How to improve employee wellbeing with data

Why use data to improve wellbeing?

Employee wellbeing is a critical aspect of any organisation but can be difficult to measure and improve without the help of analytics and data. By analysing employee data, organisations can identify areas where employees are struggling and need support. With data analysis and reporting, employers can create tailored programmes to improve employee wellbeing and deliver real improvements.

Individuals have a unique way of understanding their own wellbeing, yet this can be difficult to translate into something applicable for everyone. Experience helps identify when someone needs support, but how do we determine if our team or our organisation as a whole feels healthy?

The challenges of a data-led strategy for wellbeing

Employee wellbeing is a complex topic by nature, with many factors that can affect the individual's and team's wellbeing.

With an understanding of what's possible and some clear objectives, data can be an invaluable part of your wellbeing strategy and guide your decision making process.

Let's take a look at some of the possible problems when developing your wellbeing strategy.

Lack of available data to assess wellbeing

The data you have available to you can make it difficult to develop an effective strategy. Without accurate and recent data, it can be difficult to measure the current state and determine what interventions will be most effective in promoting employee wellbeing. Additionally, without good data, it can be difficult to track progress over time and identify areas of improvement.

Sometimes, the data you need is difficult or impossible to find, for example, if you haven't been capturing what you need. Other times, it's just a matter of figuring out where this information is and how to get it into your analysis and reporting platform.

Difficulty of measuring 'soft' attributes

Many of the factors that contribute to an individual's wellbeing are inherently difficult to measure. Mental health, psychological wellbeing, and interpersonal skills have no obvious direct metric you can track.

This can make it difficult to get an accurate measure of how your employees are really feeling and behaving - and what areas you need to focus on improving. However, there are still some useful ways to collect data on these softer measures.

Measuring personal attributes can be imperfect and subject to the measurer's own perspectives and biases. For example, looking at data on employee absences may lead to the conclusion that people are taking too many sick days. However, this may ignore the underlying issues such as stress or poor mental health that are causing these absences.

Complexity of analysing changes to wellbeing over time

As HR professionals, we are often tasked with analysing changes over time – whether this is trying to understand how rates of sickness have changed over the past year, or assessing the impact of a new policy.

However, it can be difficult to know how to best analyse this data. Time can be a tricky thing to analyse, not least because it keeps changing! This presents the challenge of comparing different versions of the same data and keeping your data sources current. There's also the inevitability that analysing change over time requires dealing with much more data than when providing simple snapshots.

Complexity of combining data from multiple sources

Information useful for analysing employee wellbeing often requires the combination of data from multiple data sources. For example, the core HR database might contain the list of employees, their teams, locations etc, while performance or productivity data is held within an operational database or account management system.

Combining data from multiple sources is often a complex and time-consuming task. The analysis process will require a multi-stage approach to obtain and transform the data so that it can be combined to achieve accurate results. This becomes particularly time-consuming and potentially even unreliable when the same processes must be repeated over time when the next set of reports is required.

Data from different sources must be accurately matched based on one or more identifying values (such as employee number). Identifiers are often not obvious and results may appear correct where they may not be. Care must also be taken to correctly aggregate records where there are multiple records in one data source for a single record in another.

Difficulty of interpreting data about people

Determining the factors that are the most important indicators for employee wellbeing can be challenging. Judgement is often required to determine the meaning behind the analysis.

Indicators are often not as clear-cut and straightforward to understand. Some indicators might appear to be positive, yet arise from negative issues. For example, it might seem like very low levels of sickness absence would be good news, but this could just mask problems with employees feeling fearful about being absent when they're genuinely unwell.

The indicators we see in our organisations can sometimes give us mixed messages - some measures that look positive on the surface level have an underlying negative cause, which will eventually come out if left unchecked for too long

What is employee wellbeing?

Many elements contribute to employee wellbeing. Factors that affect wellbeing are varied and wide-ranging but include physical and mental health, emotional wellbeing, and professional relationships.

• Physical health and fitness

  HR teams often play an active role in promoting healthy lifestyle choices, but less often proactively monitor outcomes, evaluate improvements, compare groups or assess trends over time.

• Mental health and emotional wellbeing

  Mental health is an essential factor in employee wellbeing, but is often overlooked due to being less outwardly visible and more difficult to measure. Mental health includes stress management, anxiety, depression, and burnout.

• Professional relationships

  Positive and fulfiling professional relationships are the foundation of a strong working environment. Employees that feel valued by the managers, peers and direct reports are much more likely to function well.

How to improve your organisation's wellbeing

Delivering improvements to wellbeing requires engagement across your organisation, focusing on the intended outcomes, measuring the positive and negative indicators and reviewing the effects.

Engaging your organisation about wellbeing

The essential first step to improving welling is to engage the people across your organisation. The profile of employee wellbeing can be raised by including wellbeing concepts in an organisation's existing processes.

Wellbeing concepts can be introduced during employee induction and regularly discussed during ongoing training.

Annual performance reviews and employee engagement surveys provide opportunities for reinforcing the importance of wellbeing and for collecting additional data.

Involving employees and line managers

Encouraging employees and line managers to get involved in wellbeing initiatives will help to embed a culture of wellbeing across your organisation.

Employees should understand the importance of wellbeing and how it contributes to the success of the organisation. Line managers should be aware of the signs that an employee is struggling and know how to support them. The HR team should ensure that data on employee wellbeing is collected and analysed regularly.

The more people talk about wellbeing, and particularly mental health, the easier it becomes for people with problems or concerns to find support. Open and honest conversations give people the courage to discuss their issues, which leads to them being more likely to obtain the help they need or support available to them.

Identify the opportunities for improvement

Your instincts and perceptions as a people manager should provide the foundation for wellbeing analytics. Findings based on your data will often support your instincts, and will sometimes contradict them - but starting with a theory to be disproved is a useful strategy.

Consider challenges that you already know your organisation faces. Where do you think the greatest opportunities to deliver benefit exist?

For example, has staff retention among older workers been difficult recently? Or has remote working generated challenges for employee engagement with cross-team collaboration?

It is also useful to consider your organisation's objectives and goals, as these often evolve from a desire to address underlying problems.

Sell the benefits

Senior stakeholders are more likely to support wellbeing initiatives if the business benefits such as health and safety, productivity and talent retention are positively targeted.

Individuals are more likely to willingly engage if they understand the likely benefits. Closing the gap between actions and outcomes can encourage active involvement.

Clear goals and timescales can be useful for leading the improvements.

Monthly or quarterly reviews

Many organisations already undertake annual performance reviews. Whilst often useful, annual surveys aren't frequent enough to support responsive wellbeing improvements.

Organisations looking to improve their wellbeing should consider more frequent staff consultations and surveys. Quarterly feedback reviews and short monthly 'pulse' or health-check surveys provide more up-to-date information.

How do you measure employee wellbeing

If your goal is to improve your organisation's wellbeing with a data-led strategy, the obvious first problem is determining what data you need and how to get hold of it. Your organisation will already have some data that can be used, but some data that you might want might not be available.

What data do you already have?

One way to get started is to review the data you already have. This approach usually provides the quickest route to delivering something of value. Also, the data you already have can be a rich source of inspiration.

For example, most organisations have employee survey data from annual performance reviews. If this data is available in a structured database, some simple analysis and comparisons can generate some insights - particularly when comparing between groups of employees or changes over time. If the data is less structured (for example, free text records), sentiment analysis can be useful.

Other data sets that organisations often have immediately available include sickness absence rates. If return to work interviews have been captured, more advanced analysis will be possible. Similarly, employee retention and turnover rates - particularly how they compare or change over time - can generate useful reports.

What data can you easily get?

There are some datasets that you might want to use that your team doesn't have access to. To get hold of additional datasets, you'll need to identify and engage with the custodians of the data. Be prepared to justify why you need the data and describe your plans for using it. You should also be ready to address any governance questions that arise around access and security.

You will need to formulate a plan for obtaining the information, including possibly how you will transform the data and associate the records to existing datasets.

Consider where your copy of the data will be stored, who will have access (and how access will be granted, controlled and revoked), how you will ensure the data remains secure, and how it will be removed when it is no longer required.

If you are taking copies of a dataset (rather than linking to the primary source), you will also need to plan how the data will be refreshed so it is kept up to date. How often will data be refreshed, who will be notified of any problems during an automated refresh, and whether your existing copies will be replaced, updated or extended by the updated source data.

What data can you make a plan to get?

Some data that will be useful to you simply won't be available. If you haven't been capturing data, you won't be able to analyse and report on it. This might mean making changes to existing processes or introducing new systems.

For example, if you want to use data from employee engagement surveys to measure employee wellbeing, you'll need to ensure that the surveys are designed in a way that allows you to capture the data you need. This might mean adding extra questions or changing the way that responses are captured.

If you want to use data from performance reviews, you'll need to ensure that the reviews are conducted regularly and that they include questions about employee wellbeing. You might also need to consider making the reviews more formal if they are currently informal.

Similarly, if you want to use data from return to work interviews, you'll need to ensure that the interviews are conducted with all employees who have been off sick (regardless of duration).

Finally, if you want to use data from employee retention and turnover rates, you'll need to ensure that the data is captured accurately. For a fuller picture, this might mean introducing new surveys such as exit interviews.

The plan for obtaining additional data should consider the people and teams that will be responsible for managing the systems and how you will bring this into your analytics and reporting platform.

An action plan for improving wellbeing

• Decide what data you want to acquire in the medium term. Start collecting data from new sources as soon as possible so you can build up your knowledge-base.
• Review your strategic goals to help guide your data analytics. Analyse the data you have to evaluate your theories. Develop your business case for improvement, using your analysis to support your plan.
• Continually monitor your key metrics to confirm how well your improvement plans are working.

Please get in touch if you would like to discuss how we can help you.

Helen Ramsey is responsible for strategy and operations at viewpointbi.com. Her focus is data and technology within HR. She writes regular articles on people data and analytics.

Stephen Simmons is responsible for technical strategy and platform architecture at viewpointbi.com. He leads the software development team and writes regular articles on data and technology.