Data Strategies and Tips from SHRM Speaker Cathy Missildine
Last week I had the pleasure of speaking to Cathy Missildine of Intellectual Capital Consulting about how every company can and should be using data to improve their hiring processes. Like planning a family vacation, what might seem intimidating from 1,000 feet away, can become much more manageable when tackling it in bite size steps. I focused my conversation with Cathy on what she thought some of the lowest hanging fruit is that every HR manager can and should be doing, then asked her about some of the more intermediate and advanced steps might be for those who want to take things to the next level.
The end result is some fantastic insights about how to begin building a more data-driven HR team.
Starting with Strategy:
Data and HR have come a long way in the past decade. Cathy has been a speaker at SHRM for over 12 years and remembers the days when her sessions had just 12 attendees, rather than the 1,200+ at her most recent event. The reason is—HR practitioners now face opportunity overload. With so many new tools available, leadership has become gut shy and uncertain of what tools to invest in, making a data-driven approach, not just a nice to have, but a mandatory way to navigate the noise and get budget approval for new campaigns.
While it is easy to get excited and want to start finding new data points to measure it is critical to start this conversation internally with a focus on strategy.
What are the big problems you are trying to solve with your hiring process? What do you need to measure and what are just nice to haves? A few common areas, HR teams use data to improve include:
- Quality of hire
- Time to fill
The Fun Part–Easy and High Impact Places to start:
Once your team has developed a high level strategy on what you want to improve, it is time to begin thinking about where to start. Luck for you if you are new to this process (and oftentimes, even if you are not) there will be plenty of low-hanging fruit–metrics that are easy to track and which will reveal deeply important stats that can influence your HR process in important ways. All of these numbers can be tracked a calculated with a basic version of Excel (although if you are really serious, their statistical add-on pack, is a powerful next tool to begin working with).
Retention (or conversion rate): Retention is defined by the percentage of employees who were employed at the beginning of a time period and remain at the end of that period. In university recruiting, since most internships last for short period of time, the more useful statistic to look at, is the percent of an intern class, who end up converting into full-time employees, which is a different but important stat.
To get an idea of how retention works, think about it like this: if you start with 10 employees and 2 leave you will have an 80% retention rate. Now those two new employees could also leave, but you still have an 80% retention rate of your original employees, just a higher turnover rate!
Like all data points, just knowing a single statistic (we retained/converted 20% of our intern class!) isn’t all that helpful. You will need more details to help the numbers tell a compelling story. How many of the interns who converted were engineers. How many were marketers? How many interns under manager A converted versus under manager B? What initial sources ended up sending the interns who converted the most? These slightly more detailed numbers help you make smarter decisions about who to hire, who should manage interns, and where you should hire from.
Turnover: Turnover is similar to retention but different in important ways. It looks at the number of departures over a period of time, divided by the average number of employees during that period. For example, if you on average have 10 employees and 5 left, than you had a 50% turnover rate during that period. Given that it is uncommon for interns to leave mid-internship (although it does happen) this stat is more valuable for entry-level hires, than it is for interns. We do recommend you understand the circumstances for any intern departures as that is worth investigating.
Cost-per-hire: Cost-per-hire is one of the most important numbers you can track as an HR professional and fortunately it is relatively easy. Ideally you should be slicing and dicing your cost-per-hire data to look at cost-per-hire for different role types, as well as cost-per-hire for different sources.
You might find you are spending twice as much money for hires from Michigan as you are for hires Wisconsin and while this may be okay with you, it is important to know..
Quality of Hire: Measuring the quality of your hires and what sources they came from is hugely important. Large companies like Google have found that GPA is not a strong indicator of intern success so it’s important to track this metric and begin to understand what types of interns work best for your company. So how do you do it?
At the middle and end of an internship, do performance assessments and measure this data against the university, source of hire and other variables that can help you better understand where your best candidates are coming from. Then double down on the sources that are working for you!
All five of the above metrics are relatively easy to track and are guaranteed to make your HR team more efficient and cost effective.
One part of the process that I know always holds people back from diving into data, is the fear that starting this process will mean entering into Excel hell. Most of the spreadsheets are actually quite simple to make, but knowing that is always easier to start with a template, Cathy was kind enough to share a few with us!
The below spreadsheet has been modified to help campus recruiters calculate cost-per-hire. Download it for free by clicking the link below:
The thing about using data is that it can be very addictive. Once you get comfortable with slicing and dicing data in your Excel spreadsheets you will begin to realize how deep these insights can go. A few things to look forward to as you begin to level-up your skills:
Correlation regression: Is there a tie between pay and retention? Is there a tie between manager and converstion rate? Correlation regression allows you calculate for r-squared and the closer this number is to 1, the more tightly coupled your variables are. However, correlation does not tell causation.
Structured Equation Modeling: Cathy called this the highest form of data analysis. Equation modeling involves building your own models and testing them for causation. This might be a few years (and even a few new classes) down the road, but is a powerful tool for the data scientist who wants to truly understand how to improve HR processes.
Hope this helped and happy measuring!