“Some people look at data and see integers, booleans and strings. When I look at data, I always wonder what the story is behind it.” – Stefan Groschupf, data scientist.

Context makes data useful

Like Stefan, I see the stories behind the data. This is what makes data interesting, what makes big data relevant and what has created my passion for ensuring quality data. [Tweet this]

And yet, this reality has been largely ignored by early big data vendors.

The early premise of data science:

Early proponents of data science suggested that skilled data scientists could deliver new insight simply by applying complex algorithms to vast data sources. This big data myth has been largely debunked. This is not science!

Value requires a hypothesis

Science requires a hypothesis – a business question to be asked and answered – and that the hypothesis is proven (or disproven) on the basis of examining the empirical evidence.

For most businesses, the value of big data comes when the business management and analysts are able to ask the right questions. Complex algorithms can then be applied, to vast amounts of data if necessary, in order to test these hypotheses.

Shortage of skills

The problem with big data analytics is that it has traditionally required companies to invest in scarce technical skills.

A lack of these skills and the technical complexity of early big data technologies – such as R, Python and Hadoop – has left many organisations struggling to reap the benefits that big data promises.

In Data Science Lies, Sex and Videotape and How sexy are your business analysts? I suggested that big data analytics will fail to excite mainstream support until your existing analysts and managers are empowered with self-service big data management tools. 

Tools that manage the end-to-end complexity of taking on and integrating big data sources – from Google Adwords, to web logs, to the CRM application, to the mainframe, and everything in between -, that provide intuitive and familiar interfaces for complex analytics and visualisations, and that allow the user to focus on the story rather than on the technology.

Big data requires data management

Big data suffers from all of the data management complexities of traditional data – arguably more so.

The real-time nature of many big data applications – from fraud detection to real-time price adjustment, to providing relevant tokens in-store – means that business users need to be able to adapt on the fly.

An overwhelming trend in BI over the last five years has been towards self-service – no less so with successful big data. [Tweet this]

“Data360 Analyze provides a level of standardization, a level of predictability, and a level of analysis that frees us up to provide a better experience for our external and internal customers.”

Executive Director of Finance, Comcast

When you empower the people in your business that are passionate about data with tools that allow them to integrate complex data sources, perform complex analytics and almost instantly see the results you create data science. [Tweet this]

I am one of those empowered people.

I am a self-service data scientist. You can be too! [Tweet this]

For more information about how self-service big data analytics tools can empower you, watch the video or contact us.

Image sourced from http://en.wikipedia.org/wiki/Albert_Einstein