Five Thoughts That Encourage RH to Start with Workforce Analytics

Embracing the growth in data-analytics remains an HR’s attitude more than a technical question. Based on the experiences of many IBM clients who are now realizing the benefits of cognitive analytics (artificial intelligence), we listed do’s and don’ts to aid your first steps with your cognitive analytics journey.

1. Don’t be deterred by size.

When HR departments procrastinate with analytics, it’s usually the “too big, too small, too late” syndrome. If you run a pay check every two weeks, it’s enough to start an analytics study. You can start small and still discover some great insights. A smaller scope will help you to learn without major risks; you will be able to develop a comprehensive understanding of the context, the science, and the story you are writing.

2. Do not be deterred by complexity.

The challenge in analytics is no longer about the data, it’s about the speed with which you can act based on your data insights. Speed is the new normal in turbulent environments and agile organizational cultures. The results of analytics may suggest actions that are easy to implement, they may just as likely suggest actions that require courage, patience and significant change management to address. Understanding the potential actions at the start of the analytical investigation is important, after all what is the point of analytics if the actions the insights suggest could never be implemented?

3. Data nirvana is not about 100 percent accuracy.

Clearly, you need good enough data to have a reasonable basis for effective decision-making, but new cognitive computing can provide confidence level information to better inform your insights. Of course, better (and perhaps more) data will likely be your next step, but, in the meantime, there are ways to reduce variance and bias problems before modeling, complexification and generalization.

4. Start small, think big.

As cognitive systems can rapidly analyse large amounts of data, it may be tempting to want to go big with the analyses too. But, at the beginning, it is better to look for low hanging fruit. A pilot program with a smaller project scope reduces political, technical and change management risks. As you learn, you can incrementally add new projects with more expansive or sophisticated analyses.

5. Do start where it matters most.

To gain momentum for analytics, start with what is important to your organization and think about how analytics could help. Question your business partners regarding their business strategy. The low-hanging fruit for analytics are closely related to your strategic goals. Organizations that start work on high-return analytics projects tend to see the most success[i]. In particular, they tend to focus on the following:

  • Revenue and growth challenges. Think sales performance or customer loyalty improvements. What factors drive high-performing sales professionals and high customer satisfaction? What factors do the opposite?
  • Capacity of delivery and quality of service challenges. Focus on employee retention. What creates high levels of engagement and retention and who is likely to leave soon?
  • Workforce cost and trust challenge. Address safety issues. What factors are likely to generate accident claims?

References

[i] IBM (2015) Starting the workforce analytics journey. The first 100 days. IBM Smarter Workforce.

[ii] Discover how one of our client is transforming HR (and Business) with cognitive analytics.

[iii] Learn more with Watson Talent Videos on YouTube Channel

[iv] See Watson Talent Insight demo video

— — — — — — — — — — —

Jean-Baptiste Audrerie is organizational psychologist, M.B.A, executive advisor for IBM Kenexa & Watson Talent in Canada. He is author of the HR blog FutursTalents. He guides clients into digital and cognitive transformation for talent attraction, acquisition, development, and engagement.