Blogs

We live in a very smart world… especially when we remember how far technology has come in a short period of time. So much technological achievement in our lifetime has made many things that we would have considered miraculous a few years ago are commonplace.

Just think about how streaming video has taken the place of DVDs in the delivery of media content to the extent that have revolutionized how people consume media and traditional media companies have had to evolve to survive. That same broadband connectivity has led the cloud to become our storage and computing methology of choice rather than larger hard drives and more computing power in our client “machines.”

That broadband and computing power has been leveraged to create new shopping, marketing, and even HR applications. There may be plenty of places in Human Resources where computers can be brought in to make employee experiences better, but is the world ready to consume more from what the machines can teach us… yet?

I say, “yes” and “no.” There are many cases where you can imagine the world of HR and the employee being helped by bringing the power of computing and broadband making their experiences better.

AI in the Employee Experience

One example is homeshoring… why can’t people stay at home either as a choice, perquisite, work at home on a regular basis, or permanently? The ability to work at home saves time, money, crowded highways and cities, and reduces stress. But, what does this have to do to AI?

Consider that Employees in an office environment have people that they can turn to in order to make their lives easier. We had office managers, receptionists, assistants, and a mail room. Now, the world of homeshoring brings many problems that WE need to solve on our own, not the least of which is having someone to smile at, talk to and befriend. AI can’t get you coffee, unless you are ordering Starbucks. It can’t send a package unless you use the UPS skill on your Amazon Echo device. It can’t tell you a good joke over at the water cooler, but ask Siri a joke… she will give you a weak, but politically correct one. So, maybe it can be useful to employees finding themselves at home during the pandemic, or afterwards.

Another, more valuable use of AI is that it enables the employee to find answers to questions they ALWAYS have! Questions like: how much time left in my vacation balance? When does my 401(k) vest? Can I take tomorrow off? What is my next career move based on my skills? These are all good questions, some take more assumptions than others. Can today’s AI answer them?

HR and AI… why not?

The problem is that AI is just not that smart. Artificial Intelligence in HR is getting there, and so are the people that are training these AI brains, and that is a HUGE prerequisite. Those that are focused on those problems need to know how to find the right answers. That’s key… the right answers to the questions that employees have.

One problem is that the data that we can use to train these AI brains are not ready to make employee decisions. The current data on decisions around pay and career that could be used as training sets. But the data needs to be scrubbed so as not to continue pay inequities and lack of diverse promotions. Frankly, that effort is entirely too big. It means reviewing decades of decisions on pay, promotions, hiring across hundreds if not thousands of companies.

So where can AI go in the world of HR?

Health and Welfare, Workforce Management, Learning and Development, and Analytics are all good candidates for AI assistance. They have plenty of data available, provide centricity for decision making with good benchmarks for comparisons. Examples include:

Open Enrollment – when an employee is going through their enrollment in benefits, we can look at their personal situation and review what their claims and costs would typically be for different plans offered (HMO versus PPO) and provide for them cost estimates of how expensive the options would be in the future… AI improves this by knowing the patterns of the employee’s absences (sick, vacation time) and enrollment of a new baby with the costs patterning over the next year for sick and well-baby visits to the doctor. The AI infused decisioning makes the decision better and more realistic for the employee.

Vacation time – We all struggle to budget our vacation time wisely throughout the year. The AI can plot our annual vacations, school calendars and business trips on an yearly agenda enabling us to plot out our vacations well in advance to make sure we don’t leave precious days on the table as we are too busy to focus on them.

Enrollment in skill-based training – knowing that an employee just got promoted to a managerial job can provide the opportunity to give that person critical skills to mentor and lead. AI can offer that person courses to help provide these skills or auto enroll them when their calendar is available.

Analytics in decision-making – providing the best examples of metrics that are abnormal for your population versus the rest of the organization has been popular AI targets. You can easily measure outliers for users. The real analytics gold for AI would be to track manager decision patterns to point out the best decision outcomes and provide alternatives that might be more effective given past outcomes. Even this example is using previous decisions for training that might provide inappropriate outcomes instead of the BEST outcomes. Until the AI can have judgement built in that protects the company from making bad decisions, then we are destined to repeat the past mistakes.

Human Resources is extremely complex and we can’t simplify the solution nor the problem in a blog post. Many concerns exist from employee privacy to legalities of making decisions based on previous decisions. EY has penned a fascinating paper on the benefits and concerns around AI in HR. Northeastern University has AI as a component of their MS in Human Resources.

Obviously, there are many other use cases. Maybe you have AI use cases that are valuable to the world of HR. Put them in the comments. Let’s start the discussion so that the robots can listen and learn from our experience.

Stay safe.

I am sure you can’t. I cannot either. It is a common problem we have these days where we cannot leave our computers for a moment. This is amplified by our 24x7x52 access to information through our smartwatch and smartphones. We are NEVER disconnected.

This dependence on information and insight can cause us to be TOO dependent on the answers these gadgets solely provide us. As I have discussed in previous articles, it is all too evident that our underlying, core data is flawed and requires constant care, feeding and nurturing even with the best processes and core systems. When we simply report on the “facts” in our analyses without proper context, we face the errors of unforeseen lack of insight. So what do we do?

We drop the mouse, put on our facemasks, step away from the computer to observe and listen. Yes, I mean good, old-fashioned observation and listening skills. Let’s talk to people, maybe not face to face (yet); maybe via Skype, FaceTime, or Zoom, but uncover what is happening the old way… by going to the sources of our data, our leaders. That way we can literally connect the dots between offline and online. This also provides you with opportunities to start conversations with business leaders about my favorite three letter word…“WHY?” Why you are talking to them leads to their curiosity about what you can provide and how you can provide it to them.

Imagine finding just 2 leaders a week to reach out to. Create a script to follow. Start with your old friends and contacts in the business that may not have ever talked to you about People Analytics before. Talk to them about the current challenges they are facing. Then show them what the statistics show from your perspective. Give them a taste of the insights that can drive their decisions and future conversations.

Each week find new items to discuss with new people but keep the conversations short and on track. Make notes of the meetings and refine the strategy. Make sure you let HR leaders in those areas know what you are hoping to accomplish so as not to cause undue suprises and turf wars. Maybe try not to invite those HR leaders as it may change the conversations, since you want raw feedback and their presence may shift it. But, be flexible. You know your leaders and how they might react to direct contact without their presence.

Now that you have started to make these connections and have an open channel you can uncover why major changes in data occur. You will be able to footnote and develop the story behind the numbers that provide the critical context for what the pictures, graphs and data mean. It also will provide the audience for the analyses with the full understanding of what is going on before those dreaded calls happen on “Why do these patterns look this way???”

My favorite reason for stepping away from the data and talking though, is creating that linkage between the analysts and the owners. You are the data analyst and the business leader is the people leader. They own the decisions they make and with your help and the right context, they can make better decisions with your help. After a while, think about the cool visualizations you can create between the offline and online data. I am sure those few minutes away from the computer will be the inspiration of brilliant things to come.

Stay safe.

Never.

Of course I could just leave it at that, but brevity is not a skill I currently possess. Whether you have 1, 100, 1,000 or more than 10,000 employees in your organization, your People Data is dirty. Even if you JUST finished a People Data audit, I promise you that there are pieces of data that are incorrect, missing or duplicated. It doesn’t even matter how meticulous you are.

Why? Because your organization is a living organism. It is complex and constantly changing. Don’t misunderstand, I am not saying that you need to be in growth mode to have incorrect people data. Other causes could be: layoffs, voluntary terminations, transfers, promotions, new hires…. the list is endless. Even every single time punch gives the opportunity to have incomplete, missing or incorrect data and that process drives payroll and other downstream processes.

DON’T PANIC!

At the end of the day, we typically find these errors and omissions before they go to payroll or impact significant business processes. The questions are: who finds them and when? If you have rolled out People Analytics tools to managers and executives, let’s pray that someone in the HR world finds them first so the fixes can be made before angry calls or tense meetings are called. Believe me, I have been in those meetings. They are not fun.

So, what do we do? We cannot audit People Data every day. The answer is actually pretty simple. Set a routine. Either have one person or a group of people from each business unit, spend a small, but routine amount of time each month going through their unique sets of data to comb through what might seem like a needle in a haystack, but an important needle.

Make it a fun thing… a friendly competition. Have rewards like donuts, iTunes gift-cards and/or public recognition for the person that finds the most or the most egregious errors. The goal is to make sure the constant checking provides you with the peace of mind to sleep at night knowing that there is someone looking after your data and making sure that that call or meeting invite never comes from the Executive office.

The goal of having a robust People Analytics function/practice in your organization should be to have people around you who value the data and insights they bring. It only takes one big data error to destroy that goal. So, keep a regular check on that data. Make it part of your people process hygiene. It will be well worth it and in the end make your analytics more reliable.

Stay safe and sleep soundly.

Yeah, we are tired of Covid-19 related stories, but what about the upside? Maybe, we can finally see what we have neglected, or I can see what I have neglected for so long… Stepping outside every day, smelling the Spring air. Walking through nature… with my son.

I have connected with my son again in ways that I am partly ashamed of. Why didn’t I take him for hikes before? Why didn’t we go out and throw the football or play street hockey? Not that I NEVER did these things before, but why not regularly?

The answer is that I was always working. Always too busy when he was home from school. That seems like a joke now that he is home, working on schoolwork in the next room and I am working hard to get a business off the ground. So, now I need to work harder and try harder. There needs to be a happy medium. Work-life balance is a thing!

So now I will try harder, and I will dedicate part of my day to hanging out with my son. Whether it’s hiking, hockey, skating, playing Just Dance (2017,2018, 2020) on the WiiU, or just watching Lego Masters (yeah Tyler and Amy). I will do better at being a Dad to my kids not living with me, like seeing them on FaceTime more often with longer conversations that probe into how they are dealing with this craziness.

Maybe throughout this crazy, messed-up world we are living in right now, one good outcome is that we may appreciate what we have and who we have right now; not tomorrow, not next week, not next month. Because of the recognition of the frailty of life, maybe tomorrow is here today. Maybe we need to appreciate what we have right now and this would be an awesome lesson to learn.

Stay safe while you live on.

During the first World Trade Center bombing in 1993, I was working for an investment bank in New York City. At the time of the bombing, we couldn’t believe that terrorism could come so close to home and that it could impact our daily lives so dramatically.  In response to that crisis, we created an urgent need for updated personal/emergency contact details and reporting relationships.  But it also signaled a renewed need to understand what we do in the case of a major disaster.  Since then there have been several crises that have affected us… both nature and human made.

What lessons can this Pandemic teach us?  Have we found holes in our people data that prevent us from making a smooth transition between office-based and home-based work?  How about our time systems? What about tax filings now that more people are working home-based?  Do we even know who was home-based before the crisis and who is home-based now?  Do we know who and how we can communicate effectively to all employees through appropriate channels?  The answers start with taking some efforts from all of us:

Practitioners

Can we take the time to make sure that we look at the core structures underlying our data: jobs, earnings codes, pay codes, hierarchies, and locations? Analyzing your people data without these key tables being accurate could lead to wrong conclusions and thus inappropriate or even damaging resulting actions.  So, while you may have been busy with the updates required by the COVID-19 regulation changes, it may be a good time to put some cycles into reviewing these key data structures.

Managers

Let’s ask managers to take the time to review their employee populations.  Have them take a look at reporting relationships, job titles, grading, maybe even have them review job matches for salary surveys.  If you have a good template, it might be a good idea for managers to look at current or even draft new job descriptions.  Does the crisis have an effect on their employee’s jobs?

Employees

It might be a good use of resources to send out requests to currently home-based employees to have them help us by updating their data using Employee Self-Service.  Have they looked at their beneficiaries lately?  Mailing address changes? Updated mobile phone or email addresses? Emergency contacts? Think of it as “data Spring cleaning.”  

If you work in #PeopleAnalytics, #HR, #HRIT, #IT, or #Finance, our day jobs still need to go on.  But if you provide insights to managers and executives that they are relying on to make critical decisions now, you may look back at this time and wonder if there were things we could have done to improve our People Data to make analytics more accurate.  In crises, our data is key to ensuring that we can make the best decisions possible for our stakeholders including our customers, employees, managers/executives and shareholders.  Any updates, no matter how small they may seem are good steps forward.  Take one when you have a moment.  Take two and you will feel a deeper sense of accomplishment.  Take more when you can… stay safe and we will get through this.

Most of us that are office-based are working remotely these days. We feel isolated, lonely and disconnected from our co-workers even if we are surrounded by our families. For many of us that work remotely, it isn’t business as usual because we have new co-workers. My son is a constant reminder that there is something very different going on. He loves to come into my office when I am working, writing or on conference calls and telling me something “VERY IMPORTANT!” which usually relates to the TV remote.

What makes this world different now than February 2020, is that many of us are working from home in a state of panic, chaos and uncertainty. No one knows when and how this pandemic will end; when the curve is flattened, a vaccine is discovered, or the virus runs its course. The questions about how this affects our families, the community around us and our jobs/places of work keeps many awake at night. So, where do we go to find answers?

We can go to many sites like the CDC, our federal, state or local governments for information, but where can we go to get answers about what is happening at work? Let’s go to our greatest data sources… ERP, HRMS, Recruiting, Time and Attendance/Workforce Management, and Benefits systems. The data in these systems are a treasure trove of insight based on the trends pre, during and post virus. But, that is only true if we are sticking to the rules and processes that were established pre-virus.

Yes, it shouldn’t be top of mind, but are people logging time? Are workers off, furloughed, working from home, or laid off? Do our systems even have the right “status” for what people are going through now? If not, we may not be able to get a true picture of this period. So, maybe we need to add new action/action reason codes, new data structures, new or updated instructions, or just a friendly reminder.

Let’s help those people who own their data to help us, by encouraging them to maintain it. Workers still employed, complete their timesheets. Those process owners that recruit, provide benefits, hire, fire, give increases, bonuses and approve workflows… must keep all of these processes moving forward. Yes, keep processes moving forward so that we can measure the true business impact to workforce management, payrolls, benefits and ongoing operations.

Someday, we will look back at this time and take a true introspective look at how it happened. How the virus spread. How we reacted. How it impacted our kids, our parents, our partners, our lives, and our jobs. But for now, we need to keep completing the processes that make our systems work. The result will be a full and complete picture of the pandemic’s impact.

For now, schools and businesses are closed or closing. Business operations are slowing down and we can expect to work from home for at least the next few weeks. In the meantime, my son and dog have interrupted me writing this at least 5 times. Hopefully, it still makes sense. 🙂

Stay safe all… take care and we will get through this.

A friend asked me why I am so deeply passionate about People Analytics? It was a very strange question because I had never actually thought about it in that context. To me it was always about providing critical insight to HR, managers, executives, and other stakeholders that care about making the best business decisions possible having all the evidence they need. So, how could people NOT want people insights? That spurred another thought… how much people analytics do they need?

Well, I sat down to think about that question and put on the TV. The local news was on. Not only were there two newscasters, but there was the 3 day forecast in the upper right hand corner and the scroll bar at the bottom of the screen. Oh my goodness, it was information overload. I have gotten used to it over the years I guess, but then it hit me that managers are constantly dealing with information overload and much of it becomes noise after a while. How do they filter out the important information from the noise that may distract them?

I think the answer is it depends… it depends on the individual, the business, the economic market situation, corporate culture and the maturity of the organization to use analytics in business decision making. We all have stories of how organizations have started using “HR Dashboards” to provide some stakeholders with “key HR performance indices” that can show snapshots of several HR metrics. It usually was spreadsheet-based and looked like a collection of fuel-gauges or histograms. It had little or no narrative for interpretation and it was old the moment that it was published. Even so, managers may have opened the email or filed it in “Important HR Emails” to be opened when the moment of inspiration hit.

So, why are People Analytics important and how to judge how much is too much? I wrote last week about what Analytics are, but to summarize: People Analytics provide a key set of insights that business leaders can leverage to help guide them to make better decisions about their world. They are not HR insights. They are not HR data. They are business tools that provide context for managers, business leaders and HR to understand the trends associated with the people side of business decisions. People Analytics provide insights that managers require to see how their decisions cause effects in the world of the people who ultimately have to perform the tasks necessary for the organization to be successful.

Critical lessons can be learned not only from “hindsight” metrics such as turnover rate or new hire turnover rate, dive deeper to look at WHY people leave with metrics such as termination reason histograms. If you pair these metrics up with demographic data such as gender, race/ethnicity, and age-cohorts and you can start to do more in depth diversity metrics on how we are treating our populace. It could lead to better management training, coaching and less employee adverse-impact actions.

The question might be: “where is your company seeing the biggest challenges and what metrics can be provided to leaders to help them see the people side of it?” So as to not overwhelm leaders, start small. One set of metrics, one dashboard may be the right starting place that provides them a glimpse of the power of the people insights. This doesn’t overwhelm, it invites more questions like… What other insights do you have that can help me?

The bar is set high with the noise we face in every day life. Newspapers, internet, and TV provide us with too much on demand information. We create the filter we need in order to avoid overload. We should be careful with leaders to provide them with the right amount of people insights to get them what they need to be successful, especially at the beginning.

What is an employee?  Does it include part-timers?  Contractors? Gig-workers? Seasonal workers? Directors on your Board?  The answers of these questions have the potential to cause problems; sometimes big problems.  But, who owns these answers?  Where are your data definitions?  When were they last reviewed?

You may not have a published set of data definitions for your people data.  Most organizations don’t have them either.  It is common for us to continue to have data entered and transactions recorded without looking back and asking “why did we design our people system of record the way we did?”

There is a story that my old friend Don told me once which made me laugh and think about how we perceive why things are done the way they are done…

Every Christmas, my mom made a roast that everyone in the family just loved.  When I was old enough, she taught me how to make it, but one thing always bothered me.  We loved the recipe and always made it the same as she did, but why did we cut off the ends of the roast.

She said, “you know, that’s a great question.  Let’s ask Grandma since this is her recipe.”  So, we called Grandma.  I asked her, “So why did we cut the ends off of the roast?”  She laughed and said “Honey, I cut the ends off to fit the roast in the pan.”

We record events in our people system of record the same way as always, not because they are the right things, the most efficient, or the most measurable.    We do it because they have always been done that way…  maybe it’s time for us to look at them again.

In order for us to accurately measure anything with analytics such as turnover rate, headcount or otherwise, the definition of data entered into our systems need to be clearly followed for each element else the measurements will be inaccurate.  So, the data clean up we perform at the start of an analytics endeavor is a critical part of the effort to provide confidence in the stories we tell from the insights we find in the now, clean data.

Think about how good that roast may be if we ate the ends, or maybe the roast will taste sweeter to continue to cut the ends off the way Grandma did.