Do You Measure What Matters?
Analytics and Its Growing Value to TA
Data is the currency of this era and when it is collected and analyzed wisely it dispels assumptions and improves efficiency. For recruiters, data analysis can provde insights on who is the best fit for a positon and quality of hire. But good analysis is hard to do and requires accurate data and the ability to analyze it correctly.
To help understand the potential power of good analysis, I have taken the data classification outlined by Tom Davenport, Jeanne Harris, and Jeremy Shapiro in a recent HBR article and adapted it for recruiting. It divides talent analytics into six types and I have added a rating for each type to give an idea of how useful each can be. A rating of ‘1’ means it has little value other than to tell you what happened. Ratings from 2-4 indicate progressively higher value. A rating of ‘5’ means it is highly valuable for decision-making and planning.
I do not expect that many recruiting functions will be able to do sophisticated analysis in the near term. That will require better data than most function now have and the employment of a data scientist. What I do hope is that this article this will give you a sense of what is possibe and spur more organizations to invest in improving, collecting and analalyzing the people data they now have.
1. Human capital facts: This includes the basic data that is collected by most talent functions. This includes the number of hires, time to hire, cost per hire, and so on. Getting valid data requires good collection techniques, data integrity, and integrated systems, which are not the norm for most TA functions. And when this data is lumped together and reported for all recruiters and hiring managers, as it often is, it is almost valueless for good decision-making. To be valid the data should be presented for each recruiter and hiring manager and would also include trend data. Even then it would only show what has happened in the past and would not interpret the data. But all other forms of analysis depend on the accuracy of this basic data. Usefulness Rating: 1
2. Analytical HR: This is data collected to gain insights into specific functions or people. For example, each recruiter’s time spent recruiting for a position could be compared to how many people she hired and then further compared to how many of the people hired stayed with the company or got promoted in a period of time. It could include an analysis of the specific skills a recruiter has compared to the time taken to find and assess a candidate. It could include the cost per hire or time to fill for each recruiter compared to the hiring manager involved. This requires a clear understanding of what is going to be measured, over what time period, as well as being sure the data collected is correct. It is also important that correlation is not confused with causation, which might require further analysis and testing. Usefulness Rating: 3
3. Human Capital Investment Analytics –This data indicates which actions have the greatest impact on the business? For example, what action improves employee engagement, what is the impact of onboarding on retention, or what effect the coaching skills of a manager have on performance. This is often referred to as predictive analytics and helps decide where to invest and where improvements should be made. The only talent functions I am aware of that do this with any rigor are Google and Unilever. Usefulness Rating: 4
4. Workforce Forecasts – Data that helps the talent function analyze and forecast future talent needs. This is really a form of strategic workforce planning and requires access to the firm’s strategic plans, good data on the supply of specific skills, data on turnover, internal mobility and promotions, the impact of training programs, and much more. This is a complex task and one often frustrated by sudden changes in the business or economy. Probably very few talent functions will ever be able to do a good job of analysis in this area. Usefulness rating: 4
Talent Value Model – This data helps understand why employees choose to stay with a company. What specific practice, manager, or program, for example, influences an employee to leave or stay? It helps to decide whether to offer a counteroffer to an employee who is leaving or to proactively intervene to prevent a resignation. Usefulness Rating: 5
Talent Supply Chain – Highly complex data analysis is required for this level. This data helps determine which positions should be permanent and which ones could be filled with a contractor, temporary, or outsourced staff. The data indicates which positions deliver the most value to the organization. This requires objective definitions of value and raises ethical concerns. However, its impact on profits and productivity could be large. No one is able to do this level of analytics yet. Usefulness Rating: 5+
Analytics will be the most important aspect of recruiting after automation. Just finding and hiring people is not enough if you cannot objectively show value. If you are only gathering and using the baisc level of reporting human capital facts, you will need to up your game to remain effective.
Convert to or Buy a Paid Subscription
You can get immediate access to our 2022 trends report, previous podcasts, exclusive interviews, and white papers by subscribing.
A paid subscription is only $2.50 per month/ $30 per year.
Need a Speaker or a Webinar Presenter?
I am available for speaking gigs either in-person or virtually. The topics I speak on cover those similar to these articles and focus on future trends and issues recruiters face today and in the near future. I am also available for webinars, workshops, and seminars. Click on the button below to find out more details.
Related Links
People analytics at Google: using data to make Google a great place to work
Oracle’s downloadable report on The State of HR Metrics and Talent Analytics
Create your profile
Only paid subscribers can comment on this post
Check your email
For your security, we need to re-authenticate you.
Click the link we sent to , or click here to sign in.