The Internet has strong opinions about reference checks:
- “Anyone who’s been in the profession for more than a minute or two knows deep down that references suck as indicators.” - Ere
- “When it comes to supplying references, we as human beings make only one choice--we supply the contact information of people we know will say wonderful things about us.” - Inc
- “(in most cases) a monotonous, time-wasting process that offers little value in its execution.” LinkedIn Talent Blog
- “You’re an Idiot if You Still Check References!” - Tim Sackett SPHR
Oh, and my personal favorite:
- “It’s a waste of time, an archaic practice that should have died years ago” Ed Nathanson, VP Talent
For the most part, we agree!
Yes, as a company whose core product creates reference data… we agree with 90% of the sentiments above.
That’s because the traditional 19+ step process for doing reference checks is too outdated, too tedious, and rarely provides any meaningful value for the time spent.
But if you do references correctly, you will reap the benefits.
Jack Altman, CEO and Co-founder of Lattice says:
"I've come to believe that one of the most underrated secret weapons for hiring is reference checks. Reference checks done well can contain real signal that can help you avoid making costly mistakes. Reference checks can be an extremely valuable part of your candidate evaluation process because they can distill years of data that you simply can’t reproduce in even the longest of interview processes."
Harvard Business Review reports that taking the time to hear from a candidate’s prior managers and colleagues often “yields vital information” because “relevant external observers are in a better position to give you an accurate estimate of whether the candidate will be able to perform.”
So what’s the rest of the population missing? Well, turns out that there are glaring reasons that make references painful to include in a hiring process:
- They take too much time (to execute and coordinate)
- The data is text heavy
- The data is collected and never used again
- They are done too late to impact the interview process
- They rarely disqualify candidates
- References have a positive bias
- No standardization (of questions, reference requirements, etc)
What a list! No wonder so many people believe that the reference check as we know it today needs to die already.
Enter Searchlight.
We make reference checks relevant again.
In the words of a Chief People Officer at a publicly traded company with 1000+ employees:
Searchlight is a brilliant complement to our existing TA stack for the purposes of further integrating Diversity, Inclusion, & Belonging, turning reference checks into a truly useful stage of our process, and bolstering our structured interview model.
Some outcomes we’ve helped drive for our awesome customers:
- Zapier replaced an entire interview round (saving them 3 hours of interviewing time per hire) with the Searchlight reference assessment because our reference data was faster to get and more predictive.
- Doximity reduced the number of top candidates lost to competing offers by 12% because they no longer have to wait for references to come in before they make an offer. (Doximity invites candidates to Searchlight in tandem with on-site interviews).
- Human Interest increased their reference completion rate from <50% to 93%.
- As a baseline for all of our partners, references submit their responses within 36 hours.
How do we do this?
Let’s walk through how our platform addresses all the reasons why the old way of doing references was so terrible.
1. They take too much time
We’ve heard from countless teams that conducting 3 references the old way would take 3 hours between scheduling the calls, conducting the calls, and then documenting the notes. On average, this process would take 7 days to complete. And woe to those trying to schedule reference calls before holidays!
Do you know how much time it takes a Searchlight user to conduct 3 references? Two clicks. And our average reference turnaround time is 36 hours.
Literally.
Let me show you.
Here’s EVERYTHING a recruiter needs to do to request references from a candidate:
In Greenhouse:
In Lever:
We integrate with several other ATS as well, but you get the point.
The recruiter makes those two clicks, and within 36 hours, reference data will show up in the ATS.
No calls. No phone tag. No rescheduling. No note-taking.
Just two clicks.
2. The data is text heavy
Let me show you, rather than tell you.
Here’s a snapshot of a Searchlight scorecard for a candidate we interviewed in early 2021:
Can you see why we weren’t too optimistic about their ability to thrive on our early stage team? Hint: we deal with a lot of ambiguity.
Searchlight also quantifies each candidate’s attributes so we no longer have to guess what their strengths and areas for growth are. We let multiple references validate this for us:
And, we can visually compare a candidate’s working styles against the hiring manager’s existing team composition:
3. The data is collected and never used again
Searchlight makes reference data easily accessible. Our reference data is automatically uploaded to the ATS so teams don’t have to “dig” anywhere to find the reports. In fact, hiring managers often tell us, “When I’m making a final decision on a candidate, I pull up the interview notes and the Searchlight reports side-by-side so I get the whole picture.”
Not only is our reference data used during the final debrief before a candidate is hired (more on this in the next point), our customers also use the data to help speed up onboarding for the new hire. Onboarding is a crucial period in the employee experience, nailing the onboarding experience has enormous impact on retention and productivity. Don’t just take our word for it. Harvard Business Review shares that the best reference checks focus not only on the candidate’s strengths, but also “provide perspective on how you can support the person if you make the hire.”
We believe that onboarding is critical, which is why our platform automatically generates onboarding guides (sample PDF) and sends these reports to the manager to help kick off the relationship with their new hire.
Taken in aggregate, our reference data is even more powerful. By having talent profiles for all a company's hires and differentiating those that become top performers, Searchlight creates a dynamic model of a company's hiring patterns and surfaces insights on how to hire better.
4. They are done too late to impact the interview process
If you’re doing references after the offer is already made -- yes, of course! Then you’re running a hiring process that excludes reference data by design.
But with Searchlight, references only take 2 clicks and are completed within 36 hours 90% of the time. As a result, most of our customers opt to initiate Searchlight references at the same time they start coordinating the final panel interview.
As a result, teams regularly leverage our data to assess team fit questions like:
- Does this candidate have the strengths best suited for this role?
- Is our team equipped to address the growth areas for this candidate?
- Are their working styles conducive to doing this job well?
Our platform even helps quantitatively guide the team-fit answer by computing a match score!
5. They rarely disqualify candidates
Searchlight helps companies get the best signal on their candidates to make the best candidate-team match. We don’t make it our goal to disqualify candidates, but as a result of providing consistent, quantified, and detailed data earlier, our customers make more informed decisions and end up saying “no” when interviews make them lean “yes”, and vice versa.
We hear countless stories like these when Searchlight prevents a mishire:
- There was one candidate who was a strong yes from the entire interview panel. His resume was stellar. Searchlight data pulled out insights that showed he wasn't the right fit for the role. For example, his strengths were in strategy, but I was looking for someone who is actually looking to roll up their sleeves. It was a hard decision, but when people asked me why, I could point to Searchlight and say because of X, Y, Z. We dodged a bullet." - MasterClass Hiring Manager
- The whole point of the interview is to sell yourself and what I’ve noticed from Searchlight’s results is that people tend to be more honest and realistic on attributes and areas of improvement...after seeing it in practical practice, it helped us dodge a couple bullets and probably make some better decisions on who we are hiring! - DoorDash Hiring Manager
What is the cost of a mishire for your team?
Here's some food for thought:
I’d rather interview 50 people and not hire anyone than hire the wrong person.
- Jeff Bezos (Amazon CEO)
On the flipside, we also get happy customers who find a hidden gem who would have normally slipped through the cracks:
- We used Searchlight and a candidate we were going to pass on was ranked first based on Searchlight’s rules engine so we brought them in and ended up hiring them. They’re now a top performer. - Snapdocs Hiring Manager
- There was a technical candidate in which our hiring panel wasn't sure the candidate thought about code quality the same way our team did. In all the interviews, not a single person got a sense of whether this engineer would help us reduce the number of rollbacks. But, in the Searchlight report, a reference said that the candidate single-handedly designed and implemented the whole testing framework for the company because she realized it was the thing holding them back. Because of Searchlight, the hiring panel hired her. - Standard Cognition Hiring Manager
What is the value of finding a great hire that you normally overlooked? Some would say it's priceless.
6. References have a positive bias
After gathering 100,000s of answers on our platform, Searchlight’s response to this objection is “yes, and no.”
Remember one of the quotes I mentioned in the beginning?
When it comes to supplying references, we as human beings make only one choice--we supply the contact information of people we know will say wonderful things about us.
We dug into 60,000 data points of how references rated their colleague to see if references are actually always glowing:
As you can see, a reference only rates a candidate as “The Best I’ve Seen” 22.8% of the time.
Even if we combine that with the “Exceptional” response, that’s 71.5% of total responses. That means that nearly 29% of the time, references AREN’T saying wonderful things. This is a lot higher than most of us would expect.
References are less positively biased than you might think.
And even when bias might be happening, we make it easy to catch it through visualizations and corroboration:
The “Peer” for this reference clearly has a different perspective than the Supervisor and Manager. Why might one reference say wonderful things across the board?
Digging deeper, the answer becomes more clear:
The reference who was overwhelmingly positive only had 1 year of work experience versus 10+ years of experience from the other 2 references! It becomes easy to see that the years of work experience is probably a factor in the reference’s evaluation of the candidate!
John, the peer reference, may have rated this candidate as “Best I’ve ever seen” across the board because that truly might be the case given his limited work experience.
After seeing this happen multiple times, our engineers got to work. Our platform automatically weighs reference feedback based on corroboration and relevant details about the reference, like how many years of experience they have, how long they worked with the candidate, and how long it was since they last worked with the candidate. Our algorithms are better than humans at distinguishing between good vs. great, and we provide alerts to help our users tell the difference.
7. No standardization
In addressing the first 6 concerns, I’m sure you can see how the digital nature of our platform lends to much more structured data and visual outputs.
Another way we add structure is through reference requirements:
Teams can specify how many references are required and specific working relationship types for every level of employee to ensure uniformity.
Reference questions are determined with recruiting leadership beforehand so references get a consistent experience.
Some teams make customizations by department, allowing them to straddle the middle ground of structure and personalization.
Conclusion
As a result of all these innovations on the traditional, archaic process of reference checks, Searchlight has overturned many, many skeptics.
I was a bit of a naysayer in the beginning. Biggest concern was that in this format, people would not actually get into a lot of detail on the candidates or any of the things they could actually work on. But the feedback has come back very well. They have kind of answered a lot of questions I didn’t expect them to, everybody gave great feedback. *DOUBLE THUMBS UP*
- Hiring Manager at KeepTruckin
I’ll admit to it, when I was first pulled into this, I was like, “What are we doing…??” but now, after seeing it in practical practice, what I’ve noticed from the [Searchlight] results I’ve seen in this process is that people tend to be harsher judges; they’re a little more honest and realistic on attributes and areas of improvement.
- Hiring Manager at DoorDash
We had been a little hesitant about automating the reference process for executive roles. We recently ran a test for a (L6) Director level role and the team really enjoyed the process and results, so we're rolling Searchlight out to all executive hiring moving forward.
- Zapier Recruiting Team
It’s time for traditional reference calls to be replaced by a reference experience that actually moves the needle for modern day recruiting and hiring.
And there's no time better than now, when we're increasing hiring remotely, in new geographies with different cultures, without ever meeting someone in person. Interview are ripe with bias, so rather than interview someone 12 times, learn what it's actually like to work with them through the people who've worked with them before.
Is your team ready to transform your recruiting process?
.png)
