HR Metrics: From Data to Results

Today I had a really nice lunch with a friend who is a Director of Human Resources.  We ate at Mahoney & Sons, right down by the water at the edge of Coal Harbour.  It’s a sunny day today; perfect for enjoying lunch on a patio with views of Stanley Park and the north shore mountains.

Among the many things we talked about, we discussed some of the merits and pitfalls of “HR metrics”.  For the benefit of those readers outside of HR, HR metrics are often toted as the means by which to show the relevance and legitimacy of the HR function.

As a bit of background, HR continues to languish in this struggle to achieve the status it aspires to, among other areas of the organizations of which the functional group is a part.  Common complaints include:

  • HR’s lack of understanding of “the business”
  • A focus on transactional administration, rather than providing strategic leadership and advice to the business
  • A reactive, rather than proactive stance
  • An inability to communicate how what HR’s costs do to drive top line and/or bottom line financial performance

A popular article on this topic can be found on the Fast Company website: Why We Hate HR.

So, during the conversation with my friend she shared some stories of her experiences implementing HR metrics when she first assumed her role as the leader of the HR function.  As the Director of HR, she reports into the Chief Operating Officer and wanted to demonstrate her team’s ability to be a value adding partner to the organization.  A truly great idea!

However, she quickly realized that while the metrics she was able to capture and report were interesting, there was very little that could be done by HR to impact these measures.  This brought me to share my thoughts on the trajectory from data to results with her – and I thought I would take a moment to share them with you here as well.

Here’s how I see it…

1.  You start with DATA

This is the raw material you are working from.  You need to consider what data to capture, why this data matters and what it might tell you, who and how it will be captured, stored, and kept clean.  This could be qualitative or quantitative, however, the goal is typically to have quantitative HR metrics.

2.  You then analyze the data to arrive at INFORMATION

By looking for trends and patterns, doing statistical analysis, and comparing to other data sets you can start to arrive at new insights.

3.  You can use this information to make different DECISIONS

In most cases managers are looking for information that will allow them to make better decisions.  Or a cynic might suggest, managers are just looking for information that will justify the decisions they’ve already made.  Either way there is an opportunity for the data to tell a story when it becomes information – a story that can (1) increase the effectiveness of a decision or (2) decrease the effectiveness of a decision.

4.  Your decisions are sometimes followed by ACTION

Robert Kegan, Harvard Professor and author of Immunity to Change was in Vancouver recently to speak at a BC Organization Development Network event.  He started his talk, which you can see here (Video: An Evening with Robert Kegan and Immunity to Change), with a question: “If 14 frogs are sitting on a log and 3 decide to jump off, how many are left?”  Dr. Kegan encouraged us to resist the temptation to respond, “11!”  His point being that there is a big difference between deciding to do something and actually doing it.

In an organizational setting this is extremely common.  This issue is often labeled as an “inability to execute strategy” or a “failure to implement change”.

5.  Your actions will lead to RESULTS

Your actions will definitely lead to some kind of results.  Let’s hope they’re the kind of results you were hoping for!  When they’re not, we kindly call them “unintended consequences”, but really the outcomes may be disastrous in some cases – many executives have lost their jobs based on the results achieved being different than the results desired.

Linking this back to the topic of HR metrics specifically… if you are gathering HR data (e.g., turnover ratios, absenteeism rates) you should be aiming to be able to do two main things:

(1) move the needle on the HR results – to influence them in the desired direction

(2) demonstrate the link between moving the needle on these HR results and moving the needle on business results (e.g., revenue, profit) in the desired direction

How do you see it?


Lies, Damned Lies, and Statistics

Yesterday the closure of the downtown HMV ‘flagship’ location was announced.

Oddly enough, over coffee earlier in the very same day a friend and I had been discussing the historic and current challenges of brick and mortar music stores.

Listening to the radio, this morning, News 1130 reported on the HMV closure. The host invited listeners to visit the News 1130 website to answer their daily web poll question: “do you still buy CDs?”

As of roughly 9:00am, 63% of respondents had clicked “no”.

Do you see any possible problems with the validity of this number?

Here is my perspective:

People who are frequently on their computers, tablets, or smart phones are the voters. Simply by knowing they were able to vote online, we know they have the technology and internet access required for listening to music via streaming radio and music downloads.  However, the people without this technology and internet access are also more likely to rely on other methods for accessing music and therefore are more likely to be CD buyers – but, of course they are less likely to vote on an online poll.

It’s like going to a bowling ally and asking the people there, “do you still bowl?”

Another way to skew the figures about consumer behaviour, when it comes to CD purchases, is to ask the same question, “do you still buy CDs?” as people are walking out of an HMV holding an HMV branded shopping bag.

Either way, the sample is not reflective of the range of people who listen to music.  Similar critiques have been made about the results of studies in Psychology that have only had the readily available college student population as a sample. When we then assume we can extrapolate these conclusions about human behaviour to the general population, we are making a big and potentially dangerous leap.

News 1130 cited a Communications Professor at SFU who says 1 in 3 people currently buy their music in CD format. He shared that people now buy through a wider variety of channels and also in multiple formats. Some people like to buy certain albums on CD, while their main listening may occur via a SirusXM satellite radio subscription. Therefore a binary, yes/no question such as, “do you still buy CDs” hides this level of granular detail. I have to admit, I am even left wondering about how the 1 out of 3 figure was determined.

There is a relevant saying often attributed to Mark Twain: “There are three types of lies: lies, damned lies, and statistics.”  I encourage you to spend 6 minutes to watch Sebastian Wernicke’s amusing TEDTalk on Lies, Damned Lies, and Statistics.

This is just the tip of the iceberg. We could go into much further analysis, such as discussing the use of the loaded word “still”, which for many people will suggest a negative judgement about your ability to keep up with current trends, if you answer “yes” to the question, “do you still buy CDs?”

Do you have a healthy scepticism about the statistics you see and hear every day?