HR analytics is an area that is growing, but still has some barriers to adoption around it.
An AirBnB article about how they democratize data science got me thinking about this. The article outlines several important concepts about applying data to business decision making, but in an HR context, we need to proceed with caution.
Whenever you’re talking about collecting or using data in an HR environment, it’s understandable that sometimes people get nervous. HR departments have access to some very sensitive personal data and it’s important that trust is always maintained.
This is one reason why I see slow, but steady growth as a good way to manage HR analytics. Here are some points to consider:
Download our resource for finding HR datasets here
Protecting HR data
Our ability now to collect and process massive amounts of data has many potential applications for business use. Importantly, decision-making can happen from an informed position based upon data, rather than going purely with instinct, or with some incomplete data.
AirBnB talks about a fundamental belief that every employee should be able to make data-based decisions. “This applies to all parts of Airbnb’s organization — from deciding whether to launch a new product feature to analyzing how to provide the best possible employee experience,” they say.
In theory, this is a great way to make better decisions in any department. In an HR environment though, we have reason to be cautious. There is always a large element of confidentiality attached to HR data, something which we need to respect. It’s not just about being compliant with relevant laws, but maintaining the trust of the people whom we gather the data from.
Different parts of the world have different regulations, and AirBnB are obviously able to operate within the context of the regulations that apply to them. In Europe, that means adhering to the regulations laid out in GDPR. From Wikipedia’s GDPR page, these are the most relevant “data protection” points in my view for the introduction of HR Analytics:
“Business processes that handle personal data must be designed and built with consideration of the principles and provide safeguards to protect data, and use the highest-possible privacy settings by default, so that the data is not available publicly without explicit, informed consent, and cannot be used to identify a subject without additional information stored separately.
No personal data may be processed unless it is done under a lawful basis specified by the regulation, or unless the data controller or processor has received an unambiguous and individualised affirmation of consent from the data subject.”
Data protection is something that I take very seriously, and most other HR professionals do too. Recently, I was appointed Data Protection Coordinator for all administrative purposes at my workplace. For me, this means observing the highest ethical standards when it comes to protecting that data.
Any data processor in this sort of role needs to clearly state what they will be doing with data collected in advance. The concept of “consent” is important here and should always be taken seriously. There needs to be an effective advantage to the data subjects giving out their information. Automated profiling into categories of employees without the explicit consent from employees is absolutely a no go.
So in saying “every employee has the opportunity to make data-based decisions” like AirBnB, this could be translated in HR Analytics with the following: “every company takes decisions in full transparency of the data subject and for the common benefit of all stakeholders”.
HR Analytics at the workplaces will be able to grow with the increase of trust.
Data protection is a role to take very seriously in HR analytics Click To Tweet
Making data informed decisions
Managers should be making data informed decisions to help improve trust. When team members can see that decisions have been made from a basis of good data, it helps them to see the value of it. It also helps to prove that data is being used for good purposes and that decisions have merit.
I find a good way to do this is to follow a solid, working methodology. I like the process shown in the slide below, taken from the authors of The Power of People: Learn How Successful Organizations Use Workforce Analytics to Improve Business Performance. (This is a great resource on the topic of HR analytics, by the way).
It’s important to start with a clearly defined business question that you’d like to answer. This helps to guide you through the rest of the process and stay on-topic. Whenever large amounts of data are involved, it can be easy to get caught up in the weeds, looking for insights that aren’t relevant to what you need right now. Note that in this process, gathering the data is the third thing on the list, rather than the first.
HR processes and data science
A good data science methodology should empower managers to understand and work with data. Of course, it’s also important to ensure that HR business processes comply with GDPR. Once HR business processes comply with GDPR requirements, a lot of progress can be made in analytics using data science tools.
For example HR processes can be optimised using data science types of analysis. Visualisations of HR data can be improved a lot. Lawful HR data analysis can be made faster and more efficient using typical data science tools. Normally a lot of savings can be achieved through HR Analytics alone by reducing employee turnover. Savings can also be made by improving the storytelling part of a data science project.
The diagram below (also borrowed from the authors of The Power of People), shows the principles that make for effective storytelling in HR analytics. The importance of the story is in connecting with other people (for example, your boss or project sponsors). Most people don’t connect with data itself – in fact, you may lose them if you go too far into data. They will, however, gain a better understanding from a story that relates to the data.
Data education
When AirBnB examined barriers to scaling their use of data, one of the main “blockages” was user knowledge of how to utilise data and data science tools. Data education is crucial for driving informed decisions.
Training can be offered to data scientists working on HR data. Managers can do their part by helping to formulate very clever questions for which data scientists will look for answers in existing data. I am very much in favour of offering training via “datacamp for business“. I have used it myself for over a year now and find it very helpful.
Another great way to learn is by using sample datasets that have been already made public to the HR community for training purposes. I am aware of about 30 already available and these are very suitable as public datasets for training.
Final thoughts
HR analytics promises to help transform how we work in the field of HR and how we reach decisions. Nonetheless, not all HR decisions can be taken solely based on data – there are going to be limits to data-driven decisions and still a lot of intuition is needed.
The initial business question and the analysis are important, but do not need to be perfect at the beginning, as companies can iteratively improve from one year to the next. This demonstrates an important aspect of HR analytics – that you learn from it and make adjustments over time.
You could look at the use of data analytics in HR as a marathon, not a sprint. It’s about making steady progress over time.