How DirectlyApply is using AI to elicit and understand feedback to enhance the job seeker experience.

TL;DR

In today's rapidly evolving job search market, the traditional job board is undergoing a transformation. We believe that the future points towards an interactive experience where job seekers feel like they are engaging directly with an AI recruiter. The shift towards a highly personalized, efficient, and responsive AI is already seen across platforms such as TikTock, Netflix and YouTube and DirectlyApply is building job search assistance that is similarly able to adapt and prioritize the unique needs and preferences of every job seeking individual.

Introduction

Since 2019, DirectlyApply has been at the forefront of empowering job seekers by providing extensive data to help them make well-informed career decisions. With the introduction of Threads and our innovative intelligent feedback loops, we’ve taken a step further. We're not just presenting information for job seekers to consume; we're now actively engaging with their inputs, understanding their needs and sustaining ongoing dialogues. This enhanced interaction is fostering significantly better outcomes, as it allows us to tailor our guidance and support to each individual’s specific career aspirations.

Encourage, Elicit and Understand Feedback with AI

With Threads , job seekers on DirectlyApply are already benefiting from an exceptionally personalized experience, as search results are intelligently organized according to their priorities. However, whilst we've leveraged implied signals in the past—such as the time spent reviewing a job or whether the job seeker applied—we've lacked insight into the underlying reasons behind ‘negative or non-actions’.

To bridge this gap and gain a deeper understanding of the job seeker's journey, we've recently developed an AI enhanced experience. This innovative approach allows us to capture and analyze feedback in real-time, enabling us to utilize these insights to enhance outcomes for job seekers.

Understanding the job seeker journey

The three indicators that enable us to best understand a job seekers needs are as follows:

1. Job seeker views job, started application process, completes application

2. Job seeker views job, starts application process, but does not complete the application

3. Job seeker views job, but does not apply

As stated above, Threads was built to take the learnings derived from the completed application (point 1) so we now needed to understand why a job seeker made a negative or non-action in relation to the apply (points 2 & 3).

Powered by AI

We therefore wanted to create a feedback loop that could contextualize the job seekers response and provide a meaningful response.  Obviously, a reason for not applying to a job can be straightforward, e.g. ‘I do not have a CDL license’, but often the reason is more nuanced, for example, ‘I have 20 years of experience in HR but I don't have a degree which I believe is a requirement for this role’.  The challenge we set at DirectlyApply was to solve for the latter.

By analyzing thousands of hours of job seeker interactions, we were then able to create custom vector embeddings of feedback received from real job seekers and then run comparison functions in real time as the job seeker types in their feedback to return an appropriate feedback handler. This approach means we have been able to combine common reasoning alongside free form feedback to create an incredibly powerful feedback tool.

The result?

DirectlyApply is now able to, in real time, elicit and understand job seeker feedback and push this knowledge back into their job search.  Enhancing their job Threads and delivering employers with qualified applications which they would not have otherwise received.