Wednesday, July 15, 2015

Three Aspects of Talent Analytics

The goal of talent analytics is not to get the report built, but to capture business insight holistically.

The role of talent management is to put the right people in the right position at the right time for both maximizing business value and unleashing talent potential. Therefore, talent analytics is an important tool to analyze business/talent growth trend, practice data-based performance management, and do workforce planning more scientifically. But more specifically, how can you frame the right questions and solve them via data analytics, and how can you leverage technology to transform raw data into metrics?


A good place to start Talent Analytics is to question: Identify a problem and work from framing the right questions -- probably without spending a lot of money. For example, you might go to leadership with the question "Do we know what impact retirement of top talent is going to have on our business in the next five years?" or "Do we know if the staff who are leaving are low performers or high performers?" If you can find a clear business question they care about, then you can come up with an approach on how to answer it. Once you've done the reasoning, then it will be natural for the firm to invest in analytics so that it can do more inquiries to dig through from both a strategic level and tactical level: What is your strategy and data planning, and how will you use it? Who are going to be the customers or consumers of your reports and information, and what format do they need it? What data do you have, in what format, where the strengths and weaknesses and in what state is it in? What are some of the business problems or issues you are tackling? What is your IT infrastructure and what resources do you have and not have? What reports does the business need rely on? and what is your current state of development in analytics/evidence-based decision making? What is your budget? When having some clarity around these questions, you should be in a better position to judge whether you would like to utilize in-house BI systems, get an on-premise analytics tool and build internal data management capability? Or whether you would like to go down the Analytics as a Service cloud offering. They usually come with a suite of basic measures and, more importantly, be ready to deploy fit-for-purpose data framework to organize and access your workforce information.

The goal of talent analytics is not to get the report built, but to capture business insight holistically. The problem is far more broad and relates to the need to source, combine clean, structure and present an increasingly complex set of metrics and outcomes to more and more stakeholders in the organization. It is not just about getting the report built - it is about getting insight to the right user in a way that supports a better decision about talent life cycle. The basic data collected but not analyzed is futile; good low hanging fruit is attrition analysis; if you could establish a strong trend of attrition among a set of the level that is crucial, you would get enough points for proposals to build you case for HR analytics. So delivering a high level of impact to your organization is not about finding a way to replace the old analytics system. It is about looking for an enterprise level solution that cleans and combines data from multiple sources, supports the secure sharing of data, presents data in an "end user-friendly" format, reduces the manual work required of the analytics team and ultimately delivers insight to your organization.

The key to predictive analytics is not just making predictions, but actually showing the real ROI from what you implemented. An important consideration for firms implementing HR analytics (whether using data that are "Big" or not) is how to appropriately document the business reasoning behind modeling decisions that drive the outcome. There are too many variables to indicate how predictors can be used to drive future people processes; therefore, it will depend on context. For example, say you discover that the fixed personality traits neuroticism, extraversion, and conscientiousness reliably predict employee performance in a certain role. Since personality is a relatively stable attribute, it is not useful for development because you can't easily modify personality through training or coaching etc. The practicality of analytics more lies in job fit, succession planning, and recruitment. The first type of focus is on the prediction of 'fit for purpose,' as in attracting the right type of candidates in recruitment by researching the type of traits you need to succeed in a job. The other type of model is used for predicting the effect of an intervention, like introducing incentives for certain functions. Now the goal is changing the behavior of employees. In this type of model you need, is to establish a causal relationship before you can move forward. So before setting out on creating a model, you must agree on the general possibility of an ROI, it can not be just an academic hobby.

Overall, it is a journey to build a solid analytics capability in HR and in the whole organization. Regular discussions about HR data, reports, definitions, and dashboards should go a long way in creating a sense of awareness. Having the data isn't worth anything if HR is not conscious enough to provide the context and interpretations of the data. There is a positive shift around the business and beyond, to build a culture of analytics and manage talent and business as a whole more effectively and intelligently.


0 comments:

Post a Comment