For some time, we’ve been surveying job advertisements and asking why they all seem to follow a similar format. It suggests that many recruiters are simply copying the structure and wording they have seen elsewhere. But how can they be sure the industry standard approach produces optimum results?
Curo Talent decided to conduct a survey and discovered that 72% of Recruiters never test their job advertisements (see chart below). If you never test Advert A against Advert B, how can you be sure which text produces the best response – and how can you be confident of improving results.
Why is this important? Because today’s job market is candidate driven. Skills, especially IT skills, are in demand but short supply. Your marketing department will tell you that each tweak of your advertisement can improve response by 1-2%. That may not sound much, but if you regularly test and improve by 2% each month, by the end of the year you will be getting 27% more job applications. That’s worth having.
But testing does not just increase volume, you can also test to improve quality. Using words to filter out the skills or experience you don’t need in favour of the critical skills that are important.
A fatal mistake when testing Advert A against B is to assume that whichever one gets the highest response is the winner. Wrong. The results have to be statistically valid. To give a practical example, if both Advert A & B are posted on a website with 20,000 visitors per week, can you be confident of the results if Advert A get 2 applicants and Advert B gets 1 applicant?
The good news is that confidence can be measured, the bad news is that it uses a complex formula and requires a large number of applications. All of which takes time and know-how. We believe this is why so few job advertisements are tested. It’s quicker and more convenient to just guess what will work.
This is perhaps why recruitment adverts tend to look a bit odd when compared to other forms of advertising. For example, it would seem odd if Carphone Warehouse used the following text at the beginning of an advertisement:
Our mobile phone supplier is seeking customers to purchase their latest smartphone with face-recognition security, and…
Yet how many recruitment agency adverts have you seen using that formula? Starting the advert with ‘we’ or ‘our’, like this:
Our enterprise client is seeking an IT Manager with security skills, and…
Good advertisements start with ‘You’. It’s not about what you want to sell, it’s about what they want to buy. You are trying to persuade them to ‘buy’ your vacancy.
There is also a lack of ‘power’ words in a lot of job adverts; such as You, New, Now, Who, What, When, Where and How, as well as Get, Opportunity, Learn, Easy, Grow, Unique, Start, Be, Love, Discover.
We have also noticed that many are fixated on trying to please the algorithms used by job boards and search engines. As a result, they write for robots, not humans. Getting to the top of the search results is important, but that’s only half the journey. You still have to persuade the human to apply, and a series of ‘keywords’ will not excite them.
Over the next few months, we are going to be testing different approaches in our recruitment advertising; different headlines, new layouts, varying lengths of an advert and a variety of ‘calls to action’. It’s a massive project because to have confidence in the results you should only test one element at a time. But the reward will be worth it.
For more information on this and other talent acquisition advice, subscribe to the Curo Bulletin.