Jobseekers are firing off hundred of applications. Few even get a response, let alone an interview. Inevitably many jobs turn out to be fake and the applicant’s data is shared with scammers and spammers.

Employers meanwhile are inundated with applications. Many are from bots or people in entirely different countries. Wading through this to find real, qualified, motivated applicants has become slow and arduous.

In short the market is failing in its primary task of connecting real people to real jobs. Both sides have a lack of trust and both sides are drowning in noise.

How did we get here?

To understand what is happening we need to look back a few years and understand how the economics work.

Step 1: One-click apply

In 2004 I applied for my first job, a Saturday shelf stacker at the local Co-Op. I completed a paper application form dropped it off in store in person and was given a short interview by the manager straight away.

By the time I was applying for banking internships in 2008 applications were thankfully online. Each one required a CV, Cover Letter and responses to several written questions. The first one took best part of a day and after that I found I complete each additional one in a couple of hours.

Applying for jobs took time and as a result was an indication of commitment.

Fast forward a few years and two jobs boards were dominating the UK market - Linkedin and Indeed. Both introduced their own application features and both allowed businesses to post jobs for free but charged them to promote their listings.

The metric for success was the number of applicants. Initially this makes sense - if each applicant has a 1/10 chance of being a good fit, 10 applicants means a 65% probability of success and 50 applicants means a 99.5% chance. The jobs boards therefore worked hard to remove any friction: applications felt more like an onboarding flow, designed to maximise conversion and minimise clicks. 1 click apply was born.

Step 2: Volumes rise, quality falls

Unsurprisingly if you make it easier to apply for jobs people will apply to more jobs and give less thought to each one.

The result has been an explosion in the number of applications. Users on Linkedin and Indeed could submit dozens of applications in a few minutes, and they did. Even if the applicant didn’t fit all the requirements or wasn’t in the right area or was only half-heartedly looking why not apply with a single click and see what happens?

This posed a problem for employers: how could they sift through hundreds of applicants to find the ones that actually were actually qualified and serious? Doing so manually involved hours of laborious, repetitive work.

Step 3: Enter AI

Laborious, repetitive work is what AI likes best. It is no surprise therefore that HR departments, inundated with hundreds of applications, turned to AI tools to do the heavy lifting.

AI, often integrated in Applicant Tracking Systems (ATS), reviewed applications and matched them against the requirements (increasingly also written with the help of AI).

Applicants started noticing this and responding accordingly. If your application is going to be read not by a human but by AI, then it changes how you approach the task.

At the end of my CV I used to mention my interests, including running ultramarathons. More than once this led to lengthy discussions with similarly inclined interviewers to the point where I’m pretty sure it helped me get the job. With an AI reviewing I would have been better off adding a string of keywords.

Who best to advise applicants on how to tailor their applications for AI review? You guessed it, AI. With the release of ChatGPT in 2023 jobseekers started asking AI to edit or even write their CVs, Cover Letters and other applications for them. Soon there were a whole range of free and paid tools for tailoring applications to specific jobs.

Step 4: Zero-click apply

Once AI could draft my application it was a small step to ask it to find the jobs and hit submit as well. Enter zero-click apply. In the last month one such tool has applied for hundreds of jobs for me without me lifting a finger (well, actually it’s applied for John Smith).

Step 5: Things fall apart

Applications per job have unsurprisingly gone through the roof. The chances of getting an interview has got even smaller. Any jobseeker not using AI is probably only able to submit a few applications each day which will likely get drowned out in the noise.

And companies? If their metric for success is the number of applicants per job then happy days. Unfortunately however that 1/10 chance of an applicant being a good fit isn’t fixed. Many jobseekers barely know what they are applying for. A 1/1000 chance now seems more probable. Find that 1, even with AI, is a long and painful process.

No one can afford not to use AI but the net impact of AI on both sides of the market has been to make the experience worse for everyone.

What is the solution?

The first step is to rebuild trust

Jobseekers need to be able to know that jobs are real and it is worth their time to submit a real application

Employers need to be able to identify who the real, motivated and qualified applicants are so they can invest time and energy in assessing them.

We are building a new job board where all companies and all applicants must be verified. This is in everyone’s interest (except for the spammers and scammers of course).

Quality not quantity

Job hunting is hard and can be demoralising. Submitting an application feels productive. But there is a danger that jobseekers fall into the same trap as employers of thinking that volume of applications is the benchmark for success.

Fewer serious job applications rather than hundreds of half-hearted ones would be good for both job seekers and employers.

AI can play a valuable role but it should be a co-pilot not an agent

There are aspects of job hunting where AI can play a valuable role - searching jobs boards for suitable roles, preparing documents for submission, tracking application, reminding people to send follow-up emails, preparing for interviews. But this should be an assistant to rather than a replacement for the applicant.

AI has broken the jobs market. Rodeo is trying to fix it.

Get in touch to tell us more about your experience and give us feedback or ideas about how we can help.