[Update 2019: The book is now live and is receiving rave reviews! Check it out here.]
Over the past month I’ve been working with a variety of vendors to better understand their artificial intelligence (AI) capabilities as part of the research for the book. If you’ve been hiding under a rock for the last year, you might have missed this new wave of marketing hype that threatens to overwhelm us all. Despite some of the overblown comments and ideas from the general population, there truly are some amazing AI capabilities in the marketplace that are worth mentioning. They run the gamut from video interviewing tools to screening and matching applications. This post isn’t an all-inclusive list–just a small sampling of what’s available out there as a sort of guide for those of you looking to take advantage of these new types of tools, listed in no particular order. And if you have a solution you’d like to see added to the mix, be sure to comment on our LinkedIn post so you are on our radar.
The Restless Bandit system is designed to do one thing really well: resurface qualified candidates that may be buried in your applicant tracking system data. Any recruiting leader knows that over time, the ATS becomes a storehouse for all kinds of junk, including resumes from ten years ago that are irrelevant because they don’t have current data. The system removes duplicate entries from the ATS, pulls in the person’s latest work history data from the web, matches the person against potential openings, emails the most qualified candidates, and follows up with those top candidates by retargeting them across Facebook and Google ads.
Beamery is a recruiting CRM that focuses on building better, more human relationships with candidates through the use of its platform. The system’s AI components (Watson and Sherlock) help prioritize candidates based on qualifications, suggest the best times to reach out to candidates for optimal responsiveness, and trigger automated reminders to get high-quality candidates back on the recruiter’s radar (a common problem when carrying a heavy requisition load). While the system has other capabilities, these are the core elements driven by the AI components.
HiredScore performs automated matching, sourcing, and candidate screening with a special algorithm designed to prevent bias and adverse impact to candidates. One of the unique features of HiredScore is the deep integration to post-hire systems to track performance, retention, etc. for a better feedback loop. All too often talent acquisition vendors stop measuring when someone is hired (or not), but HiredScore continues to measure that person to determine their ultimate contribution to the business.
Swoop Talent presents a single source of data by integrating your various inputs, and the use cases for the tool are nearly limitless. For example, let’s say you are losing a few of your software engineers to a competitor. The system allows you to “follow” them by seeing their current role, skills being acquired, etc. by pulling in current data about the individual. Later, you can approach and attempt to rehire that individual based on any new skills that might make them a stronger candidate at your organization. Again, just one example, but one that is valuable since boomerangs or alumni hires are typically shown to be a quality source of candidates.
Virtually everyone is familiar with video interviewing solutions and their general capabilities. However, mRoads uses machine learning to automatically review recorded candidate interviews to see if there is any cheating or other strange behavior that is worth exploring. For instance, the system can tell if the person is looking away from the screen regularly (reading cue cards?), if another voice is on the recording (phone a friend for help?), etc. These signals trigger a flag on the video as a potential source of misconduct.
Textio offers an augmented writing system that helps recruiters to craft better job advertisements. The system analyzes millions upon millions of job ads and hiring outcomes to find out what words, phrases, etc. are more likely to succeed in connecting the company with the right candidates. The scoring system helps recruiters to see how fairly minor tweaks can lead to better hiring results. Clients of Textio are seeing not just higher quality candidates, but faster time to fill as well–a double bonus.
Despite their popularity, many bots are not yet incredibly sophisticated because they are simply a Q&A interface where the bot pulls from a library of preset responses. However, Abby has a really interesting use case worth mentioning. One of the company’s clients needed to hire individuals with a specific social following in order to boost the company’s brand awareness on social channels. During the application process, the chat bot would interact with applicants automatically, but it also asked them about their preferred social channel and specific handle or username. It would then automatically pull data from the social network on the candidate’s frequency and recency to see if there was suitable social capital to make the person a fit for the job. This automated piece is not the only AI-driven aspect of Abby, but it was the one that truly made me believe that bots can do more than serve as a FAQ for candidates.
randrr is an interesting solution because it focuses on the career opportunities beyond what you might see in a normal job board. By analyzing millions of resumes and job transitions, randrr hopes to be able to support candidates by giving them more valuable insights into potential career paths and opportunities. Instead of what many career pathing solutions offer (internal-only looks at job opportunities), randrr helps candidates see that maybe leaving and coming back to the company would enable them to make that leap to the required skills necessary to accomplish that next step up the career ladder. By the way, if you missed it I recently interviewed randrr’s founder and CEO, Terry Terhark, on the We’re Only Human podcast.
As I mentioned above, if your system, whether in talent acquisition or otherwise, is using some type of machine learning, natural language processing, or other AI-based technology, please feel free to leave a comment here. Thanks!
Ben Eubanks is the Chief Research Officer at Lighthouse Research & Advisory. He is an author, speaker, and researcher with a passion for telling stories and making complex topics easy to understand.