We’re surrounded by data. It is data produced in both private and public sectors. It is data generated by individuals or machines. The variety of Big Data is large and growing rapidly across both internal and external sources and offer immense opportunities for those who will take advantage of it.
Behind the firewall, Big Data is stored in databases, file systems, Hadoop frameworks, documents, archives and legacy on-premise applications for ERP, CRM, content management and more. Externally, outside the firewall, it is available in cloud-based applications (like Salesforce, Marketo and Ariba), partner and supplier portals, public web like government websites or competitor websites, social networks and many more. External data on economic indicators, public finance, healthcare, regulatory compliance and more also play a central role in many Big Data use cases. Everyday consumers are creating Big Data across Facebook, Twitter, LinkedIn, blogs, reviews and forums, offering insights into their preferences and behaviors.
To add to this variety, each data source has different characteristics. Data can be structured, like machine data or sensor data, well organized in fields within a record or file or it can be unstructured, like social media data with no pre-defined data model. It can also be somewhere in the middle between structured and unstructured. Some data sources will have APIs and others will not, requiring alternative methods in order to access and extract data.
The bottom line is that there is really no one “killer” data source. It is the unique combination of sources that you tailor to address your business needs that will be the “killer” data for your organization. The ever-expanding pool of data sources and types is becoming available and opening a whole new world of possibilities. Start taking advantage of it now, experiment with new sources, augment traditional sources, iterate and find the data gems that matter to your business!