Thanks to ERE for its post about building an analytics function in talent acquisition. Personify’s perspective is that data analytics in recruiting is a lot like digging for gold. It takes effort to mine gold successfully, but if you know what to look for and mine the right areas, you can strike it rich. The same is true for data analytics in recruitment. There is a ton of data out there, and if you aren’t careful, you’ll dig a lot of holes that only produce dirt. But, with the right knowledge and expertise, recruitment data analytics can deliver a motherlode of insights making you more effective and efficient. At Personify, we collect and analyze critical data across the recruitment funnel to improve performance. We track investments and measure ROI on an on-going basis to optimize efforts and investments. We look for correlations to identify variables connected to desirable results. And when we change inputs, we use regression to model “what if” scenarios to help predict and manage potential outcomes. It all comes down to collecting the right data, running the right analyses, and scrutinizing results with the right eyes to interpret it correctly and act. See ERE’s article ‘How to Build a Better Talent Acquisition Analytics Function’ below.
How to Build a Better Talent Acquisition Analytics Function
Shot during the hundred days leading up to the Mr. Universe and Mr. Olympia competitions, Pumping Iron is a 1977 American blockbuster about the world of professional bodybuilding. The film revolves around the lives of bodybuilders and contrasts each man’s personality with the environment he trains in. As a result, Arnold Schwarzenegger and Lou Ferrigno became household names.
In 2016, during an interview with Graham Bensinger, Arnold reflected on his Champion approach during the Pumping Iron days, describing how he would look in a mirror to find weak points in his physique, after which he’d implement appropriate exercises.
We can extend the same approach to talent acquisition through analytics. They are the mirror into which we can look to help us transform the chinks in our armor.
Here’s what to consider as you build your TA analytics function.
Excavation: Data Discovery
“We are drowning in information but starved for knowledge,” said best-selling author John Naisbitt.
The ability to collect an extraordinary amount of information on everything from hiring patterns to marketing campaign efforts is tremendous. Data is everywhere and scattered across various systems and platforms within an organization.
Therefore, when setting up an analytics function, the initial hurdles to overcome entail gathering data from all sources (across the entire organization) and identifying what is statistically relevant. A visionary leader who bridges across functions and silos can pave the way forward.
The Foundation: Building Master Data
Once you have your data, you’ll need a massive cleanup exercise to organize and structure it effectively. Data extracted from multiple sources needs to be transformed and loaded into a consistent format. This would serve as the master data, which you can then use to create basic reports in a primary stage (like creating pivot tables in Excel to monitor progress, recruiter productivity, and utilization). Investing in the right technology here can help in structuring and cleansing data. Additionally, your ATS or ERP can help ensure consistent structure.
The Superstructure: Visualizing Data
Building the right visualizations can bring life to data in ways that make it easier to identify trends and patterns. A robust data visualization tool can be leveraged to build powerful dashboards and insights, particularly when presenting information to senior executives.
You can look at various dimensions to examine key metrics and KPIs. Some extremely powerful visualization tools are Tableau, Power BI, Qlikview, Infogram, and FusionCharts. This is the stage in which you can draw comparisons between various geographies, business units, roles, sources, etc., as well as identify problem areas and root causes.
Finishing: Predicting the Future
You can now use statistical techniques like regression, correlation, and hypothesis testing to build predictive analytics capability. This is where machine learning can help to enhance prediction and accuracy on an ongoing basis in real time. The more the data, the better the chances of getting accurate predictions. And here again, there are a plethora of AI tools to help with building this capability. (You’ll want to invest in coding capability at this stage, especially in languages like R and Python, to reap the benefits of machine learning.)
A Layered Analytics Approach
Using a three-layered analytics framework can help define value metrics in a more organized manner. The below illustration depicts the framework along with examples of some key metrics in each layer (keep in mind that you must build layers sequentially, as they have strong interdependencies):
A growing area of analytics in the recruitment space is digital analytics, which consists of data and metrics related to the visitor traffic to the careers’ website, jobs posted on various platforms, and overall engagement on social media platforms. More and more companies are leveraging this data to run targeted campaigns to attract the most relevant talent groups for specific jobs in the most optimized manner possible. Such data can sit on top of your recruitment funnel and expand your outreach in a far more efficient manner.
For instance, you might compare dropout rates across sourcing channels and even from within sections of your company website. All of which can help answer questions like: Where should I place a particular image or a hyperlink on my website? What color schemes are most attractive? How do I improve visibility of my jobs on a particular job board? Why did a particular campaign do better than others?
By building the right kind of TA analytics function, you’ll position your business to make informed decisions that help steer it toward greater success.
Credit: ERE Media