AI Will Dominate Every Element Of Recruiting – A Snapshot View Of The Future Of Recruiting

Ouch, hiring needs AI because 45% have rejected jobs after bad interviews, and 46% of new hires fail. Because of these stumbles CEO’s rank recruiting as the #1 overall business issue holding their organization back (source). Currently, the recruiting function makes lots of errors, but it has no formal process to systematically learn from each one. In direct contrast, AI’s primary charge is to focus only on learning. Learning literally from every failure and success. And because learning and then changing lead to more rapid continuous improvement. AI usage will proliferate throughout recruiting.

This is a think piece… for advancing your thinking about AI’s role in recruiting.

Start By Understanding Why AI Will Quickly Dominate Recruiting

It’s a major strategic mistake for smart recruiting leaders to assume that the approaching AI and machine learning wave can be held back. Five of the primary drivers behind this impending wave of change include:

  • Adopt AI everywhere because members of the executive committee expect it – almost everyone in a leadership role now realizes that AI has recently become the “go-to application” for solving a wide range of business, scientific and medical problems. And the sudden popularity and success of ChatGPT have further pushed everyone in business realize that it’s now time to embrace AI in every aspect of their work.
  • Adopt AI because the recruiting function needs to make up for its past major talent shortcomings –as I noted previously. CEOs view recruiting as holding their business back due to various recent strategic shortcomings. The primary shortcoming is our failure to provide our organization with a competitive advantage approach. So we didn’t supply it with sufficient quality and volume of talent that was needed just to maintain our current level of operations throughout the business. In addition, our painfully slow hiring processes have hurt us by unnecessarily keeping critical positions vacant for months. Add to these shortcomings our inability to hire sufficient diversity talent. Finally, and more recently, our failure has been the inability to measure and warn executives that we were significantly “overhiring.” A practice that is now forcing us to implement painful and costly layoffs.
  • Adopt AI because we need to dramatically improve our recruiting results – during tight times, the leader of every business function is being pushed to improve their results dramatically. Unfortunately, because most recruiting functions don’t measure their “hiring failure rate.” Most of our recruiting leaders are not currently fully aware of the extent of the shortcomings of your hiring results. The weakness of your results can be illustrated with a single number. A mere 19% of hires “become an unequivocal success.” Fortunately, the effective use of AI may be able to raise that success right to well over 50%.
  • Use it as a driver for becoming a data-driven function – recruiting must join every other major business function that has become data-driven. In fact, HR and recruiting are literally the last high-impact business function that operates primarily on intuition and the continued use of long-established practices that are no longer effective. Fortunately, because AI/machine learning is, by definition, a 100% data-driven process, AI becomes the ideal driver for accelerating this long overdue transition into data-driven recruiting.
  • Use it to create a much more powerful hiring process for filling AI jobs throughout your business– in this continuing fierce “War for AI talent.” Perhaps the recruiting area that will cause the most business damage isn’t the need to hire AI talent for work within the recruiting function. Instead, it is the inability of your current recruiting process to successfully hire sufficient AI specialists outside of recruiting. So that your organization can successfully populate every major business function with the high volume of AI talent that each one will need (I highlighted this company wide AI shortfall back in 2018).
  • The key to every recruiter’s job security may be learning everything about AI – I predict that the upcoming wave of AI adoption in recruiting is literally unstoppable. And that this AI wave will dramatically change the work of everyone in recruiting. But those impacted the most are those that currently spend a lot of time screening resumes, sourcing both active and passive talent, and participating in many interviews. However, on the positive side. This AI wave will also expand the number of jobs in the recruiting function involved with data science, results metrics, AI/machine learning implementation, A/B testing, employee referrals, and delivering meta visuals.

—————–

Next, Visualize How Every Element Of Recruiting Will Be Improved By AI

Once you research the wide range of business processes that have already been successfully dominated by AI (i.e., supply chain, marketing, finance, production), you will also realize that, as most new approaches in recruiting do, the best are borrowed from the business side of the enterprise. So expect this business tool to creep into every individual element and step of recruiting. Below I have provided a snapshot list of the dozen-plus major elements/steps within recruiting that will eventually be dominated by AI-driven processes. The names of the top six most impactful recruiting elements that AI will change are underlined.

  • AI will improve employer branding – because your employer brand’s relative strength has a large impact on attracting quality applicants. Machine learning will be used to determine the best media and sites for placing branding and recruiting information for each job family. Machine learning will also sort through survey information that is gathered from a sample of potential applicants. In order to identify which employer brand factors (i.e., culture, product, flexibility) have the most attraction power in each job family. Improving the “convincing element” of employer branding will also be enhanced by the addition of multiple lifelike work scenarios from the meta-verse. Each of these will allow potential applicants to truly “feel and experience” what it’s like to work here as a member of our team.
  • Sourcing will be raised by several levels – rather than just relying on traditional sources for talent. Machine learning will sort through every failed and successful sourcing effort to determine the most effective sources for each job family. Surveys of new-hires will reveal additional sources that we never considered. There will also be a special sourcing focus on generating quality referrals. And on hiring diverse talent (Less than 2% of HR leaders surveyed confidently said that they were actually achieving their DE&I goals). And AI will also enable recruiting to conduct a continuous automated search throughout LinkedIn, all social media, and the Internet. This continuous global search capability will allow corporations to build and maintain a talent pipeline of desirable prospects. And this AI process will also be able to find “pre-job search indicators” on the Internet. So recruiting will have the capability of alerting recruiters just before one of your highly qualified pipeline targets is about to enter job search mode.
  • AI will allow you to sculpt job titles – most job titles are created without fully understanding the attraction value of the different possible titles. So an AI-driven process will be developed that can sort through the results from your A/B testing on the possible job title choices to determine which exact job title has the most attraction power for top candidates.
  • AI will allow you to use only the most predictive job requirements – unfortunately, most lists of job requirements are created without the benefit of data. However, machine learning will study every hiring success and failure for each job in the future to determine which specific hiring requirements do or do not accurately predict on-the-job success for each critical job.
  • Where and when you place job postings will be improved by AI – rather than just repeating current posting practices that are likely not optimal. Machine learning will sort through your past hiring data and your A/B tests to determine the ideal times and places to post your open jobs in each job family.
  • The critical step of resume sorting will have a much higher accuracy rate – currently, little effort is put into determining the accuracy of your sorting by recruiters and your ATS sorting algorithm. However, a sorting error on a quality candidate will mean that they will never be reconsidered. Over time, your machine learning algorithm will be able to identify each of the accurate and inaccurate “sorting out factors” (Especially those that may have discriminatory impacts). And to further improve resume sorting accuracy. Periodically your sorting system will be tested with good, questionable, and bad sample resumes. And that will further aid in continuously improving its sorting accuracy.
  • Utilizing AI will improve technical skill assessment– because interviews can have a predictive value no greater than a coin flip. Some essential technical and business skills should not be assessed within your interview process. Instead, interviews will more often be supplemented with individual skill tests in the areas of technical and soft skills. This testing will become more palatable to applicants because virtual reality tests are becoming much more lifelike and acceptable. Over time, machine learning algorithms will determine which supplemental tests do and do not accurately predict success on the job. And that will dramatically improve the accuracy of your overall assessment process.
  • Candidate interviews will be improved – interviews often produce inaccurate results because of human biases and a lack of interview structure. Fortunately, machine learning can provide the interviewer only with the appropriate questions (and their answers) that most accurately identify whether the applicant meets each of the job requirements. Interviewers will also be provided with a “score sheet” that only allows for the assessment of the specific requirements for this particular job. In addition, automating asynchronous preliminary interviews will allow you to get interview answers without taking up a great deal of recruiter and hiring manager time. These automated interviews will be automatically scored with the machine learning algorithm. With this automation, in many cases, only “live” interviews with a candidate will be provided to the finalists. And because most skill testing has been done outside of the interview. Half of the remaining live interview time can now be devoted to “convincing the candidate to say yes” if they are offered the job (which is essential in a tight labor market).
  • AI will limit the number of premature dropouts – as I noted earlier, 45 % of applicants have turned down a job offer primarily because of a bad interview experience. Unfortunately, currently, there is no formal process for determining at which step (and why) top candidates prematurely dropped out of your hiring process. The use of AI will allow recruiting leaders to identify and then fix each major dropout cause.
  • AI will improve the effectiveness of reference checking – currently, organizations collect little data on the predictive accuracy of their reference checking results. Fortunately, machine learning algorithms can reveal the predictive value of each type of reference check (former managers, personal references, social media, criminal, educational, and credentials. Note: credentials are usually the most accurate content area in resumes). Automated continuous reference-checking technology is already available. And that’s a positive thing because the hiring process can be dramatically sped up (and bad candidates can’t be dropped earlier, saving time) by initiating these initial automated checks early on in the hiring process.
  • AI will improve how you sell your finalist – few conduct a formal failure analysis to determine why their job offers are refused. However, a delayed survey (two months after the rejection) of your finalists will produce data that your machine learning algorithm can use to determine the most and the least impactful offer content factors for each job family. AI can also be used to personalize your job offers so that they more closely fit the expectations revealed by each individual finalist.
  • Machine learning-driven failure analysis will identify the root causes of each hiring failure – most recruiting processes literally have no way to formally assess the root causes of each hiring failure in one of your critical positions. However, if a failure analysis is conducted. The machine learning algorithm will be able to educate recruiters and hiring managers about the primary causes so that many future hiring failures can be avoided.
  • AI can help to determine the best recruiting metrics to use – because becoming a data-driven function generally means that the number of “results metrics” that are calculated will be significantly expanded. AI can provide advice on which recruiting metrics (e.g., the work performance of new hires, discrimination risks) garner the attention of executives. And which ones closely correlate with business success, so they provide the highest contribution to continuous improvement.

————————————

And Finally, AI Will Add Some Startling New Capabilities To Recruiting

Below you will find four areas where AI technologies will be able to provide completely new information. Which will help to improve your hiring and its impacts dramatically. They include:

  • AI can identify the best times to be hiring – by analyzing hiring success and failure patterns during different parts of the year. AI-driven processes will be able to alert hiring managers about upcoming times of the year when your talent competition will be exceptionally low (i.e., December and August). Shifting your hiring to those time periods will increase your chances of landing highly desirable talent.
  • AI can reveal if the odds are reasonable for landing a highly desirable candidate – many hiring managers “take a chance” on hiring a top candidate that multiple major talent competitors are actively recruiting. And that “stretch try” often results in an unnecessary waste of recruiting time and resources. However, AI can use historical data covering the winning and losing of highly desirable candidates. To calculate in advance your organization’s odds of actually hiring an individual “high demand candidate”. This information provides recruiters and hiring managers with the opportunity to “pass on” highly desirable candidates where you realistically have no actual chance of landing them.
  • AI will help you identify your most effective recruiters – machine learning can analyze your data covering all of your top candidate hiring successes and failures. And from the data, it can alert recruiting leaders about which one of your recruiters (or hiring managers) is the most likely to be successful in filling this job. Where highly desirable candidates are likely to be “bid on” by multiple recruiting competitors. This process can also identify the common recruiter skills and capabilities that your newly hired recruiters should possess.
  • AI will reveal a new hire’s trajectory – not only can machine learning determine which hires are most likely to be successful in each job. If you continually track the job movements, the promotion rates, and the retention rates of all of your new hires. Machine learning will also be able to predict the new hire’s “career trajectory” over the next few years with your company. Where their career trajectory is the future growth in capabilities, their internal movement speed, and their years of retention.
If you only do one thing – realize that you are not the only recruiting function implementing AI-driven processes. So proactively put together a process that continuously tracks the AI implementations at recruiting powerhouse firms like Amazon. And then, use that benchmark information to guide your own implementation direction and speed.

Final Thoughts

Unfortunately, despite the many benefits a company accrues from adopting AI processes, you need to expect some significant resistance. The recruiting function has traditionally been one of the slowest to change. To begin with, you must learn to expect a great deal of resistance from HR and recruiting staff to any AI implementation proposal covering any of the different recruiting areas. You should also expect many hiring managers to resist its implementation, even though it will dramatically improve their hiring results. So as part of your AI implementation selling approach. I recommend that you seek out the active support of several “executive champions” who have successfully used AI in their business processes. You should also survey all likely resistors in order to identify and counter each of their concerns. And finally, you should also widely spread your percentage improvement data from your initial AI implementations to further dissuade any of the remaining AI skeptics.

Author’s Note

  • Please share these solutions by sending this article to your team/network or posting it on your favorite media. 
  • Next, if you don’t already subscribe to Dr. Sullivan’s weekly Aggressive Talent Newsletter, you can do that here.
  • Also, join the well over 12,000 that have followed or connected with Dr. Sullivan’s community on LinkedIn.

About Dr John Sullivan

Dr John Sullivan is an internationally known HR thought-leader from the Silicon Valley who specializes in providing bold and high business impact; strategic Talent Management solutions to large corporations.

Check Also

Bad HR Can Cost You Billions, Just Ask Apple (How ignored HR process vulnerabilities can lead to business catastrophes)

Retention and recruiting were factors in the downfall of Apple’s $10 billion EV car project! …