In this special guest feature, Scott Parker, Director of Product Marketing at Sinequa, discusses the evolution of data search and analytics techniques and how modern technologies are helping employees find the right data faster and spend less time searching through irrelevant content. He also discusses how information-driven organizations are harnessing big data to improve operational efficiency and accelerate innovation. Scott began his career as a software engineer and systems analyst with Bloomberg BNA. While at BNA, Scott earned a graduate degree in software engineering from Carnegie Mellon University and then went on to become a senior director at Vivisimo where he spearheaded the implementation of the company’s go-to-market strategy.
As someone who makes a living in the world of data analytics, I see more organizations reaching a tipping point of just how much data they can comprehend and act on. And it’s not only data – it’s information and insights that are overwhelming their ability to make proper business decisions.
According to a report by this publication, we can expect the size of the digital universe to double from 2010 to 2020, as human and machine-generated data grows 10 times faster than traditional business data. As a result, we’re becoming rich in data and poor in insights. I like to call this a cognitive burden.
This incessant tide of information and insights that flows from ubiquitous access points and collection methods is overwhelming both business and IT. And because of it, we’re missing opportunities that can lead to tremendous success or horrible outcomes. The question isn’t about “how” much information to capture and analyze but “what” is actually pertinent to an organization’s overall goals.
We can change the way internal business clients select, absorb and process information by helping them look at it from a holistic sense that transcends beyond the mere numbers.
Think beyond the “data” point
As a first step to alleviating the cognitive burden, let’s take a step back from the term “data” when talking with our internal audiences. When I hear the word data, I immediately think of structure, which we all know accounts for just 20 percent or less of the content we analyze. Potentially valuable content comes from all manner of sources such as videos, online reviews or recorded conversations.
Before planning any analytics initiative, IT has to engage with the business to understand and agree upon potential content sources. It’s incredibly likely these sources will be disjointed and disparate, which drives the need for contextualization to produce connections and insights that serve business goals.
Speak the same language
By moving beyond the term “data,” we can help drive the conversation around relevant content sources. So, for example, if the project is designed to improve customer service, then the initiative should cover a combination of Twitter posts and Yelp reviews in addition to quantitative data like delivery times and the number of minutes customers might wait on hold for support.
This combination of structured and unstructured data sources results from having a broad conversation with the business client. Some of them might be unaware that you can collect and analyze the tone and dialect of social media conversations through Natural Language Processing (NLP) or that you can scan and evaluate recorded conversations over tech support calls. As the data professional, you have to help the client understand that nearly any type of information can be leveraged. The challenge is in understanding and qualifying in advance what is important to them.
Understand the critical areas of knowledge work
Part of this conversation includes advance homework to understand the information sources business clients use at work. If it’s a pharmaceutical or healthcare-related company, then we know it can be a combination of drug tests, X-rays, MRIs and patient records. If it’s legal, then the work could encompass surveillance videos, depositions and DNA.
Take the time to identify the day-to-day work and ensure the client understands that content from their daily sources can be collected and analyzed in conjunction with other sources. Let them know it’s entirely possible to produce insights from as many sources as necessary, as long as it’s relevant to their goals. Look at their activity logs and usage logs to get an idea of what people are going after and the systems they’re using.
Discourage data hoarding
Walking a business client through potential content sources can be a little like going to the grocery store while you’re hungry. At some point, everything looks good. Content hoarding innocently starts with an excessive acquisition and a reluctance to delete electronic material that is no longer valuable to anyone. It stems from a variety of individual traits and habits, corporate conditions and societal trends.
Some individuals experience anxiety when facing the prospect of disposing of digital items, especially if they fear losing something important. Many content hoarders don’t know how to organize their digital content or lack a means for determining which content is worth keeping. Volumes of material naturally grow from email alone. And because the content is digital, many don’t perceive it as clutter.
As data professionals, we have to counsel business clients on two fronts when it comes to potential hoarding. First – what information are they legally required to maintain and for how long. Second – is it really relevant to future projects? It’s highly likely this information can still be kept somewhere, but on less expensive storage that doesn’t need to be quickly accessed.
Convey technical capabilities
While discussing content sources and analysis goals, it’s important to help internal clients understand the full capabilities of technologies like machine learning and NLP. Many of them might not be aware that NLP can cover more than 20 languages spoken by 95 percent of the world’s population, in addition to the ability to factor in images, video and audio. Nor are they thoroughly versed on how machine learning algorithms can provide collaborative filtering and recommendations for deep learning capabilities to analyze and structure content. Introducing them to the art of what’s possible could uncover new content sources that they previously had not thought relevant to their goals.
Helping internal clients focus on the information that truly matters and can be analyzed will significantly reduce the overall cognitive burden that obscures new insights and discoveries. Our responsibility as data professionals is to help them focus on what matters and what’s truly needed to help them achieve their goals.
Sign up for the free insideBIGDATA newsletter.
Speak Your Mind