Five Things You Always Wanted to Know About Analytics But Were Afraid to Ask: A Skimpy Primer for Today’s HR/ Human Capital Professional
Don’t worry! This is NOT another article about analytics predicting your toothpaste or moving everyone to a higher plane of being.
I don’t believe every last bit of the hype and neither should you. I do, however, see abundant evidence that analytics can drive improved individual and organizational performance.
Unfortunately, many tend to treat analytics as something akin to nuclear physics and best avoided by anyone but the experts.
As a first step to overcoming this unnecessary barrier, I want ask and answer five fundamental questions that people typically have about analytics but are afraid to ask. Once you understand these foundational concepts, you will have a new way to understand analytics, how it works, and what it can truly provide for you, your colleagues, and your business. You won’t be an analytics guru just yet, but the good news is that you don’t need to be a guru to use analytics more profitably. Let’s get to it.
1. What exactly does “analytics” mean?
The answers will differ depending on who you ask, but in the context of human capital and HR, the core of analytics really comes down to this: a systematic analysis of available data to drive strategic and operational talent decisions.
Here, “systematic” just means using the same core methods for collecting, analyzing, and visualizing the data. If the process is not systematic, then it will be difficult to interpret the data in a way that makes sense. It will also be impossible to replicate the analyses with new data.
Let’s take the simple case of measuring job satisfaction. It is important that the measures (typically survey questions) and the kinds of people responding remain the same over time. If the questions or the participants systematically change from one administration to the next, troubles abound because you have no way to really know whether job satisfaction has improved, declined, or stayed the same. As obvious as this sounds, this is a common issue.
Changes in the basic analysis and visualizations can also wreak havoc with interpretation. Subtle shifts in a sample, the presence of a few extreme values, or even changes in the scale of a plot can lead to strongly different conclusions even when there are no underlying “true” changes over time. Again, consistency and replicability are key to an honest and interpretable analysis.
(Note: A future post will tackle the all-important “strategic and operational talent decisions” portions of the definition in future posts. For now, we will just focus on the analytics piece.)
2. What are the different kinds of analytics?
I find it most helpful for beginners to break down analytics into three layers:
Level 1- Descriptive analytics: The bottom and most critical layer is “descriptive analytics”. This is best summarized as “What happened?”. Average time to fill a position? Turnover and retention rates? Performance measures for customer service representatives in the last quarter? All of these and many others fall under Descriptive Analytics.
Note that “What happened?” doesn’t just mean what happened last week or last month. It might also mean what happened the month before and the month before that. Said differently, understanding needs context. Good descriptive analytics provide that context to tell you what really happened. Looking at measures over time instead of isolated snapshots can help separate the signal from the noise.
Level 2- Relational analytics: If descriptive analytics means asking “What happened?”, relational analytics means asking “What else happened?” As its names suggests, relational analytics is about finding informative relationships in the data. Instead of simply reporting on average performance for a given role, relational analytics asks how other factors such as education, experience, previous performance, or specific leaders are related to that performance.
Likewise, looking at annual turnover might be informative, but using relational analytics to reveal that millennials are quitting at a 43% percent higher rate in a given business area is another thing entirely. In both these cases, we are using relational analytics to find relationships that lead to additional questions or possibly suggest a targeted course of action.
Level 3- Predictive analytics: Predictive analytics asks “What will happen?”. Ignore the jargon for now and just remember that predictive analytics really tries to predict only one of two possible things: a category or a number. Predictions want to say “I know your kind” (category) or “I got your number” (number). If you can remember that, then you understand the fundamental goal of predictive analytics.
3. How does predictive analytics work in practice?
In a nutshell, data scientists take a subset of data (called the training set) and train a computer using “machine learning” algorithms to predict a category (“I know your kind”) or a number (“I got your number”). Then, they test that model on a new set of data (called the test set) to see how well the model works with examples it has never seen. If this model provides accurate predictions for the new data, then it is a good predictive model.
Suppose I am interested in predicting which registered nurses are likely to stay in their current job (stayers) and which are likely to quit (leavers) within the next year. This is a categorical, “I know your kind” prediction.
To do this, I will first select some measures like years of experience and patient satisfaction that might predict whether a nurse will stay or quit. If I have selected the right predictors, then my model should first learn to successfully predict the stayer/leaver status of the nurses in my training data.
But even if my predictive model works really well on the training set, I still don’t know how accurately it will predict “stayers” and “leavers” with a new set of data from a different group of nurses. To be useful, my model needs to generalize to new cases, not simply learn the training examples.
As a follow up test then, I will this same trained model to see how well it works on a new set of new examples in the test set. If the model still successfully predicts “leavers” and “stayers” with the new examples in the test set, it’s a keeper. If not, we start the process over again and try to construct a better model. Predictive modeling for the “I got your number” scenario works the same way.
4. What do data scientists and analytics types really do all day?
Popular accounts of analytics in action typically project an image of an all-knowing, all-seeing team of nerdy engineer-/data scientist-types surrounded by huge, figure-filled displays, churning out models that predict when you will wake, eat, drive, get sick, die, or get the mail.
The unsexy reality? Analytics people spend upwards of 80% (yes, EIGHTY PERCENT) of their time engaged in “data wrangling.” Also known as “data munging”, this is essentially taking raw, messy data (think raw HR survey results in different Excel files) and converting that mess into tidy columns and rows of properly formatted data appropriate for analysis.
Conceptually, data wrangling is often like an old-fashioned “cut and paste” session in Excel, only with a few million rows of data instead of a few dozen. Replace your computer mouse with some programming that does the cutting and pasting for you according to some set of rules and, voila, you have “data wrangling”.
(Note: There are many cases where data is unstructured and the goal is not necessarily to pack it into rows and columns, but the bulk of the data in HR and Human Capital is of the rows-and-columns nature.)
5. I don’t do analytics. Who cares?
Even if you don’t officially “do” analytics in your day-to-day role, there is a constant flood of data that directly impacts your decisions, those of your leaders, and the bottom line. If you can process, interpret, communicate, and informatively challenge this information, your value will grow.
Conclusion: Next Steps
First, the next time you see some data just ask yourself “What level of analytics is this?” This simple step will help you quickly see the critical kinds of descriptive and relational insights that are not being satisfactorily addressed or even considered. It will also open your eyes to the predictive possibilities. If you are a more informed and challenging analytics consumer, you will be a more valuable member of your organization.
Second, realize that you are now in a better position to truly collaborate with your analytics colleagues who don’t necessarily know what you are looking for. You might not speak the same language yet, but you have at least eliminated some of the mystery surrounding “analytics”. Work with them, them talk with them, explain your talent and business problems, and let them leverage both your expertise and theirs to uncover the insights you deserve.
Third, demand more, not less from your HR/ Human Capital analytics team. And by more, I do NOT mean just more reports. I mean thinking through a problem with them, asking what the outcomes might reveal at the descriptive, relational, and predictive levels. Finish by asking yourself “What would the next action be?” to insure you are digging into actionable stuff that matters.
Like this post?
Get our FREE Turnover Mini Course!
You’ll get 5 insight-rich daily lessons delivered right to your inbox.
In this series you’ll discover:
- How to calculate this critical HR metric
- How turnover can actually be a GOOD thing for your organization
- How to develop your own LEADING INDICATORS
- Other insightful workforce metrics to use today
There’s a bunch more too. All free. All digestible. Right to your inbox.
Yes! Sign Me Up!
Comments or Questions?
Add your comments OR just send me an email: firstname.lastname@example.org
I would be happy to answer them!
photo credit: <a href=”http://www.flickr.com/photos/37826605@N05/4300107736″>24/365</a> via <a href=”http://photopin.com”>photopin</a> <a href=”https://creativecommons.org/licenses/by-nd/2.0/”>(license)</a>
- © 2022 HR Analytics 101