April 01, 2015

Measure What Matters

Talent analytics and big data are hot topics right now. At DCR TrendLine, we’ve been talking about different aspects of HR analytics, big data, and workforce intelligence since 2013. Research conducted by Bersin by Deloitte in 2014 found that the HR analytics software market grew by 17 percent in 2013 and is over $5 billion in size. With countless conferences, books, articles, and software focused on HR analytics, it’s easy to forget what’s really important in this arena.

What is Talent Analytics?

Talent analytics is the application of a defined methodology and integrated process for improving the quality of talent decisions with the end goal of improving organizational performance. It involved both descriptive components and predictive components. Descriptive components include headcount, time to hire, workforce demographics, and turnover, which is typically put into context by using external benchmarking data. Predictive analytics, on the other hand, seek to pinpoit the unique aspects of an organization’s environment that drive business outcomes.

Talent analytics helps answer important questions about an organization’s workforce including:

  • Reasons for turnover in particular areas
  • How to improve productivity
  • Skills gaps in the organization and how to fill them
  • The effectiveness of worker orientation and onboarding programs
  • Predicting the right people to hire for specific roles
  • Predicting and managing attrition rates
  • The accuracy of performance rating systems
  • How to attract a more diverse workforce

Using Talent Analytics

Talent analytics have become the go-to approach when leaders need accurate statistics or data-driven predictions in order to better business decisions. All areas of the HR function have  use for talent analytics, including recruiting, onboarding, training, succession planning, retention, engagement, development, and compensation. Talent analytics projects can be used to understand and link measures from the workforce to key business indicators.

Common Metrics and Big Data Myths

With talent analytics trending in a wide range of arenas, many companies are starting to invest in big data analysis to sift through their volumes of data from various sources including financial data, mobile data, transactional data, behavioral data, social media data, and more. However, the popularity of the topic also leads to managers blindly trusting some commonly heard myths about big data and metrics.

Big Data is Big: It’s a common myth that big data is huge, but “big” is misleading; big data is actually diverse. Big data is a large volume of data points that are updated at high-frequency in real-time from various sources. It is very granular and composed of lots of very small data.

You Need to Apply It Now: Countless articles on talent anlytics would have one believe that to truly gain value from metrics, companies need to act right now. However, the analysis of big data is difficult and better approached with small steps, starting with very specific objectives.

All Data is Good Data: Many companies assume that all the data they have is useable. But there is a distinction between lots of data and lots of good data. Poor quality data has a vast number of errors, and is often missing key details. To make sense of data, analysts sometimes have to throw some of it away. To truly analyze data, it is important to figure out which data is really the good data.

Metrics Give Concrete Answers: Analytics experts know that ambiguity is the dominant characteristic of big data. The more data that there is, the more likely there is to be contradictions and ambiguities that require resolution. Big data requires human judgment to intervene and resolve seemingly conflicting evidence. The true value of analytics is about combining, weighing, and judging multiple sources of information and different analyses.

Anlytics Can Answer Any Question: It is possible for analytics to answers a huge variety of questions about an organization, however, the questions have to be phrased the right way. Applying analytics with a lack of precision or detailed hypothesies can lead to inaccurate answers.