The automation of job searching and candidate sourcing

David Leithead 19.11.2018

From the perspectives of both job seeker and recruiter, Artificial Intelligence has the potential to make processes much easier.

Current tools are outdated and inefficient, to be honest...

The recruitment industry has used technology in the form of applicant tracking systems (ATS) to aid processes for some time. ATS tools increase efficiency, accuracy, bring control, and governance, but also organise the recruiter into following winning sales and marketing routines and workflows.

Yet despite heavy investment by recruiters in these tools, the current methods of candidate matching by ATS can be inaccurate and the systems difficult to use, for example relying heavily on exact keyword matching in order to form a suitable pool of applicants. Most ATS tools can’t contextualise and they don’t learn as they go. Due to the sheer diversity of skills and backgrounds of the workforce, and the variety of ways in which the workforce chooses to express their skills and backgrounds, often multiple searches have to be carried out of the same database in order to reveal appropriate and relevant results.

Benefiting the recruiters and job seekers

AI tools, of course, are marketed as being able to solve some of these challenges. But predominantly, only large recruitment agencies with a sufficient budget have already integrated Artificial Intelligence into their daily processes. So in general terms, the recruitment industry is at the scoping stage - aware of the tools but wondering how to apply them. AI has advanced at an impressive rate over the past few years and perspectives have firmly shifted from what might be possible in the future? to how can we do this now? Already, some employers have adopted tools that schedule interviews, screen applicants and even carry out background checks.

There is also the opportunity to improve the overall candidate experience - a happy applicant is beneficial in many ways. A lack of communication and transparency when it comes to status of applications is a common bugbear that can be easily overcome by introducing good tech tools; maybe interactive AI tools that seem like humans, such as simple chat functions on a website, could be the solution.

The 2nd article in our Artificial Intelligence series

Streamlined recruitment automation

In terms of looking for work and hiring employees, Artificial Intelligence is already making job searching for jobseekers and candidate matching for employers more automated processes. When an individual is uploading their CV, machine learning tools can instantly draw in relevant jobs based on the contents of the CV - beta versions of these matching tools are already on some of the key job boards. Likewise, when a job is uploaded, the tool suggests candidates that have skills matched to those in the job description based on the analysis of CVs, but also based on analysis of what’s happened to those CVs that have already been trawled and matched to similar jobs. A search can be left running in the background without being monitored and any new CVs added will automatically match with roles, reducing much of the manual labour. Ultimately, this means that a shortlist will be delivered of the most suitable individuals before there’s been any human intervention and interaction.

The machine will continue to learn. Input a full job description to get a list of relevant candidates returned instantly. Use a tool to re-write the job description and systematically improve it through machine learning to find ever better candidates, suggesting what to include in the job (err...sorry, in the job description) based on previously analysed and successful descriptions.

And then, here’s a thought, let’s develop a tool that helps write CVs and learns how to write them better, and better, to be best found by the matching tools.

Battle of the machines! Now that’s ridiculous...No, in fact, Microsoft and LinkedIn have developed one already. In this scenario, patterns and words will be recognised as ‘successful’ for a certain type of CV, and a similar version that is moulded to the individual will be produced.

(Machine enhanced) Candidate, meet (machine enhanced) job. This could lead to some serious disappointment on both sides when the person actually starts the job!

The potential with machine learning is infinite

Ok so where this is all headed creates a lot of questions. But back to today, the benefit of machine learning is that the tools are constantly evolving, using vast volumes of data to improve their processes, for example gaining a better understanding of every single applicant that enters the system. For every piece of data the tools encounter, they are continually learning - that’s quite exciting.  

There’s absolutely no doubting the value these tools bring, and the adoption and effective implementation of innovation is going to be the key to the success of recruitment agencies over the coming years. Investing in the data-driven approach should now be a priority.

Get in touch

If you have any questions or opinions about AI in recruitment, click here and fill out the form to get in touch today.

Other Artificial Intelligence in recruitment articles

Keep an eye out for our other articles in the series...

David Leithead's picture
Chief Operations Officer UK
dleithead@morganmckinley.co.uk