Welcome to the 11th edition of ‘Most Trending Articles’!
Once again, we’ve selected the best (HR) analytics articles of this month. I put HR between parenthesis because two of this month’s articles are about more general analytics. However, I am sure you will appreciate these articles as they provide valuable lessons to anyone interested in HR analytics. We’ve got an article about the impact of AI on the workforce, a business case for workforce planning, and more.
#5 Workforce Planning That Really Is Strategic
For most organizations, strategic workforce planning is not strategic as it usually focusses on the next 12 months. For example, strategic workforce planning would look at where the banks’ services are growing the most, taking expected increases in demand into account. The organization would then plan on hiring more account managers where the forecasted gap is highest.
Having the right people at the right place is just a means to an end – the end being better business performance. One has to consider the broader ways the organizational system creates the customer experience.
To do this, a tool like job analysis could help. By diagnosing the job design one can identify and fix the bottlenecks to performance. The second, vital element is to take a holistic view of the organization.
Alec and Alexis describe these elements using multiple examples in the healthcare sector. A must-read when you want your workforce planning to be more strategic.
#4 Artificial intelligence will create new kinds of work
This article, published in the Economist, describes how the way we work today, is changing. A big part of this is caused by algorithms and AI.
Although this article is not about HR analytics in specific, it does bring up a number of good points in how digitalization is changing HR – and, more importantly, why.
Algorithms are often seen as a stable ‘thing’, that, once in place, replace any need for humans. This is, however, not always the case. Especially when algorithms are used to make difficult decisions, they need to be fine-tuned. This is where we come in.
The algorithm (or AI) compares situations with its database with past occurrences. When a situation has an easy answer, the algorithm doesn’t hesitate. However, when the situation is difficult and ambiguous and the algorithm is not certain of the outcome, it can ask humans to help.
In the article, these exceptions are illustrated using multiple examples including how social networks do content moderation. To read this and other examples, check the article here.
#3 Six reasons why People Analytics will be central to the future of HR
People analytics is shifting to the center of HR. According to David Green, in the next three to five years analytics will become a core and central discipline of HR.
There are six elements that drive this shift:
- People analytics is the center of a digital HR agenda
- Those that aren’t good at people analytics are not ready for AI in HR
- People analytics underpins organizational design and new work models
- New and emerging data sources will help companies improve competitive advantage
- People data needs to be put in the hands of the people in the business that need it
- There is some serious talent in the people analytics space
Using multiple examples, David shows how each of these drivers makes the adoption of analytics in HR increasingly likely. Analytics is no longer a good idea. It’s a core component of future HR and absolutely critical. It’s no longer a good idea, it’s mandatory.
Read David’s full article and all the examples he uses here.
#2 With Advanced Analytics, It’s People (Not Data) That Stand in the Way of Change
In a very relevant and interesting read for anyone interested in HR analytics, Bain dives into the requirements for embedding data and analytics in an organization.
This is hard. 30% of executives say they lack a clear strategy. Of the 70% that do have a strategy, many will lose because of one reason: People. The article explains in what key ways the execution of this strategy goes wrong.
The adoption of a data-driven approach in an organization requires a well-defined strategy. By bringing together people from different part of the company and by enlisting critical leaders (or sponsors), change can be made – and made to stick.
In addition, success should be orchestrated and adoption should be measured relentlessly. This is illustrated below.
The article continues to give five reasons why analytics adoption is slow and what organizations should fix in order to become more successful. To read these reasons, open the article here.
#1 Machine Learning Vs. Statistics
When someone asks me the difference between machine learning and statistics, I often jokingly say that machine learning is statistics on steroids.
Machine Learning is Statistics on Steroids
There’s some truth to this. Machine learning is much easier to understand to someone with a solid background in statistics as many of its basic principles are the same. According to the article, Robert Tibshirani, a statistician and machine learning expert at Stanford, once called machine learning “glorified statistics”.
The big question is: what’s the difference between the two? That’s what this month’s top article dives into. Keep in mind while reading the article that machine learning doesn’t exist without statistics: the latter is the basis for the techniques used in the former.
You can read the full article here.
But why end at #1? While researching this article I came across a presentation on visual analytics by Prof. Dr. Bart Baesens at KU Leuven University. In his presentation, he walks you through the analytics cycle and shows you how visualizations can help in delivering your analytics results. Check it here (pdf). Below we attached some images from the slides.
This last visualization is an example of an outlier. As you can see in Bart’s presentation, yellow colors indicate a p-value between .1 and .05, orange colors indicate a p-value between .05 and .01 and red colors a p-value that’s smaller than .01.