COVID-19: A Guide to the Global Economic Impact

A customer visits your company’s website for a few moments. They click on a link or two, find what they were looking for, and move on. Now, what have you learned? In the age of Big Data, you’ve learned quite a bit.

A decade ago, you might have learned your customer’s geographic location, which browser helped them navigate their way to your site, and which search engine they used to find you. Slightly more recently, you could have also learned which social media sites they were on and what other products they’d searched for.

Today, if your customer performed their online search using their smartphone, you might be able to tell which of your physical store locations they had visited or were near when they searched for your site, as well as the type of phone they used if they pinged the store’s Wi-Fi. If they interacted with your customer service, you’d also have access to that conversation, whether audio or text. And all of this data would have been produced by just one phone.

The mass of information currently being gathered from internet users’ day-to-day activities is astronomical. Most people and households have multiple internet-connected devices, including routers, smart TVs, fridges, smart HVAC systems, and home computers. Yet none of these even comes close to producing as much information as is created by the regular use of social media networks such as Facebook, YouTube, and Twitter. If someone uploads a photo to a site, a company might be able to accumulate not just that photo but also the metadata within the photo, such as the type of camera used, the location, and the time of day it was taken.

Now multiply all this information by millions or even billions. That should just start to acclimate you to the scale of Big Data. Next, add even more breadth and depth to the information being gathered. In today’s world, access to the internet is so ubiquitous and continuous that Big Data is less about finding, gathering, and accumulating data than it is about organizing, analyzing, and extracting value from it. So, how does one begin to understand such a vast topic?

 

How to Think about Big Data

To start, we need to “un-buzz” big buzzwords such as Big Data. It, like artificial intelligence (AI) and other “new” business technology terms, refers merely to a tool. Big Data can be used intelligently or poorly, either allowing a company to harness previously unused, or underused, assets or drowning it in so much information that it can’t see the forest for the trees.

Put simply, the practical application of Big Data through Data Science accelerates both the creation of various types of information and how that information can be used to make decisions with increased precision. This information includes the following:

 

• Texts and transcripts, such as messages, social media posts, and blog posts
• Media such as photos, videos, and audio, along with the metadata within such media
• Geolocations, driving directions, pings from smartphones
• Demographic and personal data
• Other passively generated information

 

Everything people do on the internet creates information, and that information can be recorded, stored, organized, and put to use. Now we also have the growing Internet of Things, in which physical objects and products can be embedded with internet-connected sensors, enabling them to continuously collect data for companies to study and use. As a result, everything we don’t do on the internet seems to create info as well.

The data collected is often divided into three categories. Structured data is organized data that is labeled, tagged, categorized, and contextualized. Unstructured data, which is the vast majority of data and the most difficult to deal with, consists of things such as text messages and photos, which require technology such as natural language processing and deep learning to label and organize. Semi-structured data has aspects of both. For example, emails can be categorized by date and time but include text that can’t be organized without processing.

The total volume of information, in various forms, has become too large for any one person, any large team of people, or even basic computer programs to tackle. Addressing, organizing, and making use of it requires the use of AI programs and sophisticated algorithms. The term Big Data, therefore, refers not just to the enormous data sets involved in the field but also to the way those data sets are used.

 

Three Basic Components

The most common aspects of Big Data are usually characterized as “the three Vs”: volume, velocity, and variety.

 

• Volume: The sheer volume of data is so great, it is almost beyond human imagination. Terabytes of data are common, and petabytes are not unheard of. Because collating and organizing it by hand is impossible, automation must be implemented.
• Velocity: If a business’s IT department was simply handed an immense amount of data, it could find a way to organize and analyze it. It would take time, but it could be done. One of the great challenges of Big Data, however, is that the data keeps coming. Information is constantly being generated, and the rate of generation is always increasing. Managing Big Data is less like jumping in a lake and more like trying to redirect a great river that is flowing faster and faster.
• Variety: Big Data encompasses everything from numbers to words, sentences, photos, videos, audio, geolocation data, and so on. Having a variety of data also introduces all the links between the different kinds and pieces of information. Some of this is easy—a birthday is a birthday—but some of it is very difficult, such as using image recognition software and algorithms to identify and categorize the content of uploaded photos and videos.

 

Two additional aspects of Big Data that are not always included with the three Vs but that must often be addressed are value and veracity.

Data is an asset. It has a value, and it can be used and exploited. Some types of data are much more valuable than others. And because of the considerable volume of data available, determining which data is the most useful can be difficult. Veracity measures the “truthfulness” of data. For example, does the demographic data of your customers actually reflect real life? How old is your data, and will it change quickly over time, losing its utility before you have a chance to dig into it?

Inaccurate data can skew analysis results, leading to bad (or at least poorly informed) business decisions. So do not allow your data to grow stale. If you have access to a continuous stream of new data, a good strategy for dealing with it is real-time analysis, which describes processing of analyzing data as soon as it enters the underlying database. Users then receive insights and can draw conclusions immediately or shortly after initial measurement. This process can save your company from making decisions based on outdated information and help it stay on the cusp of change, allowing for more constructive predictions and planning for the future.

 

Using Big Data

As overwhelming as the idea of endless information can be, a good Big Data strategy can be invaluable to an innovative business—like looking into a crystal ball. Analysts might examine (or use algorithms to examine) a massive customer service data set and find that a significant number of customers are asking for the same thing, thereby providing ideas for timely product strategies and responsiveness to customer needs. The data could reveal trends in customer demographic changes that can be of great value to a company’s future business strategy.

With respect to customers, the ability to examine larger patterns of customer service interactions can help a company gain and retain customers. Many businesses now use AI to address customer needs, such as via smart chatbots capable of answering basic questions or providing appropriate recommendations when customers search for products. A strong Big Data strategy can determine the AI’s overall effectiveness through linguistic analysis of the “tone” of customer interactions, allowing a company to address any gaps in its customer service.

Big Data can also address significant in-company issues. Patterns in data can help a business discover inefficiencies in its organization or logistics. In fact, almost any department in which a company wants to increase efficiency can benefit from the application of Big Data. Analysts can identify patterns associated with fraud or waste and advise the company on how to excise or reorganize departments to improve operations and outcomes.

But Big Data is not just for business. Predictive analysis, for example, allows analysts to recognize patterns in large data sets related to disease prevention, crime statistics, and scientific research. These patterns can then be used to make intelligent predictions about risks in the spread of a disease, the prevention of crime, and the automation of certain repetitive research tasks.

As stated, Big Data is a relatively new tool for business, one that when used right, has great potential.

 

The Difficulties of Big Data

Creating a Big Data strategy and policy for a company can be challenging. The mountainous volumes of information involved require serious storage capacity and maintenance. Although storing digital information is becoming cheaper and cheaper, particularly in the age of cloud computing, it can still be significantly costly. The related analytical software and necessary technological investment are not inexpensive either, and all must be regularly maintained and updated.

In addition, working with Big Data demands highly specialized training, knowledge, and expertise, and as any HR professional will tell you, hiring the most qualified person for the job is difficult. Incorporating a new team, or multiple teams, for a full Big Data policy implementation can be a long, demanding process.

Finally, Big Data, like so much that is technology-related right now, is changing at a rapid pace. Investing in one type of Big Data software or creating an in-house team to create and implement a new policy can be risky. You don’t want to find out that the software is obsolete or the policy is dead in the water before you even got a chance to benefit from it.

For many companies, the best option is therefore to outsource their Big Data–related efforts. A long-term partnership with an outside specialist can be extremely useful because such an individual can examine a business’s needs objectively and craft an approach that fits the company’s overall strategy without the complexities involved in creating a whole new department. In short, an outside agency can hit the ground running.

 

A Bright, and Risky, Future

As powerful as Big Data is, it is not without real risks. Anything created by people almost always includes the limitations of those who developed it. Data sets concerning people’s information can often reflect inherent human biases and prejudices, however unintentionally. A layer of oversight must be part of any Big Data strategy to keep these kinds of issues from decreasing the accuracy of data analysis and complicating strategic policy choices.

To be absolutely clear, companies must not let themselves be ruled by what their data tells them. The public impression of many companies, such as Facebook and Google, is that they are controlled exclusively by data. This perception can do long-term damage to a brand.

Furthermore, one thing that is of immense importance is that data gathered from your customers, clients, website visitors, and other sources be securely protected and ethically used. Every few months, another large company announces a major data breach that has left the personal records and identifying information of millions of people out in the open or for sale on the dark web. This kind of incident can be incredibly damaging to a brand.

 

Conclusion

Plenty of legitimate criticisms of the Big Data paradigm exist, and the implications that come with its adoption are still emerging. Location and demographic information can be used to better address customer needs, but they can also be used to violate someone’s privacy.

Big Data is a new tool, and tools can be misused. It has massive potential to make everything—business, logistics, healthcare, science—more efficient and effective. Therefore, any business implementing Big Data policies and practices must never lose its human touch.

 

AUTHOR’S BIO:

Sameer Somal is the CEO of Blue Ocean Global Technology. He is an internationally recognized speaker and a subject matter expert witness. His passion for learning and sharing information underpins how he helps people make more informed decisions.

 

Website: https://sameersomal.com/

Twitter: @blueoceangt

Linkedin: Sameer Somal