Talent is the most valuable resource that any company has, as is retaining that talent. While this has always been true, it cannot ring truer in the economy of the future.
Decades of falling interest rates and loose monetary policy have resulted in an economy flush with capital, all of which are chasing a shortage of new innovative ideas1. Talent has thus become the new scarce resource and is also arguably becoming one of the most expensive.
The loss of talent would, as a result, be one of the highest costs for a business to incur today.
It has been estimated that a single employee’s departure could cost an organisation up to 250% of their annual salary, excluding hidden ‘soft costs’ such as productivity loss, decreased engagement and cultural impacts.How to improve employee retention. CIO, 2019
Many leaders acknowledge this problem and have spent a tremendous amount of effort attempting to reduce their turnover, but statistics show that in most cases they have not been successful. In the US alone, total voluntary turnover rose from 23.7% in 2015 to 27.9% in 2019. Among younger workers, this was more pronounced – 38% of millennials reported that they would leave their jobs if given the choice to in 2017, this increased to 49% in 20192.
Retaining talent has proven difficult because turnover is fundamentally driven by complex human behaviours and decision-making processes. Effectively, retaining talent requires the acknowledgement that each employee is different, has a unique set of needs that their roles need to cater to, and that the decision to leave or stay is driven by every experience throughout their entire employee journey.
What does Data and Analytics for Employee Retention look like?
Technological advancements in data storage, data mining and computer processing mean that organisations now have the resources to treat employees as such. Effective People Analytics teams should provide businesses with the power to understand each employee and provide targeted strategies at the employee level.
In fact, this type of transition has already been very successful for another function present in almost every other business – Customer Relations. Many modern CRM systems are extremely powerful, collecting data about customers at every step of the customer journey and integrating data from other systems. This supports customer-facing teams in creating extremely personalised customer journeys, and in some cases, personalising the product itself, tailoring to individual needs at a scale that was not previously possible.
The amount of data and metadata that can be recorded about an employee and their work mirrors, or in many cases, trumps, that collected on customers. The level of personalisation that is available for customers should thus be possible for employees too. Businesses that are serious about retaining their talent should harness this data to reengage their employees as if they were their customers.
Using Data and Machine Learning to Reengage Employees
By acknowledging that talent retention involves every element of the employee journey, from job application to termination and from their social networks to the software they use, businesses can start putting together a strategy for effective talent retention. At the minimum, this strategy should provide the company’s first line of defence – its managers – with the tools they need to improve employee engagement and reduce turnover. These tools would include algorithms capable of:
- Identification of an employee’s needs and flexibility to cater to them
- Real-time performance and engagement monitoring
- Identification of at-risk employees and intervention options
The following sections will explore what each of these capabilities should entail, and will enable, along the organisation’s path to reengaging its employees. However, it must be noted that there is foundational groundwork that must be laid before this can be achieved. Disparate data sources must have already been brought together, data silos knocked down and a culture of trust in the data already been built across the organisation, or at least the function. A separate article explains how TrueCue has helped its clients accelerate this process, check-it out here.
1. Identification of an employee’s needs and flexibility to cater to them
For an employee to be engaged, their role must not only be one that they like and enjoy, but one that caters to both their professional and personal needs as well. These needs, which may range from flexible working to promotion opportunities and rewards or even growth opportunities in other fields, vary greatly between individuals and throughout an individual’s life. Needless to say, there is no one-size-fits-all solution.
Historical data including exit interviews and engagement surveys can help to inform on the multitude of requirements that need to be catered for, however, this on its own is a blunt instrument that cannot be implemented effectively. For example, it is no longer sufficient to know that a certain demographic of employees requires flexible working options. Fully harnessing the power of data to differentiate your company as a better employer means identifying the extent and type of flexibility required by each employee, the time of day at which it is needed, and finding an appropriate team that can complement this schedule, all before they have even asked.
This level of specificity allows for working conditions to be tailored and modularised for each employee. In the talent-scarce economy of tomorrow, businesses must move away from the supply-driven model of job creation and begin designing work processes around the employee, just as products are being designed around the customer. AI-driven recommender systems, as is already being used in CRMs, can support businesses and managers in doing just that.
One type of recommender system, such as those used by Netflix and Amazon to suggest new products to customers, can be used similarly to suggest required tweaks to each employee’s benefits and working conditions that will cater directly to their individual needs. These predictive algorithms can go on to identify the best work allocation, scheduling and team formations that is personalised around its understanding of each employee’s needs.
Another type of recommender system, such as that used in Oracle’s CRM system to provide recommendations to sales reps for best next actions or surfacing relevant records at the most opportune moments3, can be employed as well to support managers more directly. These can help managers better recognise, engage and communicate with team members by guiding managers on the best courses of action, tailored for each team member.
2. Real-time performance and engagement monitoring
Employees need to be noticed and feel noticed, to feel that they are valued by the company. Reducing the time lag between an employee’s actions and their manager’s reaction is thus crucial in retaining talent. The occasional employee engagement survey or performance review should no longer be the standard in this new age of data. Instead, managers can be provided access to data that may indicate changes in performance and engagement scores, habits or emotional states (such as efficiency data, access card or login/out timings and sentiment analysis) in near real-time, such that they may identify changing needs in their employees. This enables managers to provide support in a timely manner.
It is also important that this goes beyond the typical scope of performance monitoring, providing a full picture of the employee’s interactions with the company. Advanced techniques such as Organisational Network Analytics can be applied to employee data to identify the individuals who are highly performing in non-traditional metrics, such as being knowledge brokers in the organisation and improving performance of others through network effects. This would help build a strong foundation for employee engagement, allowing individuals to feel both noticed and valued.
3. Identification of at-risk employees and intervention options
No matter how successful one may be at creating an engaged workforce, some turnover will nevertheless be inevitable. The identification of at-risk employees in a way that allow managers to react appropriately is thus the last key element of a successful talent retention strategy. This means providing managers information on who is at risk, what value they bring, and what options they have.
Once again, predictive technologies and machine learning can be employed not only to quantify the risk of each employee leaving the organisation, but also predict when. Furthermore, new techniques have been developed to demystify black-box algorithms, allowing models to also be capable of suggesting some of the most likely push factors for each employee.
Used together with advanced analytics techniques that are able to accurately model the impact of each employee’s departure (in areas such as rehiring or training costs, changes to the organisational network or loss of institutional knowledge), these analyses will arm managers with the relevant information required to develop appropriate responses that caters directly to the employee. It is only with this support that managers will be able to make informed decisions on the need for an intervention, the most appropriate intervention, and the cost-effectiveness of the intervention.
The idea of treating employees as customers is not new – many Japanese businesses have been running on ‘employee-first’ models since the 1980s4. As the war for talent continues to intensify, businesses need to rethink their strategy to stay competitive in the talent market of the future. Harnessing new advanced analytics and predictive technologies and applying them to employee data will enable businesses to treat employees as who they really are – unique individuals making decisions based on a complex web of factors both in and out of work – and tailor each employee’s experience around who they are.
Mingyang's background in economics and finance complements his expertise in delivering advanced analytics and data science solutions. His experience in both industries provides him with an astute business acumen that allows him to drive change and create value for organisations using data-driven methods.
Capitalising on this unique skill set, Mingyang has delivered automated end-to-end people analytics platforms encompassing data warehousing, diagnostic reporting, machine learning and network analytics