Emerging Technology Collects, Matches Job Candidate Information for Recruiters

Symphony Talent CEO shares excitement about innovations in the use of talent data

Roy Maurer By Roy Maurer November 29, 2017
Emerging Technology Collects, Matches Job Candidate Information for Recruiters

​Roopesh Nair, CEO of Symphony Talent

Many recruiters today are collecting data on spreadsheets, if at all, and creating reports on measurements like cost-per-hire and time-to-fill. Soon, though, they'll use tools that match candidates to open jobs, personalize candidates' job seeking experience and optimize the hiring process in real time, leaving recruiters to focus on the parts of their job that require human interaction.

Roopesh Nair is CEO of Symphony Talent, a New York City-based recruitment technology firm that builds engagement between a company's employer brand and job candidates. He recently discussed innovations in the use of talent data with SHRM Online.  

[SHRM members-only online discussion platform: SHRM Connect]

SHRM Online: What's the leading-edge technology for candidate data?

Nair: Before you can even talk about the use of data, you must first address a critical aspect which is just emerging. You must have a candidate's full data set in one place to be able to make meaningful sense from the connected data. This means being able to follow a candidate on his or her job search, careers sites browsing, clicking on job posts, responding to recruiters' outreach and so forth. One of the current complexities of dealing with candidate data is exemplified by visualizing the candidate as a whole pie, and the talent acquisition ecosystem with its various solutions as having visibility into different parts of the pie. For example, Indeed has insight on how people search for jobs. A CRM [candidate relationship management] platform has information about how people are engaged, the open rates on outreach e-mails, and what type of content is most effective to attract candidates. ATS [applicant tracking system] providers have visibility on the transactional part of the process. Being able to collect all of that information and then self-optimize the recruiting process based on that data is an exciting innovation in the use of talent analytics.

Employers are currently reviewing data to see what has worked and what didn't work and planning on what they'll do next quarter or next year. Many organizations are still struggling with even this, going into analysis paralysis over what is the best source of hire, for example. We're on the cutting edge of having the data determine the best response in real time, for example, analyzing which touchpoints work best to bring candidates in and reallocating recruitment advertising spend based on that analysis.

Another innovative application of a candidate's full data set is to create a personalized experience for each candidate. Companies can assign candidates a certain persona based on their online behavior, their resume, or whatever information the employer has on them. Interested job seekers can then be served the content they would rather see than have to search for it. For example, there's a nurse who values benefits. Why have her search for the benefits page on the company's website and click through menus to find the information she's looking for? There's a good chance she'll give up and never find it. Why not take the content each candidate values and bring it to the home page? Or plug it into the job posting?

SHRM Online: So, each candidate would have a tailored experience searching for a job? Each person would see something different when they clicked on a company's careers site or job ad?

Nair: Yes. You match the attributes of candidates with the content that fits with those attributes and then put it together like a puzzle. The core content would be the same, but a video would pop up for candidates interested in X, or branded content would display for another candidate who has shown interest in Y.

Job seekers are craving rich, personal engagement and technology, and employers can deliver it by aggregating content specifically created by an agency or the employer or curating it from the web. The content is collected and pushed through a CRM based on the preferences of the candidates, making it more meaningful and less random or generic.

Another emerging application of data is being used to help recruiters and hiring managers make decisions on the best candidates to hire. Typically, recruiters start a search to fill a role looking at the candidates who applied, and searching for resume keywords. We can do more than just keyword-based searching thanks to the availability of data and the application of machine learning systems which can automatically determine better matches. A job has a certain set of attributes, and a candidate has a certain set of attributes. Based on everything known about both the job and the candidate, systems can present all of those who are in the 90th percentile match range (as an example). It also eliminates a lot of human error from this part of the recruiting process, allowing recruiters to spend more time building relationships and closing the deal with candidates.

SHRM Online: At what stage should employers begin analyzing talent data?

Nair: As early as possible. Anyone who touches the employer is a candidate. If you click "like" on a social post, you are a candidate. That doesn't mean a recruiter needs to start looking at all of those people, or that HR should get involved, but that's where data can begin being collected. Even minimal information like how to reach someone and what type of job they would like is important for future engagement. We're not too far away from that type of initial engagement being completely automated. If you have technology to get you to the point where recruiters only become involved when there is a job-candidate match, that's where you want to be.

SHRM Online: Should data be used to make the final hiring decision?

Nair: That's a little more complex, and depends on the company. If you're looking for entry-level hourly workers and not really interested in cultural fit, you can hire prospects immediately, based on data output. But for those companies that prioritize culture fit, that hasn't really been cracked yet by data analysis. Personally, unless I actually spend time with someone, I wouldn't feel comfortable making a decision to hire them. Data can tell you that there is a fit based on certain dimensions, but I think there should always be a real human connection made before making the final hiring decision. The information gathered from pre-hire assessments can be criteria for sorting through the candidate pool, for example, but I'm not sure it can be the only point for the final decision, even with the best machine learning and analytics backing it up.

SHRM Online: What will candidate data analysis look like in a couple of years?

Nair: Adoption may take longer, but I think that in the next couple of years, the possibility will exist to automate all decisions that can be automated. People will start trusting the decision-making being made by systems. Artificial intelligence will get better and better as it learns.

I think recruiters are currently overwhelmed by all the sourcing channels available. They focus more on that part of their job instead of what they should be focusing on—building the relationships and closing the deal. Data analysis is one thing that machines can do very well and it would benefit the industry to let them do it.

Was this article useful? SHRM offers thousands of tools, templates and other exclusive member benefits, including compliance updates, sample policies, HR expert advice, education discounts, a growing online member community and much more. Join/Renew Now and let SHRM help you work smarter.


Job Finder

Find an HR Job Near You
Search Jobs


Find the Right Vendor for Your HR Needs

SHRM’s HR Vendor Directory contains over 10,000 companies

Search & Connect

HR Daily Newsletter

News, trends and analysis, as well as breaking news alerts, to help HR professionals do their jobs better each business day.