Datafication of Compensation Distribution

Is your data science providing you enough indications that challenge your existing compensation strategy?  Does it reveal that the art of compensation distribution performed by your managers is not in accordance with your compensation strategy? Old habits die-hard, so you need to make sure that your plan for data-driven decision-making is not getting overridden by compensation managers’ belief system and they are not ignoring data science recommendations.

Effectively distributing compensation using limited budget in your disposal is always challenging. Your strategy can be to give performers much higher increases than rest of the people as it may still result in improved engagement and retention as concluded by some research results.

Your compensation system should empower you with actionable insights by providing sharper data analysis on some of the critical decision points like performance, potential, risk and social score of your employees. Performance analysis primarily depends on internal data, potential and risk analysis depend on internal as well as external data and social analysis (how person is perceived or represents company outside workplace) mainly depends on external data. While internal data remains structured, external data (e.g. tweets, social media, blogs etc) combines both unstructured and semi-structured data. Your data scientists need to extract insights and value by combining these internal and external data sources to generate a real value for your business and to datafy your compensation process.

By unlocking the true value of your data, effective data analysis provides you unambiguous decision points. As a result, you may even automate compensation distribution process in cases where you are not looking for any diversion from your planned strategy and where there is no scope for your compensation managers to add value in the process.

In a nutshell, you have to leverage your employee data, available inside and outside your organization’s firewall, to enhance data-fueled decision making as it will surely result in improved business performance and will also help you to attract and retain best talent.

Metrics for Compensation Allocation

Simon would like to allocate performance bonus to his workforce who got performance rating of 5/5 and 4/5. People who got 5/5 rating will get (X) % of their base salary as performance bonus where as 4/5 will get (X-1) %. He would like to define compensation metric for the same so that he can apply it, with his decided % amounts, on his entire workforce rather than allocating to each one of them. He would also like to share it with his peers or subordinates in case they would like to use same metrics with their own numbers.

Neither the use of organization wide published metrics is new to compensation process nor is the desire of a manager to build his own metrics for distributing compensation. Only question is whether your current (or potential) adopted compensation solution supports it or not.

As a manager, you would always like to create metrics that you can use to allocate compensation to your team as it will facilitate a smooth process and help you in well-informed decision-making. From ages, managers are using off-line tools to flag and store decision attributes which helps them in making compensation as well as other decisions. What they actually need is a system which not only allows them to build their own metrics based on various person related attributes but also allows them to share it with their peers or subordinates. Manager owned metrics supplements the organization wide metrics (aka HR established Metrics) and not really replace them.

Some managers will be happy if system supports basic attributes like performance rating, work location, Job/Grade whereas others may need more specific attributes like years in Job/Grade, compa-ratio and grade ranges. It will be beneficial to have embedded support for all the possible decision attributes as well as custom attributes (an extension to store business specific values) so that managers can build robust metrics for compensation allocations with great ease. It will result in compensation allocation process to reach the next level.

Can we ever be objective?

CICERO:

Indeed, it is a strange-disposed time;
But men may construe things after their fashion,
Clean from the purpose of the things themselves.
Come Caesar to the Capitol tomorrow?

– from Julius Caesar, Act I, Scene III by William Shakespeare

During this presidential campaign, we’ve been treated to many interpretations of the data provided by a multitude of polls. It’s been Big Data meets Big Auguries.

What can we learn from the ongoing debate about whether poll and their analyses are biased or not and therefore whether they are accurate predictors of events? How can this debate benefit not just how we govern ourselves, but also how we run our organizations?

The point isn’t that polls and their analyses aren’t biased – they are. So are customer or employee surveys. So is the interpretation of sales data. What matters is what are you doing about it. How are you finding out how to correct for the inevitable bias so as to mitigate its distortive effects? How do you pursue objectivity, rather than just dig into your position as the truth or just give up and say objective truth doesn’t exist?

It wasn’t that I didn’t know enough; I just knew too much

Like any other observations, what we construe from polls is influenced by what we already know (or don’t know) and what we believe. To hammer the point home, we see only what we choose to see, consciously or not. Sometimes, there’s simply nothing there; it’s just noise and what you see is an illusion.

Does having more knowledge of the subject help? Knowing more doesn’t necessarily solve the problem – it matters more what it is you do know and that’s hard to know, you know? If that extra knowledge simply aligns with and confirms your prior knowledge and/or beliefs, it just reinforces a possibly misplaced confidence and takes your further from the objective findings.

So, like any other science, we have to understand this problem going in and have a process to gradually reveal what is really going on; i.e. the purpose of the things themselves.

Think twice; that’s my only advice

A good way to do that is to always poke holes in your thinking and the data you gathered. How might your thinking be based on false assumptions? How might the gathered data be misrepresentative? You have to test your assumptions and your data gathering methods continuously.

Thinking twice helps you get out of the cognitive trap our brains are wired for – to go with the first story that fits the facts. Jumping to conclusions is a species survival trait that served us well when we didn’t have time to mull over whether our clan should go out and hunt a mastodon.

Today’s world is infinitely more complex than arranging a hunt for food. So rethink and rethink again as much as you can before making the call. And if you have to make the call before you have sufficiently tested your data and thinking, then look for ways to structure your actions so they both move you torward your goal as well as gather more data and test your thinking. “Continuous Beta” is an example of this.

Bless your soul; to think that you’re in control

Finally, why are you relying on just your view or the “inside view” of your group to question your assumptions and data? As much as you’d like to think you can control you own biases, you and your team’s vested interests, sunk costs, etc. will always influence your objectivity in subtle and not so subtle ways.

As uncomfortable as it can make you or your team, get the “outside view” from other people and teams who will give you unfiltered feedback on your assumptions and evidence. The idea is about improving your pursuit of objectivity.

Get their thinking as well. They may not know as much as you do about the data, or have your level of experience and expertise in ways to interpret it. But they may know things that you don’t that could help you see things in a new, less biased way.

——————————————————————————————————————–

Whatever your political leanings, it’s worth it to check out “The Signal and the Noise: Why So Many Predictions Fail – But Some Don’t” by Nate Silver.

Photo by Crouchy69

Be Careful with Averages (Especially with Compensation)

The Department of Labor put together this chart that compares the average amount spent on compensation and benefits between private and public sector employees.  At first glance, you might think that those interested in high paying jobs should look to public sector employment, or that public sector employees are overpaid.  However, drawing such conclusions from a simple average is premature.  In this case, the problem is that we aren’t looking at pay for the “average employee” across these two dimensions; we’re looking at averages calculated from entire groups of very diverse people.  Nancy Folbre, an economics professor at University of Massachusetts breaks these numbers down into a distribution of earnings in an effort to discredit initial interpretations of these averages, and to come up with some meaningful takeaways from the data.   

Comp is much more polarized in the private sector, where private sector employees are over-represented in lower and higher income brackets, while most public sector employees fall in the middle ranges.  43% of private sector workers earned less than $25k per year and more of them are part time (26%).  More public sector employees are college educated (45% of public sector workers have a college degree v.s. 29% of private sector workers).  The data suggests that employees performing similar jobs in the upper end are paid significantly more in the private sector than they are in the public sector.  And if you’ve got a lower skilled job, then it’s probably better for you to work for your local municipality.

Averages often offer poor and sometimes misleading insight when it comes to compensation reporting.  Too much is lost when data is aggregated.  The fact that the average salary for a US subsidiary is lower than a Mexican one may or may not be a problem; if they are the same, it may or may not be a problem; or, a problem may exist when the average salary in the US is higher than it is in Mexico.  Then you ask yourself, who cares about salary averages broken out by country, or business unit, etc..  I see too many compensation reports that just offer these higher end aggregates and don’t allow someone to look deeper into the numbers to draw meaning.  If you’re going to show an average, then be sure to allow someone to cut that average across multiple dimensions to get to some level of granularity; otherwise, an average is just a tease.

HR: Why Improve Your Analytical Intelligence?

Hey! Come back!

Before you roll your eyes on this one, start having flashbacks to terrible experiences with calculating standard deviations, or trying to wrap your head around multiple regression analysis, and then run screaming from this post, this is not about you trying to become an expert at statistics! Trust me!

It’s about you understanding how analytical tools and methods can help HR have an impact on applying talent to strategic success. Besides, no less than Josh Bersin said at the recent HR Technology Conference® 2009 Talent Management Analyst Panel, “Get used to it.” And that’s a good way to look at it. Too often, HR has been shut out of strategic input because of the perception that it doesn’t speak the language of analytics sufficiently to measure and understand the relationships between various parts of the business (e.g. Human Capital) and profit (or whatever financial result you wish.) Once you have that better understanding, it will enable you to make a stronger case for why HR can provide valuable input and leadership in business strategy and execution.

By now, we’ve had the importance of measuring pretty well pounded in, particularly in the context of Finance. Increasing your financial intelligence is key to participating more in driving strategic decision-making around applying talent to improve business results. Being able to show to senior management the link between what you know about your company’s talent to financial results entails both measuring talent in terms of levels of performance, competency, skills, connectedness, etc. as well as measuring relationships between those measures and the other parts of the business that drive financial performance. What do analytical skills have to do with measuring those parts and their relationships?

Measuring is not Counting

To help answer that question, let’s take an example from “How to Measure Anything: Finding the Value of Intangibles in Business” by Douglas W. Hubbard. Picture the problem of measuring the population of fish in a lake; let’s say in order to know if a restocking effort was successful or not (a good ROI problem). A lot of people will say, “Drain the lake and count the fish.” They could then report there were exactly 22,573 fish and we’ll say that confirms the restocking investment was a success, although all the fish are now dead.

A better approach (certainly for the fish) entails using analytical methods to estimate the population of fish in the lake. If there is sufficient confidence in the estimate of the population before and after the restocking effort, you will be able to tell if the restocking effort was a success or not. Did you have to know every tiny detail of statistics to make a decision based on these estimates and the confidence level? No. How about to show the before and after picture to some “lake executive” who had to give the green light on the restocking effort? No. You just had to know enough about analytical methods to know that the application of them made sense in this case, and either determine you made the right call or get the point across to that executive.

As the authors of “The Differentiated Workforce: Transforming Talent into Strategic Impact” quoted a general manager, “I couldn’t do a regression analysis, but I knew what one was. And the results…made sense to me.” Further, they write, “Improved analytic literacy has a direct impact on the decision making at several levels in a typical HR organization…At the highest level, improved analytical literacy changes the perspective on the financial resources committed to HR…they consider a significant portion [of the HR budget] an investment.”

Principles of Uncertainty

HR labors under the false assumption that everybody else has “precise numbers” and there seems to be a perception that HR can’t come up with the “hard numbers.” The classic story is of the CEO asking the head of HR if they know the company’s headcount and the response is wishy-washy. The thinking is that people are either working for the company or they are not, so what’s the problem? What’s the count? Sure, in a company of a few hundred people, you might actually have a very precise figure. However, we know that depending on the industry, economic conditions, etc. as the number of employees gets larger, it gets a bit trickier to know the headcount with precision. There is a lot going on and even if you are using an HRMS system, the simple fact that humans are involved and entering transactions (or not), makes the number transient and constantly changing. In other words, one minute, you could see 59,268 and a minute later see 59,273.

This is not that different from the folks in Accounting keeping track of Receivables, the folks in Production keeping track of Inventory, or the folks in Development keeping track of Project Completion. In the case of Accounting and Finance, it gets even more interesting when it’s time to report; for instance, general accounting principles require the company to estimate the amount of Receivables that will be uncollectible and there isn’t any hard and fast equation for doing that. Different methods are used to estimate these values, some of them analytical.

The point, as Hubbard writes in his latest book, “The Failure of Risk Management: Why It’s Broken and How to Fix It” is that measurement is better understood as the reduction of uncertainty about the value of something. Once you see it that way and gain enough analytic literacy to feel comfortable with the results from those tools and methods, you’ll be able to move forward more readily with driving and demonstrating positive impact on strategic business results.

Photo by The Michael

The Failure of Risk Management: Why It’s Broken and How to Fix It

Is Bacon at the Center of the Universe?

Not here either.

No, this isn’t a cosmological question regarding pork products, but really about Kevin Bacon and his position in the Movie Universe. Although not at the center, he is closer than a lot of other actors. Understanding the principles behind this can help us find ways to develop talent more quickly and effectively, which benefits both the employee as well as the employer.

We already discussed the principles that show how social networks can help form “weak ties” that foster innovation and breakthrough thinking. It turns out that the book, Driving Results Through Social Networks points out another principle* from the game, “Six Degrees of Kevin Bacon.” It can help shed light on how networks and how they are developed can contribute to the success of high performers. This is achieved by building the right kind of network, which not only benefits the individual (and thereby serves to motivate them to put effort into this activity), but there’s a big payoff for the organization as well with these better-built networks.

Authors Cross and Thomas point out that being more central in a network (where the network is the total of all the people and their connections to each other) typically means having more numerous and diverse contacts and therefore closer access to a greater number and wider variety of information, ideas, resources, and opportunities. Note that it’s the combination of number and diversity that generally lead to this. For example, having a huge number of connections to a very narrow segment of a network probably means the connections are highly overlapping, which limits access to the rest of the network.

By having that closer access, an individual can more effectively tap into the network in order to achieve more than they otherwise would, be it goals, career development, etc. In turn, the company gets more productivity, increased innovation, and enhanced engagement from having employees more effectively connected.

There is a danger in looking at this single-minded. For example, grading everyone on a one size fits all “centrality score” is apt to backfire. How central in a network one is helps some individuals more than others based on their role, for instance. The definition of the network as all the people and their connections leaves open some questions. In some cases, you may not want to include every department in the company, but rather the pertinent departments from across all the business units. For some individuals, it makes sense to include more external networks, like industry groups, along with the key groups within the organization. Other individuals might be very central in a particularly intense area of expertise within the company. Remember that all the members of the network contribute to it in a wide variety of ways and it doesn’t serve any purpose to try to force everyone to be the same – that defeats the very usefulness of the network itself.

———————————————————————————————————-

*Briefly, the principle works like this: while there are a large number of actors, there are hardly any that are more than 3 “steps” away from Kevin Bacon (he is only two steps or less away from almost 25% and three steps or less away from almost 90%). By having so few steps to so many other actors, Kevin is better positioned than the average actor to find out about and exploit an opportunity. Of course, we all know his talent, experience, “look”, etc. all affect whether an opportunity will be opened to him, but a moment’s reflection tells us that these “connections” (to use the cliché) have a big impact as well.

How does this work and how did his network develop this way? Those two are related. By virtue of a combination of the total number of stars with whom he worked as well as who those stars were and with whom they’ve worked, Kevin has a network that reaches relatively quickly to a greater share of actors. This came about by his choices on what movies to star in and/or with whom to work. It’s likely that there is more diversity in the genres, cast, etc. in each of those choices. In contrast, other actors, whether due to type casting or personal preference, had made more narrow selections and their networks are “skewed” towards one area of the network. For example, someone might select for or get typecast as the slapstick comedian or the horror movie queen, and that would restrict the other actors they work with, reducing the share of the total network they have access to, and in turn impact the kinds of opportunities they get.

Photo: Sean Munson

At the time, I thought this was an interesting way to label their crosswalks. It turns out there’s more too it: “On September 25, 2004 Wallace’s Mayor Ron Garitone proclaimed Wallace to be the center of the Universe. Specifically, a sewer access cover was declared to be the precise location of the center of the Universe. A specially made manhole cover was made to mark the spot. It bears the words ‘Center of the Universe. Wallace, Idaho.'”

(Fly to in Google Earth | See in Yuan.CC Maps)

Rich social network = rich productivity

I was browsing through last month’s Harvard Business Review and lo and behold there’s a short article on social networking. The article was about the types of social networking interactions that are required at different times or for different purposes. A centralised structure works for discovery; but a richly connected network supports integration and decision making.
But that wasn’t the important bit! The important bit was recent research from MIT showing the productivity of poorly connected workers versus richly connected workers. Those workers with the most extensive personal digital (i.e. electronic) networks were 7% more productive than their colleagues. Of course, there’s no substitute for face to face, so the same study also found that workers with the strongest and most cohesive face to face networks were 30% more productive.
So I see corporate social networks as places for:
– gathering to share information
– gathering to integrate information and make decisions communally
– building a virtual network that supports and extends the face to face network

Is the bell tolling for the bell curve?

In an entry I posted last year titled, Taking the number out of the equation: Performance evaluations without performance ratings, I extolled the virtues of eliminating the performance rating.  Well actually what I said was “I am willing to accept that assigning a rating value is an easy and (mostly) objective way of evaluating worker performance.  But I can see no need to ever share the rating assessment with the worker (me) – because the rating is not meant for me, it’s just a tool for my manager.”  Assuming, as I did, that the HR department was closely following my posts, no doubt with great enthusiasm, I anticipated my proposal would be implemented that very same week.  Alas, I am still waiting.  What’s more, in a cruel twist of irony or possibly just well deserved Karma, I was recently asked to manage an internal performance review process we’re conducting within the development organization.  I’m still trying to work out the horrors I commited in a past life to have earned this privilege, but never mind – that’s not really what I wanted to write about anyway.  Getting back to the previous post, in the sentence immediately preceding the one I quoted above, I said “I think the whole bell curve model is a pile of horse manure – but that’s a topic for another day.”  Happily, that day has arrived.

 

Over the past year I’ve been contemplating how companies facilitate their talent review meeting.  Central to the talent review process is a box-chart analytic, generally in a 3×3 configuration, which most in the industry simply refer to as the nine-box.  For the uninitiated, here’s an example:

What I like, scratch that, what I love about the nine-box model is the multi-dimensional feedback it provides; helping customers not just to see what’s happening in their organization, but what they need to do to better align their talent management strategy with their business strategy.  The nine-box discussion makes the talent review meeting a true business driver and not just another dead end discussion.  Talent review meetings help companies assess worker engagement, risk of loss, organizational diversity, candidates for succession, and development gaps and they provide a starting point for addressing these challenges as well.  By comparison the bell curve analytic just feels outdated and sadly monochromatic.

 

In the global battle to attract and retain top talent it may turn out that the people you need to succeed are already working at your company; but if you can’t discover, motivate, challenge, develop, promote and compensate them, the battle may already be lost.  Talent reviews are one way for companies to identify, develop and reward both their best performers and their high potentials; but they also help to reveal the underlying reasons for poor performance –  workers who are in the wrong role, who need additional training, who are being poorly managed or under compensated – as well as those who simply need to be managed out of the organization.  The one dimensional feedback provided in the bell curve will never help to surface these critical path issues.  The nine-box, by contrast, offers a multi-dimensional perspective of the organization that can serve as the anchor for the talent review meeting and the cornerstone of a holistic talent management strategy.

 

I’d love to hear what you think about the bell curve, the nine-box, talent review meetings, or any of the other talent management challenges facing your organization.  In the mean time I’m off to lead this internal performance review and hopefully earn a little good Karma in the process.  Wish me luck!

 

Profiles: The Foundation

Continuing Meg’s discussion about realizing the strategic value of integrated talent suites, let’s start with the foundation of Profiles.

The notion of building talent management suites on a foundation of competencies has been around for some time now. Competencies were seen as a natural mechanism for connecting the various talent applications together with a common “currency.”  An employee could be rated on competencies in the performance management application, they could locate courses in the learning management application that would help develop competencies, etc. One problem has been that it’s been very difficult for companies to develop competency models that truly impact their strategic success. As Meg described, the result has been more tactical, talent process automation in nature.

Lately, talent management suites are being built on top of a Profile foundation. The concept of Profiles is rooted in the idea behind competencies, but expands beyond competencies to encompass other characteristics (or attributes) as well. These characteristics can include things like certifications, experience, interests, travel preferences, potential, and so forth. Some look at this as a “fall-back” solution to having trouble in developing a competency model, but another way to look at it is as a way to model more things your talent should possess that matter in your company than just through competencies. If it helps to get things started by simply dealing with education, licenses, and so forth, at least it’s a start. More importantly, folks have also pointed out that attributes can more readily describe, and in a more granular way, what it is that makes a person effective in their role, beyond what competencies alone can do.

With Profiles, a company has a way to know, across the organization, who knows what, who has what skills, certifications, who has what experience or practice, etc. What’s also important and starts to make things more strategic is when a company models what characteristics are required in jobs and organizations and how effective someone can be in that role that has those attributes. We can think of this as introducing a kind of “exchange rate” that helps you understand the meaning and value of the attribute “currency.” Competency Gap Analysis has been around for a while, but Profiles takes things to another level. Having a richer set of variables to compare when searching for someone against a role, or when an individual is looking for ways to develop themselves, is very helpful.

With profiles giving you a way to track what your company’s talent has and describe what your company needs, you have the foundation from which to impact your strategic success. Now you can use analytics to find which attributes really do result in higher performance in a role. Some of these might still be competencies, but you also might discover other attributes that either more directly predict better performance or that demonstrate a positive effect on competencies that in turn result in better performance. When you couple that analysis with an analysis of what roles are “pivotal” in your organization, you are really beginning to get a handle on how your talent can improve your strategic success. Now you are starting to see what your strategy needs in order to be effectively executed. In addition, you can also uncover untapped opportunities to leverage your talent to gain even further competitive advantage. Finally, you can even go deeper and find where the “sweet spots” are in terms of how an attribute impacts performance (and how performance impacts business results). For example, at what number of hours of training, number of projects in an area involved in, level of proficiency in a given competency, etc. do benefits start to level off?

To sum up, Profiles give you a better ability to understand how (and where, and how much) improvement in attributes results in better performance and how much improved performance impacts strategic success.

So how does the integrated talent suite fit in? Now, instead of just measuring activities (e.g. number of applicants processed, number of reviews completed, etc.), we can better understand the effectiveness of our HR processes in terms of achieving our strategic goals. We can link the results of the acquisition, development, and performance processes with the results of the business. Furthermore, we can better relate those HR processes to better decision making by line managers. For example, management in partnership with HR can better understand whether to invest more or less in acquisition vs. internal development (as well as for what attributes). Together, they can better understand what works and what doesn’t in making the acquisition pipeline effective. Opening the lens a little wider, HR and management can better decide how those investments in processes should change in reaction to external forces like economic, regulatory and competitive change.

Moneyball in Cricket

In Moneyball – the author looks at the Rajasthan Royals (an Indian Professional League cricket team) and how the coach and captain took a statistics heavy approach to running the team..

Wait..you should be thinking, isn’t the Moneyball about the MLB’s Oakland team 8-/ .. yes, you are absolutely right. This is just a twist on the Moneyball theory applied to a completely different setting and a different game . 

Let me take you straight to the stats. Eight teams competed in the IPL tournament and the owners of the team bidded for players from a pool and at the end – following are the teams and how much they have spent.  The team which paid the highest was Kolkata – $6,022,500  and the team which paid the least was Rajasthan (Jaipur) – $2,925,000

Can you guess who the winner was?

If you had guessed Jaipur (Rajasthan Royals), you are absolutely on spot. The Jaipur team had won (11 of the 14 matches including the Finals) and have been crowned champions. There might be arguments about how some teams had to deal with injuries, players pulling out etc. but lets focus on the main question.. how did a team which had put the least money went the farthest?

Following is the quote from their captain:
“The 38-year-old coach and captain revealed the secret behind Rajasthan Royal’s brilliant run in the tournament. Having just four days with the squad before the start of the IPL, Warne said he along with performance coach Jeremy Snape and director of coaching Darren Berry worked day and night to get an idea of what his players were capable of. When we reached here we wanted a background of all the players through the local coaches, which unfortunately we didn’t get. We played two practice games straightaway and watched every player in detail as to how they approached the game, their shot making, running between the wickets, fitness and other aspects,”

In Jaipur’s case, the coach and captain had to deal with Talent on hand and identify the strengths and weaknesses and devise ways to put together a stronger team. The decision making for the captain was also easy as he knew well about their team members.

Isn’t this the situation we constantly face? We always don’t have the option of getting new talent, but we have to work with existing talent and identify and develop the necessary skillset to meet our objectives.