Seize the Data - Recruiting Metrics & Analytics

[Former Member]

In 1999, I was studying for a Masters degree in Political Science and answered an ad in the Washington Post (hardcopy edition, of course) for a research associate position – little other information was offered in the ad – at the Corporate Executive Board. The best practice case study firm had recently gone public and was in the midst of tremendous growth.

Shortly after a two-step hiring process – a telephone screen and single interview at CEB’s then headquarters (the Watergate building, karma to a student who was writing a thesis on President Richard Nixon) – I was hired into the general research pool and assigned to Corporate Leadership Council, the HR research business unit. Still, I knew little about the company I would be joining, other than the details included on the company's website; despite that, I had a tremendous career with CEB and greatly appreciate the coaching and training received while there.

Of course, much has changed since 1999 with respect to how organizations’ engage and recruit new employees, much of it a consequence of new processes (behavioral interviewing, personality tests, etc.) and technologies (using social media to broadcast the open position, LinkedIn to source passive candidates, recruiting marketing to interact with pre-candidates, etc.).

From my workforce analytics perspective, it made me wonder whether recruiting metrics have kept pace with process and technology transformation? Or are firms still largely reliant on traditional measures of recruiting pipeline management?

This is the second in a series of blogs that endeavor to share some ideas on current approaches to measurement in specific talent management domains, and is written for the practitioner in that field. The first blog, on learning analytics, can be found here. 

Contributing to this blog are two colleagues who are well-versed in recruiting trends. Will Staney is a Director of Recruiting for the SuccessFactors Talent Management organization, while Brent Ellis a Director of Recruiting Product Strategy & Sales at SuccessFactors.

1. What’s conventional practice for using data to measure the impact of recruiting?

Traditionally, most SuccessFactors Workforce Analytics customers have focused on integrating their other talent management data (performance, engagement, mobility, diversity, etc.) with measures of hiring activity and process effectiveness:

• External recruitment measures relate to hiring activity in terms of the volume of new recruits from internal and external sources, as well as the relative growth or shrinkage of the workforce.

• Effectiveness measures relate to the mechanics of the recruitment process, such as screening, interviewing and making employment offers, as well as the delivery of quality candidates to hiring managers.

As concisely described in the SuccessFactors Staffing Metrics Pack (part of the Workforce Analytics application), such measures can provide a wealth of important information, including:

• Demonstrating the value of the entire recruitment process and function.
• Supplying a vivid picture of recruitment costs, as well as the expected and actual outcomes from these costs.
• Providing a picture of how the organization has changed over time.
• Identifying potential talent management risks.

Of course, poor data quality can inhibit any effort to provide solid recruiting metrics. “Put yourself in the shoes of your Head of Recruiting,” says Will Staney. “They have to use the numbers coming out of the applicant tracking system but are well aware that the raw data is prone to human error. A classic example of this is recruiters who must choose the candidate source (referrals, job boards, career fairs, etc.) from a drop-down list; if they don’t know the source, or if it isn’t listed, they probably just select one at random.”

2. What’s missing from this approach?

Assuming that your organization has the ability to mine its applicant tracking data (in contrast to the VP of HR at one Financial Services organization who told me that “we have lots of recruiting data…we just don’t use it”), two obvious challenges are:

• Internal - Misaligned Functional Objectives: Current metrics tend to focus on the latter stages of the hiring process – the volume, quality, time, and cost of candidates having submitted applications. However, as Will Staney says, “early stage sourcing is the really challenging part of recruiting – it tests the effectiveness of outbound marketing. Locating and engaging candidates can mean that 70 percent of recruiting is done before the candidate physically applies for a position.” Put yourself back in the role as Head of Recruiting – are you holding your recruiters accountable, and using metrics that reflect, activities conducted across the full process of recruitment marketing (sourcing) and execution (hiring)?

• External - Lack of Visibility into the Recruitment’s Impact on Organizational Goals: Brent Ellis, who works with the SuccessFactors Recruiting Execution product, makes the point that “today, less mature recruiting organizations seek to address the question of ‘how well are we hiring?’ whereas the future question will be ‘what types of people will we need to recruit to best execute our growth strategies?” Firms may approach recruiting as a knee jerk reaction to a vacancy, engaging temp agencies or kicking off a replacement search – the Financial Services VP referenced above explained that one BU has identified 125 jobs that require strategic sourcing, with no action taken thus far to do anything about them. Turn that around – if you knew how many staff your firm was liable to lose in any given quarter, could you better predict how many candidates you would need in the pipeline to replace them (should they all depart)? It would certainly reduce the time to full productivity of incoming hires.

3. What might be examples of foundational metrics to apply to recruiting?

Beyond the basic metrics (cost per hire, time to fill), consider a range of metrics that illustrate the depth of the internal/external recruitment pipeline:

• On-Time Talent Delivery (how many recs were filled within the preferred timeframe)
• Offer Acceptance Rate (how many offers are accepted/declined)
• Top Candidate Acceptance Rate (how many of the best candidates for the job actually accepted the offer)
• Recruitment Source Breakdown (the composition of external hiring sources)
• Recruitment Source Ratio (the number of positions filled either by internal or external candidates)
• Referral Rate (the number of positions filled by internal referrals)
• External Hire Rate (the percent of the current workforce hired from external sources)

4. What about more advanced analytics?

In order to provide better visibility into the entire sourcing/recruiting process and organizational impact, here are a couple of suggestions for ideas/frameworks to use:

1. Capture data from a recruitment marketing platform that indicates the robustness of your sourcing and external talent cultivation activities; for example, what proportion of hires are sourced from your talent communities or career sites?
2. Put as much effort into collecting data on internal recruits as those sourced from outside your organization. For example, use employee engagement/satisfaction data to evaluate the propensity of employees to switch jobs or career tracks.
3. As loosely discussed in Kyle Lagunas' blog on Recruiting Analytics: The Next Big Thing in Talent Acquisition and Stephan Millard in Big Data Brewing Value in Human Capital Management, leverage big data in real-time (where relevant) and present the information in a meangingful way to recruiters that speeds mid-course adjustments to sourcing activities.
4. Over-invest in using data science to attack a subset of business/talent problems. In his blog on How Google is Using People Analytics to Completely Reinvent HR, Dr. John Sullivan shares how Google developed an algorithm for predicting which candidates had the highest probability of succeeding after they are hired. Its research also determined that little value was added beyond four interviews, dramatically shortening time to hire. This data is of value both to the company (building a strong talent bench) and the HR function (improving process efficiency). Who doesn’t like a “win-win” situation?

5. Any final words of advice?

Regardless of what kind of work you perform, information overload is a real threat to productivity and decision-making; we are bombarded with data on almost every conceivable subject. Before diving into processes for collecting, aggregating, and synthesizing data, take a step back and ask “what 3-5 metrics will offer our recruiting teams the best insights into the sourcing/recruiting process?” Start small and deliver value. 

In my next blog, I will take a look at workforce analytics in the realm of Performance Management. Enjoy the final days of summer!