Search driven analytics is a new way how you can make ad-hoc queries and get answers simply by entering search query. This works like Google search but for your own dataset, and instead of list of links you get reports with numbers. Search driven analytics is especially good when you need to enable data-driven decisions in your company by non-IT staff - they don't need to know anything about pivot tables and visualization, as reports are configured automatically by search.
SeekTable Ask a question feature allows you to perform data exploration by search. Just enter free-form natural language query and get answers in the form of pivot table or chart. SeekTable helps to form the question with keywords auto-completion: suggestions come from the dataset itself and the cube configuration.
Everyone can use search-driven analytics and get the results for data-driven decision making. Try it with online demo:
average television by age range and gender
show sales sum in 2011 vs 2012 by region and month
Sometimes you can get something different from what you expect; in this case you may back to your search query by returing to cube dashboard ("back" in a web browser or click on the cube name in the breadcrumbs) and add more keywords.
Currently natural language queries are supported only for CSV data sources; we'll enable it for databases a bit later.