Opening Cross: Test the Water Before Diving In (And Other Old Sayings with New Applications)

But is that still the case? These days, it seems like the main reasons for sticking with big, traditional vendors are their breadth of service and the potential to leverage economies of scale to get more data at a lower price point, rather than the IBM adage, which emphasizes reputation and reliability. In many cases, end-users are willing to take a chance on working with new vendors—indeed, in many cases, they need to work with new vendors to obtain the new types of datasets they want—because these may be able to offer specific datasets cheaper, and deliver greater return on investment.

But with many firms so wary of extra expense that even setting up a proof of concept trial might be nixed if it costs too much to get up and running, how does one go about establishing the risk and ROI of a new dataset or vendor before actually using it?

There’s another old saying attributed to the British SAS: Who dares wins. And the prospect of winning trades in an area where rivals won’t yet catch up is making traders more daring, and more willing to investigate new datasets—perhaps such as earnings events data like that offered by Wall Street Horizon, which startup Kyper has integrated into its upcoming Data-as-a-Service platform to correlate with other data sources. As demonstrated by strategy modeling software provider Deltix last year, changes to a company’s earnings announcement date can be predictive of that company’s performance. Correlating this with other historical data to generate patterns could help provide further depth and context to the WSH events dataset by predicting the impact and duration of that effect.

And while some are looking for value in data that’s been there all along, others are creating new data entirely. For example, to boost CDS market liquidity and data quality, bond trading platform operator DelphX is designing a new ATS to trade CDS protection contracts known as Default Swap Receipts that it says will help restore liquidity to the single-name CDS markets.

And of course, there’s the old chestnut favored by proponents of Big Data and digital imagery analysis: A picture is worth a thousand words. And once again, it’s our friends at Deltix who are helping to pioneer in-depth exploration of new “datasets”—in this case, satellite images of retail store parking lots from Orbital Insight, to assess how predictive a full parking lot is of full coffers for those stores come earnings season. Combine that with the WSH data that Deltix has already tested, and you have an exciting combination: real-time customer traffic data to plot against earnings dates and company performance.

And for every new dataset, there are metrics that must be monitored to ensure accuracy. For example, those obsessed with low latency—or frankly, just those who need to comply with MiFID 2’s timestamping requirements—will be aware of the acute importance of having consistent timestamps across the entire flow of data. After all, if you can’t guarantee the accuracy of your timestamps, you can’t guarantee accurate latency levels to make your trading strategies work. So many firms source timing data from independent GPS sources for use throughout their organizations. But a weak signal or obstruction could lead to potential errors in timing. So time synchronization technology provider FSMLabs has developed a GPS validation tool, dubbed SkyMap, that visualizes stronger and weaker signals reaching a GPS antenna. While this won’t deliver guaranteed timing, it will help firms spot if they have a problem, rather than continuing trading in blissful ignorance.

So if timing matters to you, don’t forget to test the water—or rather, the airwaves—before you dive in, lest someone else gets to ride your wave and leaves you in the flotsam.