Tag Archives: sentiment analysis

Do HR Organizations Really Need Big Data?

Big Data Funnel
Big Data photo courtesy of hrringleader.

Over the past couple of years, the hottest trend in enterprise technology has been the evolution of HR applications from basic benefits, payroll, and workforce management to a variety of applications covering a much wider set of needs – from finding potential employees and predicting the need for specific skills to optimizing employee capabilities to the appropriate offboarding and succession of core talent. To get a strategic advantage, HR departments are being asked to use “Big Data” to “Moneyball” and quantify their approaches. But is this really the right way to go?

To determine this, consider what Big Data really is. Big Data is a set of technologies designed to store, process, and analyze data that does NOT fit into traditional spreadsheets, databases, and other basic structured data sources. Depending on whether you need your data to be faster, bigger, or more varied, you may be looking for specialized high velocity messaging solutions, cloud-based and multi-tenant storage, high performance analytic engines, Social Network Analytics, sentiment and natural language processing, or video analytics. This is the world of Big Data. A practical definition of Big Data would be 5+ terabytes of data, including some aspect of machine data, interactions, video, or high velocity streaming data.

In this context, is HR looking for solutions that can hold hundreds of terabytes of data and provide real-time analysis? Or are you really looking for some analytic insight for existing data? Chances are that you need a little bit of both depending on the maturity of your current HR data, the maturity of your current analytic capabilities, and your appetite for new and emerging technologies. But you should start with basic analytics and traditional data discovery tools before talking about Big Data.

For your current HR data, you have probably invested in some level of HRIS system to keep track of your employees and define your cost or profit centers. In addition, you have some level of workforce management to track hours worked and to seek what level of resources you have in-house to carry out key projects. This data, in and of itself, is already a wealth of data that is often underutilized to provide greater insight for your organization. As a basic example, consider that all of your employees have usernames and passwords for one or more applications, including email, ERP, CRM, document management, marketing asset management, and many other potential use cases. What happens when you start tracking your application usage on an employee-specific or department-specific basis? You may find out that you need to build a new office, or at least invest in new application licenses and network bandwidth, even before employees fully realize for themselves the extent to which their own work may be hampered or degraded.

Organizations that have invested more deeply into HR, training, market research, and succession management may have even more data regarding the skills of their employees, the ability to benchmark internal talent against readily available talent, and the relative volatility of performance for each position. The value of this data can quickly be increased by integrating tactical decisions and communications with higher-level KPIs to seek how the best employees work or if there are fundamental gaps that internal employees have relative to the market. This approach can also detect potentially troubling aspects in the organization, such as if employees start losing key skills or performance as they continue to work in an organization, or if they fail to learn skills at the same rate as the general market. This may indicate a need for cultural or development changes within the company.

It is important to realize, though, that these approaches are NOT Big Data, but a combination of data integration and analytics capabilities that likely exist already in mid-sized to large enterprise organizations. To get insight like this, start working both with IT and line of business executives to build out the potential value of these approaches and take advantage of the business intelligence initiatives that your company has built out for financial and operational insight. In fact, by being the visionary who understands how to bring human resource insights to a higher level, you may get the HR “Big Data” result you are looking for with just a bit of internal resource allocation. And given the choice of Big Results or Big Buzzwords, good HR professionals should choose results every time.

Defining Social Big Data for Human Insight

Social Big Data for Human Insight
Internet map 1024 by The Opte Project is licensed under CC BY 2.5.

DataHive Consulting is focused on the concept of Social Big Data for Human Insight. But what does that actually mean? For us, there are three key components to this phrase: Social, Big Data, Human Insight.

“Social” is about the engagements and transactions that we use to interact with each other. Social media and social networking are the most obvious examples, but we are interested in the entire span of interactions. Email is a social interaction. Telephony call detail records and text messages are also social interactions. Even our interactions with locations and content can potentially be social if two or more people are interacting with the same location. In short, we are focused on personal and shared interactions.

The most important aspect of “social” is in getting good data. In this case, good data is determined by the quality of interaction. Social data will not be useful if employees, customers, and partners are not interacting with each other and getting value. So, from this perspective, the lessons of effective social collaboration are vitally important for maximizing the value of Social Big Data. Garbage in, garbage out is just as true for relationships as it is for data. Because of this, we care about setting up social engagements properly to create business value and to support ongoing social insight through Social Big Data for Human Insight.

“Big Data” is a bit of a marketing play, to be perfectly honest. We do care about the 3 Vs of Variety, Velocity, and Volume, and we work with video, sentiment analysis, predictive analytics, and data mining. But we also deal with the traditional data management and business intelligence tools that can analyze social data. Often, the key to unlocking the value of social data is as simple as connecting the correct data to the correct applications. In cases like this, DataHive will recommend a basic data integration and connection process rather than a Big Data implementation. The goal is to use the correct business intelligence and analytic tools and not to simply add Big Data tools for its own sake.

At the same time, there are specific challenges to Social Big Data as well. Companies trying to gain value from the full Twitter firehose need to have semantic strategies, sentiment tracking capabilities, focused goals, and a strong knowledge of which Twitter fields are more likely or less likely to provide value to those goals. Video analysis may require packet inspection, natural language processing, metadata creation and analysis, video surveillance, and user participation tools to translate gigabytes of video sharing into actual insight. These social data sources are much more complicated than the relational databases and structured data marts that DBAs initially grew up with. When we conduct data-driven work, it is in light of Social Big Data for Human Insight.

“Human Insight” is the most important part of this model. DataHive is not concerned with data analysis for its own sake. In fact, one of our biggest criticisms of the “Big Data” mentality is concept that the supposed “data scientist” would look at data without hypotheses and look for findings. (We know that not all data scientists take this approach by any means, but we believe in an empirical and scientific approach for analyzing data.) In our world, businesses and organizations have specific goals that they are trying to achieve and Social Big Data needs to help them to improve processes, provide guidance, and foment disruptive innovation.

Although social media is typically in the hands of marketing or customer service, DataHive believes that there are Human Insights that go across the entire organization: human resources, sales, research, operations, IT. The insight may be a simple as picking up website complaints which lead to the realization that the corporate WAN is being impaired at a key node. From an HR recruiting and morale perspective, the sentiment for your organization may be much stronger or weaker online than your internal metrics may track. And the pure insight and educational value of key thought leaders online can provide value to every level of the organization if the correct information is used.

But beyond the sentiment of the social media itself is another level of insight held through metadata, social connections, and context provided through sales, talent, and other key business capabilities. Data has a topology and a specific meaning, especially as we try to collaborate together. Social efforts can be pointless if we are not talking to the right people or if we lack a critical mass of users. There is also a bigger world of human insight associated with global challenges, but we are starting with business challenges. Today, businesses, tomorrow, the world!

As we look at human insight, we want to show you how to gain value from all three of these areas both in isolation and in combination. We use our decades of experience in using social media, our use of metadata and relational data, our social science backgrounds in understanding the sentiment and psychology of social data, and our business backgrounds to provide you with greater insight into using social data.

If you have questions on what social data is or how to take optimal advantage of your social data, we would love to discuss your social strategy and social data use cases. Please contact us at connect@datahiveconsulting.com to schedule a free consultation.