John Lewis have announced that they will publish job interview questions online so that candidates will have time to prepare answers in advance. It’s not clear how extensive this will be, but do helpful processes recruit the right candidates? John Lewis seem to think so. This well-publicised change, coinciding with the publication of government guidance on AI recruitment last month prompts a conversation on recruitment processes in the future:
- Do your current practices prove a successful recruiting tool?
- Do they give all applicants the chance to demonstrate their skills and experience?
- Could AI help in that process?
- What are the risks that recruitment by AI raises?
Are your current practices successful? Could they be more inclusive?
This is not an exact science; however, a good indicator of the success of recruitment is your retention rate. If your turnover rate is low and candidates are remaining in post for two years or more then your recruitment process is probably working well. That said, it might be effective at recruiting those people who can handle your current processes - whether that be completing an online application form and/or performing well at an interview. It might be missing a range of other candidates who would fit the vacant role well but don’t, for whatever reason, perform well according to your recruitment processes.
Publishing questions prior to an interview is helpful for neurodivergent employees or those with specific learning styles; it enables them to prepare more readily, and so prospective employees can see the skills and strengths they have, unhindered by nervousness. Disabled candidates can request reasonable adjustments to application processes, however, is it time employers took the lead and made the process more inclusive as John Lewis are trying to do?
Can AI help in the recruitment process and if so at what risk?
Depending on your perspective, the role of AI within our workplaces can either be alarming or exciting. In response last month, the Government published its responsible AI in recruitment guide. There are countless ways that AI can be used within recruitment; automatic interview transcripts, CV assessment software, and targeted job advertisement to name a few. Its usage may be valuable and may considerably reduce the time that many managers and HR teams currently must spend on recruitment. However, as this guide notes, and we all know by now, AI is not perfect and comes with risks. The most publicised being the discrimination risks. Transcription may be less accurate for non-native speakers or those with regional accents. Automated CV assessments may reinforce requirements leading to certain groups being underrepresented in the workforce.
If you are seeking to use AI with recruitment processes or are already doing so, we would advise you to read the guide and note the following:
- Undertake an impact assessment and produce an AI governance framework when procuring an AI system. This will help to ensure that the risks and opportunities of using an AI system are explored at an early stage.
- Ensure regular performance testing to identify any drops in fairness including unforeseen discrimination that may arise from updates.
- Implement training for employees using the AI system.
- Learn all you can about the system and its processes; knowledge and a willingness to engage are the most effective options when dealing with any new technology and understanding its impact.
If you would like any further information on recruitment whether that be policies and procedures or how to make it more inclusive, contact any member of the employment and pensions team.