SJV Connect
SJV Smart

AI & Natural Language Processing: The Next Frontier of Resume Verifications


The turn of the New Year has got us thinking about where our industry might be headed in the next decade. It’s also got us looking back. It’s amazing how far call center technology has come since the dawn of the digital era.

Not too long ago, a team of employment verification specialists was essentially a group of people sitting in front of telephones, dialing manually. Calling patterns for different clients were tracked with hefty written instructions — or walls of sticky notes.

For many consumer reporting agencies (CRAs), this is still the case. Here at SJV Data Solutions, however, we’ve always been fascinated by the transformative potential of modern call center technology. We invested millions in building an employment verification call center that mirrors the customer service departments of leading corporations such as Best Buy.

Thanks to tools such as call trees and call flows, our callers are never at a loss as to what to say when verifying a job candidate’s previous employment, who to contact, and when to call. The system routes the call to the right caller and guides the researcher through the interaction.

The time required to onboard new researchers is one of the primary reasons CRAs struggle to scale their verification businesses. It can take up to six weeks to get a new employee up to their full capacity.

With our modern calling system, on the other hand, we can take someone with zero experience in the verification industry and have them on the phone successfully calling on orders within three days.

What’s next in resume verifications?

So, where do we go from here?

For the verification team at SJV, improvements often come from digging into the data. As an outsourced employment verification provider, we have insight into the processes of hundreds of different CRAs. Because the underlying technology we use and the calling team is the same for every CRA we serve, it’s easy to make one-to-one comparisons.

For example, by analyzing the data, we’ve learned:

  • switching from three calls over three days to four calls over four days can result in a 5% increase in completion rate.
  • switching to four calls in three days (so, more than one call per day), you’ll see a 2.5% to 3% increase in completion rate. 

In other words, you can pick up 2.5% on completion rate by making 4 calls in 3 days without increasing turn around time.  But increasing to 4 calls in 4 days gives a 5% improvement, but at a cost in turn around time.

Data-driven findings like the one we just described help us and our clients strike the proper balance between speed and accuracy to achieve our clients’ goals. We expect to make similar discoveries as we gather and interpret more data over the coming years.

Looking farther ahead, we also believe that technological advancements will revolutionize the verification industry during the 2020s. One area we’re watching closely combines artificial intelligence (AI) with natural language processing (NLP).

How can AI and NLP make resume verification call centers more productive?

Before we proceed, a quick caveat: Everything we describe below is purely conceptual. Neither we nor anyone else in the employment verification business has integrated NLP/AI technology into their calling systems. But we are intrigued by the possibility.

If you have an Amazon Alexa or Google Nest device in your home, you are familiar with a rudimentary version of AI and NLP. Typically with those devices, you ask a question using natural language, and the device responds.

The way AI and NLP may be used in call centers is the opposite. The computerized agent will initiate the call, converse with a human source, and record the relevant information.

The technology is not quite ready for primetime. We already know that people are reluctant to return voicemails from robotic callers. Many find the artificiality of the experience off-putting.

Google’s Duplex software shows promise. It uses a very human-sounding virtual assistant to make reservations and set up appointments on users’ behalf. Unfortunately, Google does not currently license this technology to call centers.

We think that, sooner or later, AI and NLP-powered calling systems will become commonplace in the employment verification industry. And when the technology gets closer to maturity, we will be among the first to explore it in greater detail.

Imagine how much more efficient a team of researchers could be with digital assistants making most of the calls. We can envision a process in which the AI makes the first two dials — all that is needed for about 70% of verification calls — and human specialists handle the rest.

Incidentally, at SJV, we’re already experimenting with artificial intelligence for different purposes. For example, for whatever reason, many job applicants misspell the names of their former employers. We’re working on an AI tool that can quickly determine the correct spelling, so we know who to call.

The future of employment verification services is now

Artificial human-like callers may be a ways off, but your CRA can benefit now from the scale, technology, and experience of SJV’s cutting-edge call center. Contact us here to talk about how we can help your business take on more orders, satisfy more customers, and grow or check out our new ebook to gain a business perspective on employment verification services.


employment verification services