How would you measure the effectiveness of our employee referral program?
Google- Rohit Kumar
Context
First off let’s just establish a common understanding of what the Employee Referral Program is at Google and why users’ use it. Google employees are able to view a list of available job opportunities at the company and refer their friends or previous coworkers by submitting their contact information and resume to an opening. This referred user’s application is then denoted in the list of applicants as a referral and recruiters can view it then follow up accordingly.
It’s important to take a step back and think about the bigger picture. What is Google trying to accomplish as a company and how is an employee referral program going to help further that. Google’s mission statement is to organize the world’s data and make it universally accessible and useful. In order to accomplish this Google is going to need the best and brightest minds. The employee referral program aims to help Google increase it’s ability to find those best and brightest minds.
Users
There are a few different users at play here and we should be considerate of the different roles they place when discussing metrics:
- Referrer – This is the current Google employee whose submitting the referee’s application
- Referee – This is the person whose application is being referred by the referrer
- Recruiter – This is the one who reviews resumes and schedules interviews
Metrics
In terms of measuring the Employee Referral Program’s effectiveness, I’m going to pose some key questions and then list out metrics to measure those questions.
The first question I would have is are employee’s using it? This is a pre-requisite for effectiveness, if no one is using it then it isn’t effective. We could measure adoption of the Employee Referral Program via:
- Total Number of Referrals per month -> This gives us a good absolute measure of how much usage the program is getting
- Average number of referrals per employee / month-> As opposed to an absolute measure this is relative, which will help us measure usage independent of total employee count increasing or decreasing
- % of referred applications that get reviewed by a recruiter -> We want to make sure the referred applications we’ve gone through the effort of acquiring are actually being processed downstream and aren’t just being submitted into the void
Now that we’ve looked at metrics to establish a base level of usage, next I would look at how effective is it? More precisely, is the Employee Referral Program effectively helping Google find and hire quality talent?
- Do the referral’s pass the hiring bar and do they want to work here? % of Referrals who are:
- Interviewed
- Extended offers
- Accept offers
- Average performance rating of employee’s who joined via referral -> How do they perform relative to their peers once they’ve been hired?
- Average tenure of referred employees -> Do they stay with the company long enough to add value and contribute in a meaningful way to Google’s mission
If we know the Employee Referral Program is effective, the next thing I would want to assess is it efficient / optimal relative to other options.
- Average referrer payout per hired referee -> We want to be consciencious of any additional costs we incur as a result of this program
- Average decrease in employee time spent successsfully hiring referred candidates vs normal candidates -> Our employees spend a lot of time vetting candidates if we can reduce the amount of employee time needed per hire then that is a huge benefit and additional cost savings
- % of Referrals who had previously applied to Google -> Are these referrals new additions to our recruiting pipeline or are we relabeling an existing portion of our pipeline and paying a fee to do so?
Downsides
I’m confident in the ability of our proposed set of metric’s ability to measure the effective of the Employee Referral Program, but with that said not set of metrics is perfect so it’s worth spending a few moments discussing any potential downsides.
- While intentional, all of our metrics are very high level. They don’t give us any granular insight into areas where the employee referral program might be more or less effective. For example, it could work great for hiring sales people but terribly for engineers.
- We never addressed the quality of the referrals being made. Is the person being referred your previous co-worker who is a rockstar or a random person you never met who you found via Blind?
- I would like a better handle on whether or not these quality candidates who were referred would fall through the cracks without the referral. Is the program helping us more efficiently find quality hires or is it kind of redundant?
Summary
In order to measure the effictiveness of Google’s employee referral program, we’ve brainstormed a set of metrics that help monitor and address several key questions:
- Is the Employee Referral Program actively being used?
- Is the Employee Referral Program effective at helping Google hire the best and brightest?
- Is the Employee Referral Program efficient and optimal when compared to other recruiting methods?