Nobody likes applying for a new job. As much as you may despise your current role, the process of tidying up your CV and crafting an eloquent cover letter that simultaneously portrays you as an independent thinker and the world’s best team player is tedious work. At least it’s not too much of a stretch to put yourself in the mind of a recruiter, visualise the kind of traits that they’ll be looking for and structure your application accordingly.
Well, that may have been the case until fairly recently, but today your application may be rejected before anyone even sees your credentials. Increasingly, businesses are using computer algorithms to separate the wheat from the chaff. This is partly out of sheer necessity: some firms receive hundreds of thousands of applications for any single vacancy, which means going through them all manually simply isn’t practical.
Just as algorithms are influencing the items we choose to buy, the programmes we watch and the people we date, they are also affecting our career prospects
But while using algorithms helps to streamline the process, it introduces challenges elsewhere. Firstly, businesses need to think carefully about the specific algorithm they are using and the hidden biases they may be programming into it – they are still written by humans, after all. More importantly for jobseekers, this software-based form of recruitment may mean yet again rejigging their applications if they want to make sure they’ve got a chance of making it to the interview stage.
The age of the algorithm
The job market is becoming a more competitive place than ever. Back in 2012, a study by job search service the Ladders found that recruiters spend just six seconds looking at a CV. Then, in 2015, employment review site Glassdoor reported that each job posting would receive an average of 250 applications. For investment bank Goldman Sachs, the number was far higher: the company reportedly received 250,000 applications for a single vacancy in 2016. Candidates have become more qualified too, with the proportion of individuals in the US holding a bachelor’s degree or higher increasing from 29 to 37 percent between 2000 and 2018. Introducing some form of automation was inevitable.
“As with most automation coming to the business world, the focus is on removing excess administrative work and providing analytics and modelling to make better-informed decisions,” Rachel Roumeliotis, Vice President of Content Strategy at O’Reilly, explained. “Analytics and modelling are playing a key role in tracking vendors, employee attendance and workforce efficacy.”
Recruitment software, in many different forms, is already being employed by businesses. Artificial intelligence (AI) chatbots, like those created by California-based firm Mya, engage with passive and active candidates through dynamic conversations, gathering insights and building trust. The company reports that its software has led to a 79 percent reduction in time-to-interview and a 144 percent increase in productivity for recruiters.
Just as algorithms are influencing the items we choose to buy, the programmes we watch and the people we date, they are also affecting our career prospects. Applicant tracking systems (ATS), dubbed ‘resume robots’, can instantly scan CVs for keywords and essential skills, eliminating candidates so that human recruiters don’t have to. This is changing the role of HR teams significantly.
“As HR adopts technology to help it streamline its processes and get more time back for deeper analysis, this needs to be reflected in the skill sets of HR and recruitment teams,” Tom Ricks, Senior Director of People Analytics at software company Qlik, told The New Economy. “Larger recruitment teams may now even have analytics teams of their own to help them filter candidates and… analyse data on later performance in order to assess the effectiveness of the solution and tweak the algorithms.”
Mya’s recruitment software in numbers:
Reduction in time-to-interview
Increase in productivity for recruiters
HR teams are increasingly looking towards automated solutions when it comes to traditional routine processes, from employee onboarding to dealing with employee benefits queries and undertaking general admin tasks. Jobscan estimates that 98 percent of Fortune 500 companies now use ATS software to assist with the hiring process.
Cutting down the numbers
While each piece of recruitment software will operate in distinct ways, they broadly work by ruling out a large swathe of candidates without any human input. ATS, for example, can match an individual’s CV against the listed job requirements, giving the candidate a score for each one. Recruiters can then decide to only manually review applicants that score over, say, 80 percent, saving them huge amounts of time.
“If these solutions are created with recruiter input, they create ease in the recruitment process and free up time,” Katrina Collier, author of The Robot-Proof Recruiter, explained in an interview with The New Economy. “Ideally, this gives recruiters and talent acquisition more time to create better engagement with jobseekers and a better experience for managers who will be interviewing more suitable applicants. My recommendation to HR and recruiters is to always ask their peers for their technology recommendations based on the benefits delivered and not purchase [the software] solely because it looks shiny and the sales consultant says it will deliver.”
As Collier says, not all recruitment tools will live up to the hype. In fact, some of them come with unexpected downsides. Bias has long been a major talking point within the recruitment sector; believing that AI will help eliminate this is naive. Algorithms simply reflect the biases of their creators and the existing data that they have access to. In 2018, Amazon was forced to abandon an AI recruitment tool it was using after discovering that it was discriminating against female candidates.
Computer says no
If computer software is largely making life easier for recruiters, it is also posing a new challenge for applicants. ATS solutions often scan applications for keywords, so it’s important that candidates mirror the vocabulary used in the job description. Using a fancier-sounding synonym might impress a human recruiter, but may be dismissed by an algorithm. Similarly, previous job roles should be given standardised titles. Some firms use off-beat names like ‘digital overlord’ instead of ‘website manager’, but these are meaningless to computer programs that are not expecting them. This doesn’t mean applications have to be completely devoid of personality, but making sure the right boxes are ticked first will help candidates pass the robot recruitment stage.
“Ultimately, automation won’t replace the significant role of human decision-making in HR,” Roumeliotis said. “What may make it more tricky is how candidates find their way to HR’s attention. Automated recruitment matching compares CVs to a job description to find out if there is an initial fit. Only then will the CV make its way to the finalist pile. One suggestion to counteract this is to customise your CV to incorporate keywords and phrases from the job description. You’ll still need to actually do those tasks to be a success, so be sure to only do this when it aligns with your skills.”
Collier believes that regardless of developments in the recruitment sector, many things remain the same. “New technology or not, proactive jobseekers always fare better,” she advised. “By proactive, I mean those who apply for roles they are suited to and who find a way to connect with in-house recruitment teams.”
Rejection is never easy to take, despite being an inevitable part of the job application process. Candidates looking to find a new role should remember that, usually, it’s nothing personal – in fact, there might not be a person involved at all.