By Olivier Meier, Mercer
The adoption of analytics is accelerating, and its use is growing in the field of global workforce management and talent mobility. This trend is driven by the need to enable more talent mobility without increasing cost and complexity, which is whetting the appetite for process optimization, bringing to the forefront the real value of talent mobility and the importance of making more rational decisions about it. While talent mobility analytics has not reached full maturity yet, it is an increasingly important trend that should not be ignored.
Here are a few things to consider to start your analytical journey.
Metrics are quantifiable measures designed to assess and track a specific process over time. They can be used for a wide range of HR or business issues and provide tangible facts to help understand how the different parts of the business are operating.
Key Performance Indicators (KPI) are sets of measures designed to evaluate the performance of a company, a team, or an individual against predefined goals. They are focused on the question of efficiency and performance.
Analytics goes further than simple measurements describing the past activities of the company and aims to understand why specific outcomes happened and how these outcomes could evolve and change in the future. In doing so, analytics uncovers patterns and connect facts with decision making to drive business improvements. Analytics is a tool to actively drive positive business changes within the organization. There are several types of analytics.
- Descriptive analytics: Mapping and understanding patterns.
- Predictive analytics: Foretelling the outcome of new events, process changes or new policies.
- Prescriptive analytics: Providing recommendations on what action to take achieve specific goals.
In terms of scope, talent mobility analytics focuses on understanding and improving how moving talent internationally can benefit the organization and its employees. It goes beyond HR analytics, which is focused on HR processes, to encompass approaches used by business analytics (e.g. finance and processes) as well as people analytics (such as the impact on employees’ performance, their career paths, social interactions within the organization, and their personal development, to some extent).
The Business Case
Talent analytics aims to help solve a range of issues.
The disconnection between the perception of global mobility and its realities: In our recent survey, 57% of participants deplored the lack of relevance of mobility in the company’s business (“How Global Mobility is Responding to new Dilemmas”, Mercer August 2018).
Conventional wisdom suggests that talent mobility effectively supports business expansion and provides a boost to employees’ careers. The reality is much more complex, and the benefits of mobility can materialize in specific cases, for selected types of moves and employees, under specific circumstances. In some cases, the added value of an assignment is questionable.
Predicting the success of assignments or at least identifying critical assignments types, talent groups, and factors that can affect them is one of the objectives of mobile talent analytics. Analytics provides a more rigorous approach to assignment management and a way to fight decision biases and oversimplifications
Mobile talent acquisition: A better understanding of expectations, required skills, and desired personality traits that make employees successful on assignment can help improve the recruitment and subsequent retention of mobile talent.
Career evolution and attrition of mobile employees: Over time, do employees with international assignment experience have a faster advancing career compared to their local peers? What factors influence their progression and willingness to stay with the company?
Increasing assignee engagement: What’s driving certain assignees to go the extra mile? How do we connect the engagement drivers to business strategy and host country’s management needs?
Understanding cost drivers to better control assignment costs: The cost of assignments is not the simple function of direct costs of salary and allowances. The types of employee, the home-host relocation patterns, and the way the assignment is managed also have a huge impact on total cost.
Enhancing the effectiveness of the mobility function is another important objective. Is the organization of the mobility function optimal and what would be the impact on the cost and efficiency of potential changes (e.g. new processes, modifying the organization structure, increasing outsourcing)?
Fostering mobile talent diversity: Talent mobility has a role to play to foster diversity at the managerial level. The participation of women and some minorities in the expatriate workforce remains low, and, more importantly, the debate is fueled by assumptions about what these groups of employees can or cannot do in specific locations and their expectations. Bringing rationality to the debate is a crucial step in the direction of making diversity happen, and analytics can be a way to identify and put the spotlight on shocking diversity gaps.
Leveraging talent networks: How do assignees collaborate and share information? This is especially relevant for international assignees because they are taken out of their home networks and thrust into unfamiliar environments. Measuring how and with whom they interact in the host location provides insights on their integration in the host locations, their link with the home country and the rest of the organizations, and how mobility is supported (i.e. who do assignees turn to when they need help – would it be local HR, line management, or the mobility team?) These findings could help identify support gaps, avoid employee isolation, and foster exchange of information and skills.
At an individual level, understanding metrics and analytics is a pre-condition for mobility professionals to become more strategic and have a say at the top management table. There is sometimes a misunderstanding about who should own analytics. The real added value of analytics is coming from its relevance to current business issues. This is not a function that can be driven purely by the data and IT teams who are not aware of business priorities. It needs to be a shared business priority between the data analysts and mobility professionals. Analytics provides an opportunity for HR professionals to upskill and become more relevant for the business at a time when automatization and outsourcing could threaten HR jobs.
Pitfalls and Challenges
Careless implementation of talent analytics could lead to a widespread perception that analytics, at its worst, brings misleading results, or, at its best, is just a fad yielding limited business value for talent management and that should not be high on the list of company priorities.
Accessing the data and data quality: Data required for analyses might be hosted in different systems and in different countries. The data might not have been not recorded for a sufficient period of time or consistently captured. Without preparation work and support from IT, it will be difficult to access relevant information. Many HR teams give up early in the process or scale down their ambition due to this difficulty.
Data interpretation mistakes and bias: The saying “correlation does not imply causation” summarizes some of the challenges when dealing with complex data and situation. Establishing correlations and quickly seeing patterns can be misleading: a good understanding of the context and a rigorous methodology are required to avoid drawing false conclusions.
Examples of Common Biases and Errors
- Are employees more successful because they go on assignment, or are you selecting the most successful employees to go on an assignment?
- Assignee decisions and performance might be driven by the complex interactions of pay, career, organizational, and personal issues.
- Make sure to take into account the full context and recognize that some factors cannot be controlled by the company or the employee.
- Small samples can lead to faulty generalizations and exaggerated trends. Results from small sample tends to have more variations and display more extreme patterns than larger ones.
One-time results and lack of persistence
- Persistence over time is the mark of a meaningful trend. One-time exceptionally good or bad performances will be followed by average ones if luck or isolated, non-recurring circumstances were involved, a phenomenon called “regression to the mean”.
- When trying to capture opinions from employees or crowdsource information, remember that expatriates in a given location influence and talk to each other – a lot.
- The more they do so, the more negative impact it has on the validity of the sample, as respondents might influence each other. You may capture the buzz but not the true issues. Make sure you to capture feedback from diverse sources with independent opinions.
Seeing patterns and the dark side of story telling
- Storytelling is important to communicate the results, but not all data tells a story. The human brain is designed to make sense of scattered pieces of information and tends to see trends and patterns that don't actually exist.
- Rigorous methodologies and monitoring a trend’s persistence over time are way to fight this natural impulse.
Unwittingly integrating biases in the analysis
- Unconscious biases can spill into the analysis and be eventually reinforced by the results.
- For example, the success criteria used in the analysis are based on a specific existing group profile and could disadvantage individuals from other groups.
Data security and compliance: Assistance from the legal department is required to understand the implication of circulating and using data for analytical projects, especially in an international context. The question goes beyond simple compliance, it’s about data governance: i.e. what data is analyzed, stored, transferred, combined, deleted etc.? Bear in mind that one piece of information might not be sensitive in itself but combining it with other pieces of information could lead to privacy issues. Companies needs to have a clear view of what is being done with data – the analytical team needs to produce a clear “data catalogue” detailing how data is collected, analyzed and correlated.
Lack of relevance and wrong priorities: A lot of things can be measured, and it is tempting to measure them just for the sake of analyzing. What are the priorities for the business and what patterns are relevant and which ones are merely a distraction? Does the issue analyzed really matter for the business? Can the company do something about it?
Lack of follow up and validation: Oftentimes management don’t know what to do with the recommendations or fail to act upon them. It’s a question of change management – change of perception or processes – that is not specific to analytical projects. However, analytical projects can easily fall prey to this lack of follow up and absence of subsequent effective changes.
How Far Do You Want to Go on Your Analytics Journey?
Ensure that the basics are in place: Mobility metrics allow you to track accurately all types of assignments, directs costs, and assess potential compliance issue. You can then move on to more comprehensive analytics and ultimately predictive analytics.
Take it step by step and break down what you have to do into small steps to generate quick wins. Too many organizations start out with high expectations but stop after just analyzing basic assignment metrics due to lack of resources and visible short-term results, or try to launch over-engineered projects that yield little value and are not sustainable.
Analytics is about improving talent management and business outcomes – HR and business teams should drive the projects rather than the IT or data teams. First, check the organization’s business priorities with management and identify the pain points. Secure a senior management sponsor for team leaders to work with.
As a first project, pick a relevant topic that could lead to actionable results, rather than a complex issue that the company cannot totally control or will yield little to no results. Experience shows that many people within the organization tend to have strong and emotional views about analytics: some view it as the panacea to avoid irrational decisions while others dismiss it as a passing trend. In any case, the risk is that if the first talent mobility analytics project is not delivering added value, this could negatively impact the perception of all future projects.
A lot has been said about the return on investment (ROI) of global mobility – and much of it to deplore that finding a way to measure it is difficult. According to the Mercer 2017 Worldwide Survye of International Assignment Policies and Practices, 77% of companies have not established a way to determine the ROI of their international assignments. 21% of companies are working on it and making some progress, with only 2% reporting being confident of having a full understanding of the ROI of their assignments. The real added value of global mobility from companies can be measured by a cluster of analyses rather than one magic figure. Rather than trying to over-engineer some form of complex ROI formula, starting with selected metrics and analytics about the impact of global mobility can bring clarity to the debate.
Start by asking critical questions to capture the right metrics and analytics:
Organizing the Analytics Function
Having the right skillsets: Upskilling the HR function is an important step but this doesn’t mean that HR needs to become expert in all aspects of analytics. Developing relevant analytics should not be a one-man show and a cross-functional team with a diverse range of skills is needed. Tapping into the expertise of other departments and learn from their success and failures is a good starting point.
Use solutions from within the company: Reach out to marketing or client service departments who have a lot of experience with customer analytics and tap into their expertise.
Explore your options to build up your capabilities: Upskill HR business partners, build a dedicated cross-functional team, leverage existing analytical teams, or partner with an external provider.
Effective Data Management
Ensuring data quality: Effective data management requires rigorous auditing to understand what’s available, have clear definitions and labelling (data dictionary), as well as clear planning/mapping of the data (interaction between the pieces of information).
Establishing best practices: If your data sources are fragmented and inconsistent, start establishing processes to bring more consistency in term of standards, labelling, and methods used. This might take time and will require coordination with other teams, especially IT. Start the harmonization process early on and not when you are already trying to perform analyses.
Multiply sources of information to increase the validity of the resources and reach out to other parts of the organization to understand the information used. Not all relevant data will come from the HR department itself. Very often, the success of an expatriation will be measured by metrics and data that are not purely related to talent mobility or people issues – e.g. financial, customers or production data.
Validation over time: Assess the validity of your assumptions by measuring different factors over time. This is especially important for talent mobility because most of the value of assignments may come after they have been completed: the skills acquired by the assignees will boost their career (or not) in the years following their assignments. The positive changes and new expertise brought by the assignees in the host location organizations where they have been relocated might not manifest itself immediately.
Establish a baseline measure for use in tracking progress toward a high performance mobile workforce and talent mobility function. Analytics should be viewed as a way of taking the pulse of the business on an ongoing basis rather than as a one-time exercise.
Making the Best of the Project’s Output
Build trust and support internally: Educate stakeholders and ensure transparency about data collection and interpretation. Talent analytics should not be a black box where the internal workings are hidden or not easily understood.
Communicate your results: Storytelling has been identified as one of the key skills required by talent mobility managers. A clear message based on the results has to be conveyed by the HR team to top management and backed with strong evidence and a structured story. Effective visualization of the results and having well-designed dashboards can go a long way to convince stakeholders of the validity of the recommendations.
Focus on what matters and keep going: Finally, accept that it is not possible to change everything or act upon every issues highlighted by the analytics project. Do not drown the most relevant input in the middle of long list of irrelevant items. Put a spotlight on what matters most for the business stakeholders and keep refining your approach to increase the accuracy and relevance of your analytical activities.
Contact the author: Olivier Meier.