IBM, Datawatch Team Up on Data Preparation for Watson Analytics

By Darryl K. Taft  |  Posted 2016-03-14 Print this article Print
IBM Watson

Moreover, IBM statistics show that people spend a lot of time on data preparation—disproportionately more time than on data analysis, Grosset said.

"We want to enable this mixture of different types of data, which is taking people a lot of time today to do this in a conventional enterprise context," he added. "This is why tools like Datawatch Monarch are so important because they do enable your everyday person to just pick up a tool, shape their data and load it and make it ready for analysis really quickly. This is taking something like 80 percent of people's time when they're doing analysis. So people are not spending time analyzing the important things—they're spending time working on their data."

According to Forrester Research, business analysts and data scientists typically spend up to 80 percent of their time manually preparing data, and it's estimated that only 12 percent of enterprise data is used today to make decisions.

Grosset explained that while the data required for analysis is quite often structured and easy to use, often the data is not in a convenient form. It could be survey data from an online source. It could be open government data, or it could be any one of a number of different sources. Yet people are often mixing data from different sources together to gain new types of insights, he said.

"Imagine you've got your sales data and you want to understand whether the weather had some impact on your sales," said Grosset. "I could go get some weather data and combine that with my sales data and I could start to look for any type of correlation using analytics to see whether my sales went up or down based on weather. We actually know that things like social media activity go up when the weather is bad."

The ability to simply send data accessed and prepared in Datawatch directly to Watson Analytics and Cognos Analytics will enable businesses to select any data source quickly and automatically convert it into structured data for analysis, Marc Altshuller, vice president of Watson Analytics and Business Intelligence for IBM Analytics, said in a statement. "This will not only expedite time to insight, but it increases the likelihood of uncovering new insights that have the potential to transform the business," he said.

Clark Carpenter, infrastructure supervisor at Southeastern Med, an acute health care center, said the combination of Datawatch and IBM Watson Analytics is "well suited" for self-service analytics because of the ability to send data directly into IBM Watson Analytics.



Submit a Comment

Loading Comments...
Manage your Newsletters: Login   Register My Newsletters

Rocket Fuel