One of my customers had a problem. Their graduate recruitment process wasn’t delivering enough candidates to fill the vacancies they had available.
They knew the standards candidates were required to meet. They knew the number of vacancies they needed to fill. But not enough of the right people were coming through the process. They were concerned that the sifting tools they were using were filtering out too many people.
So the question they asked me was should they lower the benchmark on the sifting tool so that more candidates reach interview stage? Wouldn’t it be better to assess more at interview to ensure more people were recruited?
But there are significant risks associated with taking this step. Here’s why you should be cautious about lowering your benchmark.
1. You may fall foul of the law regarding fairness and discrimination
First off, check that you can, legally, change the benchmark during a recruitment campaign.
For my customer, theirs was a rolling programme of recruitment which aimed to fill vacancies as and when they came about in the business. This gave them more flexibility to amend scoring guides than if they’d been undertaking a single recruitment activity.
If you’re doing the latter and change the benchmark during a campaign, a candidate could claim that they’ve been unfairly discriminated against.
2. Candidates may simply fail later on in the process
My client was using a numerical test as a sifting tool, having already ascertained that numeracy was an important skill for the role.
If they were to lower the pass mark on this test, how would they ensure candidates had the necessary numerical capability before hiring them?
You can’t test numerical ability in an interview. So perhaps you’d have candidates complete a numerical exercise after the interview.
But the risk is that candidates will simply fail that test, just as they would have failed the sifting test. Meaning you’ve interviewed lots of people that don’t pass muster anyway.
3. You will spend more time interviewing people who are not suitable for the role
By lowering the benchmark, you’re effectively widening the floodgates, which means you will get more people to interview. But some of them won’t be suitable for the role.
So you’re likely to spend time and resources interviewing inappropriate candidates. If you’re using external interviewers, this will cost you money too.
4. Interviewers may become frustrated and disengaged and drop their standards
By dropping the benchmark more people will get through to interview stage. This means interviewers will spend time interviewing people who are not at the standard they need to be to be hired.
This can be a frustrating and disheartening experience for interviewers. They may lose faith in the process and in the standards you demand for your organisation.
As a result, they may drop their standards too, putting more people through that don’t meet your criteria for the sake of ‘bums on seats’. Or, they might go to the other extreme and become too harsh. They start to doubt everyone and put no one through.
5. You might recruit people who need more training, more support and more time from you when in the job
What happens if you recruit people who do not have the skills and capabilities you need to fulfil the role?
Clearly, they’re likely to need more support, more induction, more training in order to get up to speed.
And there are some skills and abilities – like numeracy – that are really difficult to train into someone if they don’t already have an innate competence in that area.
Getting more bums on seats is all very well, but are you prepared for the extra costs and extra time that will be required as a result?
Lowering benchmark scores seems logical if you’re not getting enough people through to interview, but it’s a risky business.
Ultimately, by improving the quality of candidates at the face-to-face interviewing stage, you will save time, money and stress. And ensure you have the evidence you need to make better hiring decisions.
And the only way to do that is through high quality sifting at the front end of the process.
And here’s a useful tip when it comes to making sure your sifting tool is set at the right level in the first place: Most sifting tools are designed to deselect between 30 and 40% of applicants based on a normal distribution curve. If your tool is sifting out more than that, it may be set too high. And if it’s sifting out less, it may be too low.