A Modern Take on Data-Driven Recruiting from a Google Recruiting Leader

Data-driven recruiting is one of the best ways for recruiters to prove their value to their organization. It’s also one of the approaches world-class recruiting teams use to position themselves as strategic partners within their companies. Building out a data-driven recruiting funnel and approaching recruiting as both an art and a science, allows talent acquisition teams to earn the trust of hiring managers and leadership and proactively shape company growth.

A lot of the metrics and data we use to drive recruiting have been around forever. Metrics like time to hire, offer acceptance rate, and source of hire haven't changed since recruiters first built their Rolodexes (I threw out my Rolodex in 2002, just FYI).

What has changed is the way we’ve used metrics to evolve the recruiting process and improve our decision making.

With that, here are a few modern takes on some old school hiring metrics.

The problem with time to hire

Affectionately known as TTH, this is one of the most fundamental metrics used by recruiters everywhere. The idea is that faster is better. The quicker someone moves through your recruitment process, the better your candidate experience will be and the more time you will have to find qualified candidates to fill other roles.

Sure, that’s all true, but we need to dig deeper to maximize this metric. Simply looking at TTH on its own isn’t very helpful. It tells you how quickly (or slowly) you’re hiring, but it doesn’t tell you how to hire faster.

The key to making TTH actionable is to break it down into its components, examining those for hints on where to optimize your hiring process for speed.

TTH is the aggregation of several steps in your hiring process:

  1. Application to Phone Screening
  2. Phone Screening to On-site
  3. On-site to Offer
  4. Offer Extend to Offer Accept
  5. Offer Accept to Start Date

And it’s important to look at all of these steps on both an individual and aggregate basis when building a data driven recruiting strategy.

Once you’ve broken your process into steps, figure out how many days on average your candidates spend in each phase. This is the Time between Phases. Then, with this baseline data, you can look at each phase and see which ones take longer than they should. This information will help you identify inefficiencies in your process and opportunities to improve TTH. Once you’ve identified your “problem areas,” you can investigate the source of the slow down and take focused action to remove blockers.

In this process, it’s important to brainstorm all possible sources for these “problem areas.” For example, if your TTH is too long, you might assume your recruiters are slow to schedule on-site interviews. But once you dive deeper into the data, you find that the real lag in the process is from Phone Screen to On-site because interviewers aren't giving feedback quickly enough.

Free Phone Screen Template Preview

If you hadn't broken your TTH metric into each step and pressure tested various reasons for the slowdown, you may not have discovered the real root cause.

Breaking TTH into the Time between Phases allows you to target the underlying problem. And once you’ve identified the underlying problem, you can brainstorm solutions and test them to see if they improve your time to hire.

How to optimize your pass-through rate

Another common metric that has long been used in recruiting is pass-through rate (PTR). PTR measures the percentage of candidates who make it through each step of the process. It’s a useful metric for identifying the what - any problem areas where a disproportionate number of candidates fall out of the process.

Let’s illustrate this with some hypothetical numbers. Assume you have a recruiting team that converts 1577 applications into 48 hires this year - that will give you a PTR of 3.2%. As you can see in the image below you can break that down into 5 (or more!) steps and see the conversion rate at each step.

Free Phone Screen Template Preview

When you break your PTR down like this, it’s easier to see opportunities for improvement in your funnel. In this case, you might notice that the conversion rate for Application → Phone screen is a little low at 49%.

Even though you now know what is going on with the data, you don’t know yet know why that number is a little low. To get to the why, you need to explore the factors that affected how many applicants were screened, and brainstorm ideas to improve it.

One example could be that your hiring managers are setting extremely specific criteria for passing an applicant forward to phone screen. You could discuss that criteria with the hiring manager to reveal ways to open it up without compromising the quality of the candidates that get hired.

Let’s say you open it up and now 65% of applicants make it to the phone screen. With that single change you would increase your PTR from 3.2% to 4.0% and fill an additional 15 roles from the same number of applicants!

Free Phone Screen Template Preview

What pass-through rate really measures

PTR measures the past performance of candidates in the pipeline. Because of this, it can be misleading when used to outline the volume needed to hit future hiring goals - if you don’t understand the reasons why your PTR was what it was.

Let’s say your data shows that you hire 2% of candidates who apply for jobs at your company, and you are trying to understand how many applicants you’ll need to double hiring next year. (Quick hint: It’s not double.)

That 2% PTR is how candidates performed last year. Only last year, your recruiters were calibrating. And your interviewers were new. Oh, and your hiring managers switched requirements a few times. And, there were a lot of other factors that affected your PTR.

PTR measures the outcomes of past hiring, which is useful to give direction for future hiring, but only if you identify the causes of candidates passing through (or not passing through). This allows you to identify why your interview process performed the way it did, and helps you improve it moving forward.

Why performance management isn’t just about hitting hiring goals

When I first started to recruit (I was with an agency at the time), we measured recruiter performance based predominantly on simple metrics like submittals - the numbers of resumes sent to hiring managers for review - and hires made. But as our ability to track hiring data on a more granular level, improved, I’ve evolved how I measure performance to reflect a recruiter’s full body of work.

The problem before was that sometimes weaker recruiters would get lucky and hit their hiring goals, despite the fact that they brought significantly fewer candidates through the hiring funnel than usually needed to meet goals. For example, if a recruiter’s hiring goal was expected to require roughly 1,000 applications to fulfill all target roles, but they managed to hit the goal with just 250 applications, it’s possible they got lucky and wouldn’t be able to repeat that kind of performance.

In other cases, strong recruiters have been unlucky, missing their hiring goals, despite doing great work and generating a healthy candidate pipeline.

At the same time, if you had to do double the work to hit your goals, there is obviously a breakdown in the process that required you to fill the pipeline with more candidates than you would have had to otherwise.

The key to properly evaluating performance in this case is to goal recruiters against the full hiring funnel. Measuring metrics like number of phone interviews and number of onsite interviews, in addition to applicants and hires, gives you more transparency into your recruiters’ performance and why they did or didn’t hit their goals.

The poor quality of quality of hire

Quality of hire is one of the new, trendy metrics I’ve been hearing a lot about in the recruiting community. It maps recruiter performance to the performance of the new hires they make.

The intention with this kind of metric is to hold recruiters accountable for the performance of the people they hire. But the problem with it is that it measures recruiters based on a number of factors far outside their control.

There are so many considerations that go into employee performance beyond the quality of the hire. Poor performance of a new hire could be attributed to management issues, team troubles, culture problems, strategic misfires from leadership, or some other situation that causes a new hire to struggle.

For these reasons, I don’t see a lot of value in using quality of hire to measure recruiter performance.

Analyze past performance to improve your hiring process

This is my favorite way to use data.

Taking a backward look at candidate profiles can be an amazing way to get better. Looking backwards shows you trends within the funnel, and can help you learn what a good or bad candidate looks like based on historical hiring. Not only does this approach let you source more effectively and target people you'll be more likely to hire, but it also allows you to take a deeper look at the talent you’ve added to the team and ask if that’s the right talent moving forward.

For example, say you look back at your hiring data and find that a high percentage of your hires that performed best over the past year, lacked a specific technical skill when you hired them or didn’t have a specific degree. Yet, because the team thought these criteria were important to do the job well, your recruiters often rejected applicants because they didn’t have that degree or skill. Once you make the connection in the data that your resume filtering criteria is disqualifying a lot of promising candidates, you could change screening criteria to help bring more viable candidates into your pipeline.

It’s important to segment your candidate data by various characteristics and profile details - such as school, degree, a particular type of experience or previous company type - to help you identify any patterns in your best candidates. Ideally, you want to distinguish the key characteristics that make candidates a good hire for your organization.

Analyzing your hiring data also allows you to myth-bust stories about the hiring process heard around the organization. For example, a hiring manager may say, “We only hire people from the best computer science programs.” By analyzing your hiring data, you’ll be able to see if this is true, or not.

You can compare the PTR data for different pools of talent to either support or debunk the idea that only candidates from the top programs will be a good fit.

By using this data driven approach, you can help the team evolve its hiring process to attract and find top talent. A good way to start is to share insights with your hiring managers and start a dialog around what you are seeing in the applicant pool. Over time this will make you a more successful and trusted recruiting partner.

Refreshing your data-driven recruiting for a better hiring process

When you dig in to define the hiring data that’s most important for your organization, and its individual needs, you’ll put yourself in a position of power. You’ll be the recruiting partner that learns from past hiring patterns and uses numbers to back up their insights.

Make sure as you build out your hiring funnel that you look below the surface to understand the inputs that drive your numbers and tell you why your data is behaving the way it is. Pairing the data a good applicant tracking system makes easily available, with the insights that google analytics can give you into traffic and behavior on your career site, can help you make the data informed decisions that will facilitate the best hiring decisions possible.

By using your hiring data in a thoughtful way you can make your hiring process more efficient, and hiring easier for your team.

About Hire by Google

Hire is a recruiting app by Google that uses AI to make the hiring process faster and simpler. Because it is designed specifically for G Suite users, with Gmail, Google Calendar and other G Suite integrations, Hire streamlines administrative tasks so that your team can hire the best people, faster.