Results From Our Annual Hackathon: MercerHack 3.0

Min Park
Min Park
Oct 6, 2017 · 7 min read

By Linda Chen & Min Park

By Bruno Scramgnon

If you are interested in more information about the outputs of our hackathon, and the issues they help address, please connect with Linda Chen ( and Min Park (

In August, Mercer’s Workforce Strategy & Analytics practice hosted our third annual hackathon (“MercerHack”) in Washington DC. The event aims to drive team engagement, deliver a burst of innovation, and, ultimately, better serve our clients. This article highlights the key outputs from our event.

What is MercerHack?

First off, what exactly is a hackathon? The English Oxford Living Dictionary defines it as:

“An event, typically lasting several days, in which a large number of people meet to engage in collaborative computer programming.”

MercerHack was certainly an event, which did last several days — two and half to be exact — bringing together a team of 36 analytics consultants. More importantly, we have always intentionally engaged in collaborative programming during our hackathon events. While our definition of “programming” is not limited to writing computer code (after all, we are not exactly software developers), we are a diverse mix of curious builders, creative thinkers, and disciplined researchers and data scientists who regularly wrestle with large volumes of workforce data.

With nearly 25 years of experience consulting in the people analytics arena, our team has collaborated with hundreds of organizations on a variety of topics ranging from identifying the drivers of turnover, to quantifying the impact of human capital practices on business performance, to analyzing organizational networks. Our primary focus has always been to help our clients generate strategic insight from their HR data and develop evidence-based solutions to drive actions with greater impact.

MercerHack is our annual opportunity to take a short break from regular client work, and elevate the side-projects and ideas that otherwise collect dust on the proverbial shelf. We take the time to think more broadly about common questions we have heard from clients and reflect on major trends we are seeing in HR and people analytics. With these priorities in mind, we then get to hacking.

What Did We Create?

We brainstormed nearly three-dozen potential concepts for hacks on the first day of the event, all of them revolving around key issues for our clients.

Ideas ranged from developing new technology applications, to evolving our methodologies, to conducting original research by leveraging our vast data resources. Over the course of nearly 30 hours of hacking — as we all cycled through various degrees of panic, frustration, elation, and pride — fifteen of these ideas came to life in the form of concrete outputs.

In no particular order, the following snapshots provide a high-level summary of just a few of our innovations from the event.

We developed a framework for taking a disciplined and data-driven approach to help our clients address, navigate, and plan for the impact of the Fourth Industrial Revolution on their workforce. The hack included proof of concept analyses and tools (sample outputs below).

Top: A wire frame of key questions to be addressed using our framework. Bottom-left: An interactive data visualization of an 80,000-employee workforce by likelihood of complication due to technology; based on multiple data sources, client input, aided by machine learning for accelerated mapping of jobs. Bottom-right: A network map of jobs in another organization, linked by similarity in capabilities required for the role, with color-coding by technology-impact likelihood. Both can be used in workforce planning to help navigate the risks and opportunities that lie ahead (for example, re-skilling where realistic).

The topic of pay equity is top-of-mind for many organizations as they seek to better understand their current state and respond to growing expectations around transparency and disclosure.

While we have been helping clients proactively address this issue for over a decade, clients are increasingly asking us to evaluate various adjustment scenarios and their likely impact on pay gaps before they lock down a strategy. Common questions include:

  • If pay inequities are identified, what are the various remediation options the organization could pursue?
  • What are the tradeoffs between these multiple, and often conflicting, scenarios?
  • Moreover, what is the post-adjustment impact?

The hack itself was an Excel-based tool that could quickly calculate the post-adjustment impact on pay gaps across a wide array of scenarios. By taking standard outputs from our statistical analyses, users can immediately begin testing and comparing what it would take to make progress against their gaps.

There is a great amount of press coverage on the positive relationship between organizational performance and diversity.

In reality, the story is more nuanced. Productivity can get worse with increased diversity before it gets better. At the outset, newly diversifying teams may run into challenges in learning to work together, as they are more susceptible to naturally forming non-diverse subgroups — defined by “faultlines” in the team — which can restrain performance.

One of the MercerHack teams uncovered these nuances by applying a rapid literature review technique to academic research in this field. A complementary team conducted pilot research with a blinded client data to find empirical evidence supporting the literature. As follow up, research is underway to help our clients identify and measure where they have such faultlines, and what specific interventions can moderate the potentially harmful effects so that organizations can reap the benefits of diversity faster.

Left: Theory implies a curvilinear relationship between diversity and organizational performance. Right: Our diagnostic based on multivariate regressions reveals the exact point at which increased diversity has a positive return on performance. Research is continuing to understand how organizations can identify interventions to moderate this effect, and achieve greater productivity sooner.

Individualization and personalization are gaining traction when it comes to enhancing both the customer and employee experience. HR has been taking a page from marketing in using unsupervised machine learning techniques, such as k-means clustering, to identify distinct segments in the workforce that may have different preferences. This is to ensure that programs and policies are tailored to meet unique needs. Traditionally these segments, or “personas,” have been based primarily on demographic factors.

One of our teams flipped the process upside down. Rather than using demographics, they used employee survey data to identify workforce segments based on the employees’ stated preferences regarding total rewards, as revealed through conjoint data. Conjoint surveys ask employees a series of trade-off questions to measure the relative importance of various total rewards elements (e.g., pay, type of work, career development, flexibility, etc.)

What we learned is that these preference-driven personas cut across demographics and can be effectively targeted for different interventions. Furthermore, there is significant potential to connect this survey-based information with behavioral data (e.g., actual employee turnover) to refine insights and associated actions that HR can take to optimally serve their workforce.

Top: How the team hacked the unsupervised learning process, to identify more accurate employee segments based directly on preferences, rather than demographics. Bottom: example results revealing four distinct workforce segments with unique preferences.

Other Hacks

Other hacks included research around what drives Millennials, based on statistical modeling of longitudinal Human Resource Information System (HRIS) data covering their experiences and actions; development of several additional tools to improve and expedite delivery; building out our library of R functions to streamline and scale our analytical capabilities; refinement of job matching algorithms, among others.

The winning team developed an interactive report template in Tableau, enabling our consultants to review and engage with data outputs in a more dynamic fashion, more rapidly.

As a testament to the value of team diversity, the six colleagues from the winning and runner up teams represented colleagues from five different office locations, with anywhere from less than a year to 10+ years of experience.


Innovation, design-thinking, hackathons — these are some of today’s well-intended buzzwords to signal an aspiration towards achieving something greater than the status quo. Unfortunately, they can become generic and overused in our desire to innovate quickly and often. By staying true to the spirit of a hackathon (see epilogue) and encouraging team members to pursue a passion project, the outputs remain authentic and relevant.

We thank our team for its creativity, dedication, and diversity of thought. We are especially grateful to Mercer leadership for retaining its commitment to thoughtful, research-driven innovation in an age when “quick and dirty” often rules.

We encourage you to share your emerging HR and people analytics issues with us, and to reach out to learn more about our hacks.

Linda Chen ( and Min Park (

* * *

Eplilogue | Can a Hackathon Work for You?

Hackathons may not be the right format for all teams, but it could work for you too! Reflecting on what else makes the event a success, we believe the following things matter:

  • The event is grassroots, not top-down: ideas for hacks are generated from the crowd, across all levels. The brainstorming session we have on the first day encourages open and supportive discussion to further flesh out ideas.
  • It’s free-form, not prescriptive: we place few restrictions on what teams work on. We trust in the caliber of our people to pursue great ideas, based on the belief that people do their best work when genuinely engaged.
  • Outputs have to be usable, not hypothetical: the only expectation we “impose” is for all teams, whatever they do, to produce something concrete and tangible by the end of the event. While we all love ideas, we want to use our hackathon time to bring them to life.
  • It’s an accelerator, not a destination: the event jump-starts many initiatives that are further developed throughout the year, while also producing many outputs and tools that are immediately useful for our colleagues and our clients — the “hacking” doesn’t end at the event.

Following the event, we conducted a survey asking our team to share three words that come to mind when they reflect on their experience. Here’s what they said:

Participants were asked to share three words to describe their experience at MercerHack. The top five words used were “fun”, “collaborative”, “energizing”, “exhausting”, and “inspiring”.

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store