AI Revolutionizing Recruitment: The Future of Hiring
Reading Time:
Shortlisting qualified candidates for a role is always time-consuming. Artificial intelligence promises to match candidates to a role faster and more efficiently than humans.
If you don't have the luxury of recruiting for a niche role that only attracts a few highly qualified candidates, creating a shortlist can be a time-consuming task. HR managers at companies running large-scale hiring campaigns—such as for graduates or a new contact center—face a mountain of potentially thousands of applicants. Not surprisingly, they turn to recruitment management technology to sort out the good candidates from the unsuitable ones.
Applicant matching has advanced significantly in recent years, becoming a prevalent application of AI in HR and recruiting. Numerous software companies now offer tools for every stage of the hiring process, from technologies that align job postings with potential candidates to applications that scan and analyze resumes. These tools can detect keywords in resumes or applications, score candidates against specific job criteria, and filter out those lacking necessary qualifications. They claim to accomplish these tasks within minutes, a stark contrast to the days it might take to manually review applications. Hiring software has thus revolutionized the efficiency and effectiveness of recruitment efforts.
The providers of such systems claim they can significantly reduce the workload of hiring managers, speed up time to hire, and save companies money because technology can automate much of the selection process. "While human judgment will likely remain more reliable, there are parts of the hiring process that can be automated. For hiring managers, the biggest bottleneck is wading through endless unsuitable resumes," says Marja Verbon, co-founder of careers site Jump. Pre-screening involves screening candidates against a list of criteria before they apply. Only an estimated 12% meet these criteria, meaning the time spent screening can be reduced by as much as 88%.
Sounds too good to be true? But AI's growing role in matching candidates with job openings has its limitations. Fundamentally, most algorithms search resumes based on keywords, so hiring managers may miss the bigger picture of what a candidate has to offer. "Matching candidates has gotten a lot better," explains Neil Armstrong, Commercial Director at Tribepad, an onboarding and ATS hiring software company. "But a resume isn't necessarily a good way to understand a person's skills. You get an overview of a person's experience, but not insight into their personality or potential. And people have learned to game the system by entering certain keywords that an AI search picks up. They get caught out later in the process because they don't have the skills, and it's a waste of time for them and the hiring manager."
One of the main concerns around using AI in candidate selection is the potential to introduce or reinforce bias. On the one hand, unconscious human bias can be reduced by automating the selection process, but on the other hand, the algorithms that underpin this automation can themselves be problematic. Often, employers use historical data sets to evaluate applicants, which can mean they disadvantage underrepresented groups such as women or black, Asian and ethnic minorities. Additionally, certain keywords can target certain groups. For example, asking the algorithm to identify applicants who play a certain sport or exhibit certain behaviors could inadvertently exclude a large group of suitable applicants. Outside the field of recruiting, biased algorithms used by U.S. courts have been shown to falsely classify black defendants as twice as likely to commit crimes as their white counterparts.
Kim Nilsson, co-founder of Pivigo, a data scientist recruitment specialist, explains: “One would hope that as [algorithms] become more sophisticated, they would remove bias rather than amplify it, but the slightly scarier question is: how big is the risk of bias in the process as algorithms become more integrated or widespread?” That’s because algorithms learn from “training data” based on past successes, she adds. “In HR, for example, a training dataset would be a set of resumes from previous [candidate] applications, with labels of those who received a successful job offer. The problem with this is that if fed this kind of data, the algorithm might say that someone with a different profile to that of the current workforce (e.g. gender, nationality or educational background) should be rejected because the previous ‘success cases’ don’t have that diversity. You will perpetuate the biases that already exist in your workforce and in your data.”
“Greater diversity in algorithm development teams could help mitigate unwanted bias in system design,” says Nimmi Patel, policy manager for skills, talent and diversity at UK industry body techUK. “With diverse teams – in terms of gender, ethnicity, experience and background – the likelihood of unconscious bias being identified and addressed, rather than ingrained into future algorithmic decision-making systems increases. This, in turn, will improve the quality of decisions made.
Gareth Jones, CEO of Headstart, an AI system for hiring graduates and entry-level professionals, agrees that the data going into the system needs to improve if AI is to deliver on its promises. “AI offers the opportunity to make hiring fairer, but we are still a long way from hiring being data-driven,” he says. “Recruiters still look at a CV or a profile. You can supplement this with an assessment, but they are not based on what the success criteria are at that company. That should be the starting point - not a job description.
However, there are ways to make the most of AI by combining it with other tools to more accurately predict good hiring outcomes. Introducing assessments into the process can be a means of getting a more comprehensive and objective picture of the candidate. "These can be integrated into a company's ATS at the beginning of the hiring process, after the initial application," explains Chris Platts, CEO and co-founder of ThriveMap, a company that conducts pre-hire assessments. ThriveMap's assessments take candidates through a digital "day in the life" of a job. "Candidates are automatically invited by email or text to complete their 'virtual shift' and the scores and interview reports are pushed directly into the ATS," he adds. "This hands-on approach to talent assessment means candidates can get a feel for the role and culture before they join the company; they may even opt out of the hiring process if they feel it's not the right fit for them."
Selection of candidates from the process is actually a positive outcome for hiring managers. It ensures that those who move on to the next stage identify with the company culture and are more likely to be productive and engaged if they get the job. The data and feedback that testing provides during the hiring process can also help to clarify to eager but unsuccessful candidates why they didn't meet all the criteria and how they can improve their skills when applying for another role - or even suggest a better-suited position within the same company. "Context-specific testing can show that the role is real. So when you're applying for a residency position, you should address that request rather than asking for general skills like Excel," adds Adrian McDonagh, co-founder of hireful, a company that helps recruiters improve their hiring practices. "Then you can point to the assessment and show why you made the decision, rather than just saying, 'We found a better candidate.'
More comprehensive assessments can also show hiring managers that candidates have a diversity of values and thinking styles. A 2013 report from Deloitte recommends keeping “diversity of thinking” in mind when hiring employees to avoid groupthink and encourage new ways of solving problems. One way to do this is through game-based hiring tasks or challenges, which can show not only whether candidates have job-related skills, but also how they make decisions—an important aspect of building a diverse team. The UK intelligence agency GCHQ has run several campaigns aimed at recruiting employees with cyber skills, asking candidates to crack codes or decode messages to meet the requirement to “think like a hacker.”
Employers such as Lloyds Banking Group and Police Now have now used virtual reality challenges for graduate recruitment, where candidates complete simulations and hiring managers can see how they do. Some game-based recruitment competitions even go so far as to offer a prize: Google's Code Jam, where programmers compete against each other to show off their coding skills, offers a $15,000 reward for the winner (and a solid shortlist of candidates with proven coding skills for Google).
Knowing more about how the candidate operates in a realistic environment and whether they'll fit well into the team will make the onboarding and familiarisation process much smoother, Platts adds. "It's difficult to find candidates in your applicant pool who match your company's desired behaviours, who can demonstrate the required role skills and who are genuinely committed to the role," he says. "Making better hiring decisions means candidates become productive faster. When the assessments guide candidates through a digital experience of the role, they are less likely to be surprised by the job requirements when they start and give up early.” By using algorithms effectively, companies can make their recruitment marketing more targeted.
Verbon adds: “On the professional side, job recommendations have been a big no-no until now. We have all received a completely irrelevant email with a job advert. Algorithms can help by identifying which roles would be a good fit based on a professional’s resume, rather than just random keywords. These recommendations are much more precise and can make job hunting a far less daunting task.
Ultimately, however, AI and other automation tools should be used to support and complement the work of the human recruitment team, not to replace it. “Used correctly, these systems can do wonders for improving diversity in a company by encouraging applications from people who would not otherwise make the shortlist,” says Nilsson. "HR managers often have to sift through hundreds of applications and, for obvious reasons, don't have time to review each application in detail. They can also help to compensate for the shortcomings of mechanisms such as recommendations. "While recommendations can be a good indication that the person in question could be a good choice, they miss outstanding talent who are not as well connected. So algorithms can support the decision by highlighting applicants who would otherwise be hidden in the noise without making the final decision," she adds.
A start-up called BrightHire, backed by renowned organizational psychologist Adam Grant, is offering to fill that gap. Grant, a professor at the Wharton School of Business, explored in his book Originals: How Non-Conformists Move the World why in the workplace, it's not always the conventional qualities or actions that make people good at their jobs. BrightHire's tool doesn't simply automate decisions about who gets shortlisted, but provides context to interviewers during and after video interviews with candidates. The software offers an "interview assistant" that keeps interviewers on track by showing them the predetermined questions to ask each candidate. Because interviewers ask a set of standard questions, they are less likely to make subjective assessments based on their own biases, while BrightHire's prompts make the process more efficient. Pre-employment assessments are also becoming more sophisticated, including neuroscience-based tools that can predict whether a candidate is likely to be a good fit with company culture and skills.
With an overwhelming number of applicants, the promise of speeding up the matching process is tempting - whether at the beginning of the candidate search or when inviting candidates to an interview. A dizzying number of technologies are available to help HR and hiring teams source, match and pre-screen candidates - and while this can't replace the human touch, it can make the process more evidence-based, efficient and cost-effective.
Candidate matching: five key takeaways
- Today, hiring managers can choose from a wide range of tools that can help match and select candidates
- AI and algorithms can save hiring managers time and money, but should complement rather than replace human decisions
- Be aware that algorithms can introduce bias when hiring candidates.
- Technology can paint a more reliable picture of a candidate's suitability, making them more likely to be worth hiring
- Determining a candidate's suitability for the day-to-day aspects of a job during the hiring phase can increase retention and productivity.
AI in recruitment offers significant advantages, from reducing manual workload to improving candidate matching accuracy. While human judgment remains vital, AI tools can complement HR efforts effectively. IceHrm, an innovative HR management system, exemplifies how integrating AI can lead to smarter, more efficient hiring processes, ultimately enhancing workforce quality and diversity.
