Guiding your journey to HR Analytics success

When browsing online, you can easily find the following definition: HR analytics (also known as people analytics) is the collection and application of talent data to improve critical talent and business outcomes. HR analytics leaders enable HR leaders to develop data-driven insights to inform talent decisions, improve workforce processes and promote a positive employee experience (Gartner).

In other words, HR analytics consists of gathering, using, and reporting on people data to take HR-related decisions. Those decisions should be considered carefully because they impact one of the most valuable assets of a company: people. Nevertheless, in several companies, people-related decisions are taken by HR decision makers using very little data and are then left to a more questionable science called ‘gut feeling’. Beyond any doubt, developing an effective HR analytics system is not easy.

In this blogpost, we will give an overview of the main challenges encountered during past HR Analytics projects KPMG was involved in. We will elaborate on our approach and the lessons we’ve learned.

When starting a HR Analytics project, you need to understand in which context you are working. In some companies, HR is still a ‘forgotten’ department where it might be difficult to obtain funds to improve processes, data quality and/or actionability. Too often, decision makers in companies do not allocate an equally substantial budget to the HR department as they would for other departments. The HR department needs to build a solid case to prove their needs and quick fixes are often preferred instead of deep restructuration and reshaping solutions.

Therefore, it is very important you emphasize the financial as well as the performance impact of HR analytics. If the impact can be monetized and measured, funds are more likely to be given for HR Analytics projects.

Another challenge often faced is identifying the key user and setting the goal of the dashboards. In many cases, it is not yet clear who will be using the dashboards. In past projects, our clients did not always identify the target audience at the start of the project; they were often so enthusiastic about dashboarding that they wanted to include too many users.

Furthermore, the objectives of the dashboards and the alignment with the broader strategy were not always well thought out. We often observed that clients did not clearly specify in advance the objectives of the target audience. The requested dashboards and metrics are sometimes not tailored to the needs of the HR team. For example, starting a project to build indicators about diversity while the topic is not part of the HR (and overall company’s) policy or strategy.

To tackle these challenges with our clients, we organized workshops to understand and identify the first and key group of users. As a best practice, only a small number of people should be chosen. This first stage focuses on the specific needs of these key users and aims to provide them with the first dashboards. As a second step, dashboards should be personalized for each type and group of users. That way we can tailor the dashboards to the specificities of each group of users to make the dashboard relevant for everyone.

In addition to user identification, the definition of the goal is essential. What the client wants to achieve with the reporting should be clearly stated. In addition, the goal of the dashboard and its content should also be aligned with the client’s broader business objectives.

Working with the HR department often means facing a lot of different data sources. At KPMG for example we have different tools for planning, timesheets, and performance. This requires more effort to find, access, combine and analyze data. The lack of one ‘single source of truth’ and non-automated data sources increases the likelihood of differences within the data. In other words, the same metric present in different systems might possess different values. This particularity creates confusion and raises questions on reliability. Our clients often don’t understand why they are receiving different values, and they have difficulty understand which data source is correct and which data to consider.

To solve these kinds of data quality related issues, we spend time collecting, comparing, discussing, and deciding which numbers must be considered. We worked in close collaboration with the system owners to understand where the data is coming from. The data streams and the master data were identified to get a trusted single source of information. 

It is fundamental for a HR Analytics project to understand the HR data environment, which can be extremely complex and restricted by legal aspects. Taking the necessary time for those preliminary analysis and discussions will contribute to the success of your project.

The last and most common challenge encountered during HR Analytics projects is related to the data knowledge. HR departments often possess limited know-how regarding data. They’re often not aware of all the existing data or know how to access it. If they manage to get the data, they sometimes struggle to know what to do with it and how to generate valuable insights. The large amount of information and the complexity of its structure makes it difficult for HR departments to leverage this resource with limited knowledge on the topic.

Human Resources is often a sensitive but also very interesting topic. There are many opportunities for data analytics in this domain to help you make the right decisions. Like for any project and any domain, some challenges will be encountered and, as consultants, we are probably already more aware of them and able to quickly adapt thanks to our experience and expertise. By sharing the four main struggles we faced, we hope to help you in starting your HR analytics journey.

If the topic draw your attention and you would like to benefit from the domain knowledge of our People & Analytics Team and the data capabilities of our Lighthouse Team, don’t hesitate to contact us.

Author: Estelle Pitance & Zoé Dubois