I love you mom

My mother, Elizabeth del Barrio Bernal, was an English teacher and speech and drama adviser at the Philippine Science High School (PSHS) for 35 years. She read to me every night as a child and showed me the beauty of literature, music, culture, and art, particularly the performing arts.

The following is a redacted version of the last speech I delivered before my mother went to heaven on May 3, 2019. I was able to mention her public service in passing, but did not get the opportunity to properly pay tribute to the depth and breadth of her legacy during that speech. I am praying that I will get the opportunity to do that someday soon.

Everything I am today, everything I love and believe, everything I have ever achieved is because of my mother.

Mama, mommy, I miss you. I love you so much, and I will love you, always and forever. I am so proud of you, mama. Thank you for everything, especially for being my mother ?? I know we will see each other again in heaven one day.

---------------

"HR Metrics and Analytics" - excerpt from presentation to the Civil Service Commission on March 27, 2019

Good morning!

Before this day, my only encounter with the Civil Service Commission was when I took the Career Service Professional Exam in 1999 as an aspiring civil servant. Back then, 22 years old, a new mother and barely a year out of college, I took the test in Quirino Elementary School (which incidentally is where my grandmother taught as a public school teacher until she retired in the early 80s). I took the test, understanding that it was a qualifying test for government service, but not really thinking too much about its significance. I took the test and managed to qualify - but unlike my grandmother and mother before me, who both had the privilege of being lifetime teachers in public schools, a lifetime of public service was not in the cards for me. Now, 20 years later, as I reflect on that experience taking the Civil Service Exam, I see it in a whole new light. The Civil Service Commission is charged with one of the most noble and important responsibilities in government - to institutionalize meritocracy and excellence in public service; to make sure that only our nation's best minds and hearts get the privilege of being public servants. And this is enshrined in the agency's core purpose - Gawing Lingkod Bayani ang Bawat Kawani - Make Every Civil Servant a Servant Hero.

And so I am deeply honored to be invited to this gathering of our government's best and brightest - our government's "HR" or "People Team", so to speak - or the guardians of the light, as I would like to think of those in the profession. And while I am not an HR expert by any stretch of the imagination, I do hope I am able to share with you some useful "nuggets of insight" gained over years of trying to figure out how good people and good organizations work.

And that brings me to our topic for today - "HR Analytics and Metrics" - which combines two of my favorite things - numbers and people.

Before I became an HR professional, I studied to become a mathematician. That's not because I was particularly good in math or because I found it easy. I studied math precisely because I found it challenging. I liked how it stretched my thinking and challenged me to think in both logical and creative ways. Math is more than just arithmetic, or algebra, or statistics, or calculus. It is, as Albert Einstein put it, "the poetry of logical ideas".

And numbers are very powerful - they are these invisible things that make intangible things tangible in our heads. If you're struggling to articulate a concept or bring an idea to life, throw in a few numbers in there and all of a sudden, what was once fuzzy and abstract becomes more real, with delineations and dimensions, making it suddenly more comprehensible - like an artifact you can hold in your mind. Numbers do that. They allow us to more accurately describe ideas, they enable us to contextualize things in space and time, they give abstract concepts perceptible characteristics (size, age, rate of change, etc), they make problems analyzable and easier to break down, and they make methods routinazable, repeatable and scalable. Numbers give us a sense of magnitude and direction. Having a number to wrestle with in our heads brings us one step closer to understanding something better.

And this is why HR needs numbers. The HR function has historically been at the tail end of the organizational value chain, despite the inherent importance and criticality of it's role - there is no organization, public or private, that can exist without people. This is largely due to the function's lack of or inability to show numbers that demonstrate the function's value to the organization, as well as the value of people-related decisions to organizational outcomes.

In this age of big data, super fast computers and machine learning algorithms, it is absolutely imperative for HR and every other discipline for that matter, to take full advantage of the power of analytics and metrics, not only to drive functional efficiency, effectiveness, and impact, but more importantly to create competitive advantage by driving desired cultural behaviors and strategic organizational outcomes. With the vast volume, velocity and variety of data now available for us to study and unpack, and with the computational power and learning capacities of today's machines increasing ever so rapidly, the winners of tomorrow will be the ones who will be able to unlock the power of data to reveal non-obvious insights that enable their organizations to evolve in ways that leave the competition behind.

And it all starts with data and measurement.

It starts with collecting, connecting and classifying data, and then describing "what happened" - that's descriptive analytics.

Anything that describes a data set - the sum, percentages and ratios, the measures of central tendency - mean, median, mode - the measures of dispersion, the standard deviation, etc - all of these are descriptive statistics, which are tools for descriptive analytics.

When you report performance against targets or governing KPIs - how you've executed on your strategic, tactical or operational goals as set in your plans - that's descriptive analytics.

A few other examples:

The usual measures of "HR efficiency", or the amount of resources used by HR programs, such as

• Cost of hire
• Time to hire
• L&D Budget
• Training time in days
• Time since last promotion

-- that's descriptive analytics.

The usual measures of "HR effectiveness", or the outcomes produced by HR activities, such as

• Retention rate
• Absenteeism rate
• Individual performance
• Team performance
• Quality of hire
• Employee Net Promoter Score

-- that's descriptive analytics.

The usual measures of "HR impact", or the business or strategic value created by HR activities, such as

• Market share
• Profit margin
• Market capitalization
• Customer satisfaction
• Customer loyalty

-- that's descriptive analytics.

When you create dashboards using data visualization tools to visually represent your metrics, that's descriptive analytics. They answer the question "what happened" - they give insight about whether things are going well or not in the organization, but they do not explore drivers, root causes or make predictions or recommendations based on the data.

After knowing "what happened", you would want to understand "why it happened" - that's diagnostic analytics. Looking at trends and thresholds, detecting spikes, anomalies and deviations from plans or expectations, surfacing unusual events, identifying drivers, and finding root causes - that's diagnostic analytics.

So when you present actual performance vs the targets set in your plans and explain the variance - that's diagnostic analytics. When you drill down into the analytics to figure out the drivers of unexpected outcomes or the reasons for variations from plan - when you drill down to uncover hidden causal relationships and look for the various endogenous and exogenous sources of variation - that's diagnostic analytics.

After knowing "what happened" and "why it happened", you would want to understand "what will happen". While descriptive and diagnostic analytics give you a snapshot of measures that show your execution against plan, and help you understand the reasons for deviations from plan, predictive analytics helps you make predictions about what will happen in the future based on what happened in the past. And for this you will need to use statistical tools such as regression analysis, or maybe even neural networks to find relationships between data and use that to predict or classify new data.

Prescriptive analytics, takes inputs from predictive analytics, and combines them with rules and constraint-based optimization to enable better decision-making. The major difference between the two is that predictive analytics forecasts potential future outcomes based on what happened in the past, while prescriptive analytics helps you with specific recommendations, and helps you figure out the optimal option from a suite of options by combining the power of prediction and optimization. In other words, prescriptive analytics augments predictive analytics by informing decision makers about different decision options and the expected impact of those options on key performance indicators. For example, if you're using geolocation tools to figure out which route to take to avoid traffic, predictive analytics can estimate the time it will take for you to get from point A to B for each possible route based on the experience of others who have taken those routes, while prescriptive analytics, looking at the different route choices and predicted ETAs can recommend the best possible route, if it understands what you're trying to optimize. If the goal is to minimize travel time, then it can recommend the route with the soonest possible ETA. This is an example of predictive and prescriptive analytics at work everyday.

Cognitive analytics, builds on the 4 previous types of analytics by combining intelligent technologies like artificial intelligence with human-like intelligence to perform certain operations. It tries to imitate the way the human brain processes information by making inferences from data, drawing conclusions based on current knowledge, and then codifying those conclusions into learning to inform future inferences - this makes cognitive applications smarter over time.

AI is coming, whether we like it or not....

So in closing, how do we do this?

I know it seems overwhelming - getting from where we are today - as a function, as an organization, as a nation - to where we want to be. But don't let the hugeness of the challenge stun you into inaction. Start small, and start with what you have. A lot of the jargon being thrown around these days are just new words for the same old things - data science just used to be analysis, data visualization just used to be graphing. Choose what works for you and your context - you may not need super fast computers or super complex calculations to tell you what to do - but don't shy away from trying to learn new tools and new methods. Machines and algorithms can help reduce bias in human decision making, but humans, with domain expertise, intuition and insight, contextualize the products of machine decision making. And it is this sweet spot, this partnership of humans and technology, with human intelligence exponentially amplified by technology, where "creative analytics" lives. If we can unleash the creative power of the human mind, by combining human ingenuity with insights derived from the full spectrum of analytics - descriptive, diagnostic, predictive, prescriptive, and even cognitive analytics - then we can innovate beyond our wildest imagination and unlock the next wave of growth and disruption.

Thank you very much.