Posts tagged ‘HR analytics’
Storytelling is rightly hailed as a must-have competence in people analytics. In my own competency model, it is one of the six core competencies any analytics team must have. Other models do the same. Compelling arguments are being made about the value of good storytelling. In other words; master it or beat it.
So don’t get me wrong; it is important. But my point in this post is that storytelling requires the presence of a theory to be successful. If you do not have a proper – i.e. a plausible and documented – theory behind your data, storytelling can do more harm than good.
Angela Duckworth observes in her book: Grit – the power of passion and perseverance, that “a theory is an explanation. A theory takes a blizzard of facts and observations and explains, in the most basic terms, what the heck is going on”. I could not have put it better myself. And funnily enough, this is also what storytelling is doing – explaining what the data says.
Let me give you an example why you need a theory to tell a story: ZengerFolkman – an excellent US data-driven leadership development consultancy company – has compared the combined leadership effectiveness scores as measured on 360-degree evaluations for men and women respectively at different leadership levels. The result is, as you can see below, that women score better than men at all levels and that this difference is more significant the more senior the leaders are.
I recently made the same observation within a financial institution. They had collected performance data for all their leaders and we were comparing performance data – split into different KPI groups – and it was clear that the performance rating was significantly better for the female leaders and also that difference was greater the more senior the leaders were. The data at this company confirmed the international data I had found. I had data and I had other similar data points to back them up.
So far so good.
The problem is, that although the difference between performance scores is significant the data makes little sense without a theory to explain the observations. Why are women leaders rated better than men? All we know is that the performance ratings/360-degree evaluations put women higher than men. It may be that women are better leaders than men. It could also be that women are reported to be better leaders but in reality are on par with men. Maybe there is a bias in the evaluation of female leaders. Or it could be a third reason.
Another thing to consider is the relationship between the portion of female to male leaders vs. overall performance. Is it linear or does it have another shape as depicted in the figure below? If it is linear and you conclude that females are better than male leaders, then a natural recommendation is that you should replace all male leaders with female. If on the other hand the relationship has some other shape – such as the one in the second figure below – you should identify the optimal point to reach leadership effectiveness.
My point is that without an answer as to why there is a difference you cannot create a story and a recommendation. To come up with a proper recommendation you must have a proper theory to explain the why. The basic analysis cannot explain it and you cannot go straight to storytelling because you are still left with the basic question of ‘why’. And what you will be left with are leaders sitting around a table wondering what to do. In this case, maybe there is a good theory. I don’t know of it (but would love to hear it if you happen to have one).
So you need a theory behind your data. An explanation if you will. It does not need to be verified by Harvard or any such institution. But you do need an explanation. Let’s say that you find that the talent you source from one university performs significantly better than the talent you source from another. You need to understand why. If you cannot explain why through a theory, your storytelling will lack the power it has the potential to have.
So: please do not do storytelling on people analytics without a proper theory explaining your data. It really makes no sense.
Succeeding with workforce analytics is difficult. It requires a mix of skills not found in one person only, and you should not assume, that you can do it on your own. We are all decent at most things but really only good in a few. You should therefore assemble a team, which has a multiple of superheroes each with a superpower of their own.
I described this in a previous post, where I suggested six competencies a superhero analytics team should have:
- Strong data management skills
- Captivating storyteller
- Understand the business
- Ability to visualize your results
- Strong psychological skills
- Excellent statistics and numbers skills
But what happens if just one of those skills are not present? Can’t we manage anyway? My answer is no. If just one of the skills is missing from the team, six outcomes are possible – each with a disastrous outcome – as shown in the figure below:
In essence, if you:
- have no good data, you will not be able to perform analytics. It is as the old saying goes: crap in – crap out. If you do not have good data, it is sometimes better not to do analytics.
- lack of storytelling abilities, the message will nog. As Tom Davenport describes: “Narrative is the way we simplify and make sense of a complex world” and it is the way messages are most effectively conveyed and the best way to get people to change (which is the ultimate goal of analytics).
- have no business acumen will mean that your team will perform excellent analytics on the wrong issues. Workforce Analytics should help decision making on vital must win battles for your organization. Understanding the business is vital to understand what those must win battles are.
- are not able to do visualizations you will bore your audience. Data and numbers are boring (and I am a numbers guy), but data and numbers effectively conveyed through visualization
- lack psychological skills you will misunderstand your findings, be unable to convert your information to knowledge and be subject to important challenges such as bias, cognitive dissonance, imposter syndrome etc.
- have poor numbers and statistics abilities, your analysis will just be plain poor. You can get really far with simple regression-, factor- and t-test analysis skills but at other times, you will need skills in more advanced statistics when the data set become really big or you are looking for more predictive analysis.
Analytics require a lot of skills and abilities – superpowers if you like. The best way to ensure that you have the right ones to deliver on your task is to assemble the best team. An analytics superhero team.
My second take-away from the workforce analytics case-studies and conferences I have heard, attended and experience over the last year is what I call the confusion of cost savings and value creation. While the good news is that we are starting to deliver, my warning would be, that we should be careful not to deliver on the wrong things – or more important; on the least value added things. Let me elaborate.
At most conferences and in most reports by leading consultants, we are being presented with a maturity model, which illustrates activities from the least mature to the most. It goes something like this; first there is some descriptive methods, such as reporting and trend analysis, then maturity increases and the methods goes on to being predictive and prescriptive and finally the maturity goes on to machine learning or something like this. One example of such as maturity model is from IBM shown below (but frankly they all look very similar).
I fully agree with the idea of maturity and that prescriptive analysis is better than descriptive. It is also a good way to illustrate this maturity journey albeit they could be a little more operational in terms of assessing level of maturity and suggested next step depending upon current level. However there is one dimension missing from this picture: the focus of the analysis itself. All good at being mature of the methods but we must also assess maturity on the object of our analysis.
In rough terms: If it the focus is on cost savings elements then the potential shareholder creation will always be limited (it will be the net present value of the cost savings minus the investment). If the focus is directly on creating customer value/business value then the potential shareholder creation will be great.
In fact, I will propose, that there is more value added in doing predictive analysis on a business matters than doing prescriptive analysis on an HR matter.
To be clear, let me come with a few examples. If you are analyzing sickness, employee turnover, recruitment effectiveness or training effectiveness, you are really at the cost savings end of the spectrum. There is no harm (at all) in coming up with evidence based suggestions to reducing employee turnover. Indeed for many companies there are significant money to save in doing that. It is however still cost savings and it won’t get you a seat at the table. So do some of that, but don’t put all your efforts there.
At the other end of the spectrum, you are adding workforce data to customer/profit/sales/other business data. Here the examples are less generic as they are (should be) tailored to each company’s specific strategy and situation. A few I have witnessed/been part of: Finding which service behavior adds the most impact to customer experience/satisfaction, and which training programs are most effective in embedding this behavior. In this example, the workforce data leads straight on to more sales and higher profits. Another example; how does change load (employees’ load of change relative to ability to handle change) impact strategy execution.
These two specific examples had a heavy use of non-workforce data as part of the analysis. In fact, you can test your value maturity on the cost/value axis by testing how much business data you have compared with how much workforce data. If you only work with workforce data, you are probably focusing on cost savings rather than value creation.
Some will sometime argue that “Attracting talent is always business critical and therefore what we do is value creating”. That may be true in some cases but they are missing the point. Indirect value creation is important but less straight forward to prove. In most cases they misunderstand HR processes with business matter.
I therefore suggest that we add a dimension to our maturity models. Perhaps some large consultancy company can show how this may look?
Over the last year, I have with interest read and heard a lot of workforce analytics case-studies both at conferences, in network groups and in companies by practitioners. And I find myself hearing good news. I believe, we are as a profession starting to deliver on our promises and heading towards a brighter future. Let me elaborate.
Two years ago, Workforce Analytics was in my view in grave danger of over promising and under delivering. And that is a recipe for failure and extinction. At conferences, Google and other large wealthy US-based companies were showcased as the promised land and of what to come in the near future of HR Data and Workforce Analytics. Airtime was primarily given to these companies and to vendors who were trying to over-sell software capabilities and the picture drawn was that not only was workforce analytics adding a lot of value, it would finally bring HR to the promised land: To the table.
But reality was far from that picture. Bad data, incompatible software, data illiteracy and a lack of a data-based strategic mindset was rather the norm than the exception. And worse, it left practitioners disillusioned when leaving conferences as they knew that their maturity level was far from what had been presented at the shiny conference.
Two years later and things look different. The promises made at conferences, in case studies and at presentations within companies are more realistic. Workforce analytics will not radically change HR – what it hopefully will is to improve the decisions we make in HR. And what’s better, we are starting to deliver. The data has been cleaned (somewhat), the software has been installed (to some degree) and is working (sometimes) and a data mindset is creeping in within most HR functions now. Armed with this, real projects with tangible benefits are starting to show. Nothing major. Nothing fancy. Just credible analytics projects delivered by HR to the business.
So having been a bit of a worried pessimist on behalf of the workforce analytics community, I now find myself being optimistic. If we continue to promise less and deliver more, who knows, perhaps the business will start to listen.
In my last blog post I gave five reasons why HR Analytics should not be located in HR. The five reasons I gave were: 1) It will lose its strategic potential, 2) HR does not have the capabilities – and will not be able to attract the right ones, 3) HR does not have ownership of all the relevant data, 4) Efficiency gains from pooling all relevant capabilities together and 5) HR is not credible enough to work with analytics.
Fair to say that while some agreed others did not. However, that post was only the prosecutor’s arguments. Now it is the time to ask the defendant to rise and present his side of the argument.
In this blog, I want to present the case from the other side’s point of view: 5 reasons Why HR Analytics must sit in HR:
- Nobody else care about HR data. If a central BI or Analytics unit should do all analytics work in a company my guess is that HR issues would be prioritized pretty low. In fact, it is not that many people outside of HR that thinks that engagement data can be used for much more than a pie-chart report once a year. However, in HR we know that engagement can drive business results. We also know of many ways to use analytics on engagement data. It takes an HR mind to see this and to care for this kind of data. Therefore, the more positive re-wording of this point might be that HR data has a lot of potential value; HR people are the only ones who know this and can get the most out of HR data.
- It takes HR knowledge to interpret HR data. Data is fine but where it really has an impact is when it is converted to information and later to knowledge. Computers can to some degree make these conversions but at some point they must be converted and interpreted by a human being. This is not easy and interpreting HR data requires HR knowledge. Leadership profiles, performance data, engagement survey data and many other HR data can only be fully utilized if the HR department helps. While a steering group with HR capability may go some way to resolve this problem, I would argue that either the meetings with the steering group must be very frequent or HR Analytics will have to sit in HR to get the required frequent interpretations of data and findings.
- It may make HR more data driven and improve HR impact on business. With some exceptions, I think it is fair to say that many HR departments do not have the influence in the organization (I believe) it deserves. There are many reasons for this; lack of strategic focus, difficult to quantify HR’s impact on the business results, a lack of basic business acumen and finally I often hear that HR does not speak the language of business. While I have always disliked the last argument, I hear it a lot. I think it has to do with not being able to make proper business cases, identify the strategic link, provide some kind of evidence to support the suggestions and quantify the value of the initiatives. I also think it is fair to say that HR is not very evidence-based. Do we really use the best knowledge available to design our practices? HR Analytics can help to achieve this to some degree. If used properly this will without doubt in my mind help HR gain credibility and impact on the business.
- Data ownership sits naturally in HR. Much of the HR data needed to conduct HR Analytics sits firmly in HR. It is always a problem if the analytics department does not have easy access to the data required. Many times the HR data is located on different IT platforms than the rest of the business’ and being located near the data makes access easy. HR will also understand the legal aspects of the data much better than anyone else (with the exception of the legal department of course) and this will help the HR Analytics team avoid running into serious legal issues.
- It will increase the likelihood of the analytics actually being used. HR Analytics is really about creating knowledge to make better HR decisions. HR decisions are many times made in HR by HR and therefore it makes sense to make the analytics department sit close to the people who are actually going to use it. It will make it more likely that it will be used and have the impact intended.
Did I forget any obvious good reasons to place HR Analytics within HR?
In this and my previous blog I have presented the two cases; 5 reasons for and 5 reasons against placing HR Analytics in HR. In my next blog post, I will discuss the two sides with Peter V.W. Hartmann who is the Lead on HR Analytics at Maersk Drilling and see if we can get to some kind of overall conclusion. So stay tuned…
If you were to build an HR analytics department where would you put it in your organization? The obvious answer may at first be to put it in HR. But at second glance this may not be the right place. Maybe even the wrong place.
I believe there are five good reasons why HR Analytics (or Workforce Analytics or HR Data or whatever it may be called) should indeed not be placed within the HR department.
- It will lose its strategic potential. It is common knowledge that HR Analytics has far greater potential if it is directed towards strategic issues rather than operational and tactical ones. To put HR analytics in HR will create two problems. The first is that I often see HR Analytics focus on improving HR processes only. Nothing wrong with showing the link between engagement and employee turnover rates, it is just not very strategic. Nothing wrong with improving the training programs with analytics – it is just not very strategic. Tactical at best but most often operational. The strategic mindset is often not present. The second reason is that HR is mainly strategic when it is working outside of HR’s own silo and instead fronting the business and the front line. Some even argue that it is not HR’s role to be stategic. Putting HR Analytics deep inside some (often random) HR function makes the decision making process far removed from the business – and hence strategic – side of matters.
- HR does not have the capabilities – and will not be able to attract the right ones. I have argued that to succeed with any kind of analytics you will need to attract a number of different capabilities to take charge of your analytics effort. These capabilities, competencies and skills include i) being excellent at statistics and numbers, ii) strong data management skills, iii)) captivating storytelling skills, iv) visualization skills, v) strong psychological skills to understand terms such as bias and heuristics, and vi) the ability to truly understand the business. Very few in HR master these and fewer yet have a strong team which complement each other enough to be able to build the right team. Likely they will fall short on data management skills, visualization skills, statistics and understanding the business. And worse, those people are not looking at HR job ads. Trust me.
- HR does not have ownership of all the relevant data. Many of the data which HR has ownership of (recruitment data, performance data, engagement data etc.) needs to be combined with data which reside in other parts of the organization such as finance (payroll), IT, legal and most importantly all the customer related data. These may also just be basic things such as master data. Many times HR Analytics people do want to work on strategic matters but need data which they are not allowed access to for internal reasons. In my experience these are often customer and profit data. If HR Analytics resides in a function outside of HR these data may be more available.
- Efficiency gains. Purely from an economic and efficiency point of view, it does not make sense to have several small teams scattered around the organization trying essentially to do the same thing; namely to do analytics. Instead pull the people together, let them build on each other’s experience and competencies, save money and efficiency by having one big analytics department instead of one in HR, marketing, IT, business units and where else they may be.
- Credibility issues. If HR has not been using data well in the past but instead has created and submitted poor business plans made more by gut feel than by use of good data, and if HR has produced endless of meaningless data reports on absenteeism, employees turnover, sick-days and percentage of women in workforce with questionable data quality, and if HR generally argues by what feels right rather than what is right, then the quality of the analytics work will be called into question just simply because it comes from HR. HR does not have the credibility to work and argue with data. It may require someone from finance or business intelligence to produce and conclude the HR analytics for it to be taken serious.
If you take this perspective where should you then place the HR Analytics people? One approach would be to create a central business intelligence unit where all the company’s analytics will take place. Another approach is to create one or more center of excellence so much of the BI capability can be decentralized to the business units. And there are many other ways.
But maybe HR Analytics should not be placed in HR.
Overconfidence is a term used in psychology to describe a person bias for being right. The overconfidence effect is a well-established bias and describes a person’s subjective confidence in his or her judgments is reliably greater than the objective accuracy of those judgments, especially when confidence is relatively high.
Daniel Kahneman writes in his epic book ‘thinking fast and slow’ that “neither the quantity nor the quality of the evidence count for much in subjective confidence. The confidence that individuals have in their beliefs depends mostly on the quality of the story they can tell about what they see, even if they see little…our associative system tends to settle on a coherent pattern of activation and suppress doubts and ambiguity”. What this means in practical terms is that one would assume that
How is this relevant to HR Analytics you may ask? In many important ways. Imagine that you present a fantastic ‘Predictive Employee Turnover Analysis’ to your boss who is the head of HR. You have collected a ton of data, used complex algorithms and found the following conclusion: female leaders at level 3 and 4 who have been performing among the 10% best leaders (at those levels) for three straight years and have not received a promotion are 60% likely to leave the organization within one year. Wow. And based upon this analysis you suggest a targeted effort towards that group of leaders. You calculate that this effort alone will save the organization more than $3m a year.
Your leader is truly impressed and now ask you how confident you are that your result is correct. That this group of employees are that likely of leaving and that they effort will save that much money. This is a fair question because your boss has to make a decision based in part on your findings. What do you respond?
According to the theory of the overconfidence effect, you are likely to say a very high number. Perhaps even 90%. In any case, the number will be much higher than an objective accuracy of likelihood. Frankly this number is much lower than we think it is. And this is a problem because your boss then makes a decision based on wrong input from you. If he thinks you are right, he will underestimate the risk of the project – the risk of being wrong – hence estimating a too high ROI.
Why will you make a higher estimate of the correctness of your finding? According to the theory and what I have experienced many times in reality is that because when we work with data over a stretch of time, as such a project mentioned above would imply, we are slowly developing a story in our minds. A working hypothesis if you will. And slowly we are only finding data which supports our hypothesis. This is a natural inclination. We confirm our hypothesis when we work with data.
The solution to this problem, which is quite a significant problem; always have a steering group for your projects with many types of people, continuously present your findings and assign someone to be the devil’s advocate and make it that persons responsibility to ask the stupid questions and challenge your beliefs and hypothesis’.
Overconfidence is by the way only one of many biases we carry with us and which work against us. A couple of others such as cognitive dissonance, illusory correlation, availability and anchoring. They are subtle but important and possible devastating to good analytics.
My main message here is as follows; we would like to believe that because we work with data and facts we then automatically make decisions that are more rational. We do not. We have a fantastic ability as human beings to disregard facts and make decisions based on irrational biases. If we can overcome this natural tendency then working with data and facts will make our decisions so much better. Exponentially better. But we must continue to work on our psychology.