What's New?


Big Data Conversations with Dominic: Data Exploration

Posted by Dominic Barton

30-Sep-2014 15:30:00

Does anyone feel that there are quite a few people talking about Big Data but no one really understands it? We’ve seen lots of analysts discuss the importance to companies and talent acquisition functions in particular but for the most part these are high-level conversations that don’t give the reader any real, actionable information (a bit of irony there if you ask me).

At Broadbean, we’ve distilled years of business conversations about our workplaces and the people who are employed there, and we've created tools that not only answer the questions big data brings up, but also shapes the future questions discussed over donuts at long conference tables. 

In these forthcoming articles, I’m going to talk about some of those conversations and show how having the right perspective can drastically alter your business for the better.


Let’s get started then...




Data Exploration: “How can we use the data we already have?”

So many bad big data conversations start this way. Not because it’s the wrong question but because once this question is asked everyone gets excited and loses sight of two very important things:

1)   Data is often a hot mess.

It’s true. Data is messy, it’s not easy to put into similar, searchable or comparable formats. It’s difficult and sometimes very boring work, which means it’s often foisted on an intern, an admin or an overworked project manager. When and if that person is taken from this understaffed and underfunded project, all the parameters they put around the data are lost until someone else picks up the thread and the data is corrupted in the name of organization again. In the meantime, data continues to pour in from multiple sources. It’s a never-ending cycle. In fact, enterprises store 80% of the world’s data.

2)    No one agrees on WHAT precisely is relevant.

Because of point #1, no one really knows what they might find in their big, dusty stacks of data, so everyone votes for the findings they’d like to see, using gut intuition or worse, pure hope. This not only wastes time if and when the data ever comes back but it wastes everyone’s time in nearly every company-wide meeting, bickering over what needs to be proven in the data and what is a lesser priority. This colors the findings in the end as well, as any scientist worth his or her salt would predict.

Of course, that doesn’t make the question “Can we use the data we already have?” irrelevant, it just makes that conversation fraught with potential potholes. Here’s how to steer this conversation in the right direction.


If the answers are important, then the data must be

treated with the same importance.


While you might have multiple systems to integrate and messy data points to contend with, handing it off to someone who isn’t equipped to manage the very large project of classifying and integrating large amounts of data, both incoming and historical, simply isn’t fair. Invest in the proper tools and staff to make the project worthwhile (and actually finish it this time?)

This may seem like a hard sell, but in reality the amount of time you’ll save in both man-hours and the sheer amount of information you’ll have gleaned once the project is completed and chugging along. The White House administration is investing $200 million into big data research projects. 


broadbeantwitter-01   The White House is investing $200 million into big data research projects. @BroadbeanInc


It’s worth it to have a realistic conversation about the (time and actual) costs associated with having someone with limited time and resources working on a project of such importance versus the (time and actual) savings associated with having a tool that can assess, sort and clean the data and provide near real-time baselines that answer the questions your leaders are asking on a weekly, monthly and quarterly basis. Even a 10% increase in data accessibility (which is what we’re talking about here) translates to an additional 65.7 million in net income for a Fortune 1000 company.


broadbeantwitter-01  Only 23% of organizations surveyed have an enterprise wide #bigdata strategy. @BroadbeanInc


Agree on what precisely is relevant. Does that seem simplistic? It is a little bit but this is putting the cart before the horse. What is relevant in talent acquisition is rarely given as much import as it should, especially in a people-powered economy like the one in which we find ourselves. Only 23% of organizations assessed had an enterprise wide big data strategy. It’s also worth noting that rarely do people expect any other business processes to integrate multiple systems in the way that talent acquisition regularly must. A frustrating by product of the innovation we see in our industry is that lack of streamlined functionality. So the first step to agreeing what is relevant is finding a system that will create a clean landscape of data, so you and your stakeholders can decide just that. 

New Call-to-Action

Topics: big data @us , Big Data