace" so their models can be based around multiple variables. The vendor already offers access to content from the Associated Press, The Weather Company (formerly WSI), Benzinga, BarclayHedge and several PR wires, and is in negotiations to add data from other major data vendors, Chang says.

"Kyper is a sophisticated way to take in data and create a quality analytical signal out of different datasets. Kyper finds providers like ourselves, ingests that data, and allows customers to test our data around different scenarios, and do back-testing," says Bruce Fador, president and chief commercial officer of WSH, whose earnings dates revisions content has already been shown to be predictive of company performance. "They want to make it easy for people to juxtapose one type of event with another type of event and analyze the factors contributing to stock movements, then take that thesis and look for similar conditions in future... without having to be a coder."

One way that Kyper achieves this is by reformatting and unbundling data from its vendor partners to make it easy for clients to use different datasets together, from both a technical and commercial perspective. "We built a technology called Data Logistic Technology─our API that understands a dataset's data structure, and restructures it into one machine-readable format without any manual intervention," Chang says, adding that Kyper is targeting quantitative engineers, active traders and small- to medium-sized investment firms.

As such, commercial flexibility is key to adoption, since these types of clients typically cannot afford to purchase large datasets up-front simply to test strategies. "A lot of the time, when people are testing new ideas, they need data from many sources... and testing processes can be time-consuming. You want to be able to test data, and if it doesn't work, you want to be able to move on to the next idea," Chang says. "So we're helping companies like WSH to unbundle so they can reach a different range of customers who can just pay, say, $50 per month for a sub-set of data, rather than thousands of dollars for their entire dataset."

"We've come up with an interesting pricing model so Kyper's clients can test our data a little bit at a time before committing to buying it," Fador says, adding that this model will help WSH to extend its reach to another segment of the market, such as smaller hedge funds and investment managers. "Kyper helps us get to firms that we wouldn't otherwise be able to reach.... We don't want to be in the signals and analytics business; we want to provide data to other people who do that," he adds.

Kyper was set up last year, and is currently in a closed beta testing phase with around 700 users ahead of a planned launch in May, Chang says.