Sheena Hartlooper and Claudia Zwitser are responsible for recruitment marketing at Nederlandse Spoorwegen (NS), a railway operator based in the Netherlands.
If you’ve spent even a small amount of time in the Netherlands, the chances are you’ve boarded one of our NS trains.
We are the biggest rail company in the country and also one of its largest employers, with a staff of 20,000 — and counting. But while we are looking to hire more people, job applications have been down. And we’re not alone.
The recruitment struggle is one many companies can relate to. In the Netherlands, 84% of employers report labour shortages.1
While traditionally recruitment may be seen as more of a human resources responsibility, marketing has a big role to play in getting the right job advertisements in front of the right people. And with AI-powered marketing on the rise, it creates new opportunities for brands to find job seekers they struggle to reach using more traditional methods.
At NS, we were in a transitional phase. Crucial members of our workforce — train drivers, managers, and security — were retiring, and we needed new recruits to fill the void. But with applications decreasing, it was time to find a conductor that could drive them forward again.
All change, please: The challenge of recruiting a new workforce
Following years of disruption, 2022 saw railway employees across some of Europe begin striking in order to obtain better working conditions and higher wages.
These events negatively affected the public’s perception of many brands, including NS, attributing to the decrease in applications.
Not only did we need to reach more people, but we also had to be smarter and more cost-effective.
Most years, we hire around 5,000 new people. In the past, we have created radio campaigns, used paid social, uploaded job applications to recruitment sites, and staged a number of in-person recruitment events.
But our usual methods in talent acquisition were seeing a decrease in job applications and quality visits to our recruitment portal, as measured by the amount of time a user spends on a webpage.
Furthermore, cost-per-application (CPA) — which measures how much money it costs to get a job seeker to apply — was rising due to national market trends.
For our next recruitment campaign, a new tactic was needed.
Not only did we need to reach more people, but we also had to be smarter and more cost-effective.
All aboard – how to use AI in recruitment
In our roles as recruitment marketers, we work closely with an agency called Dept, who assist us with our recruitment drives.
Dept recommended custom bidding, which would offer a more cost-effective solution by identifying the value of each potential applicant interaction. This AI-powered tool would be integral to understanding how different parts of the process on our website contributed to an application.
To do this, we assigned points to each impression, informing the system of what we truly value:
This tactic favoured an impression that leads to job applications, rather than just a landing on a page This is something we kept in mind as we set up our digital campaign: display banners across all social platforms and job application sites.
“Within the campaign, we created two sub campaigns,” explains Daniël Landman, senior programmatic consultant at Dept.
“The first was aimed at maximising conversions, which means increasing the number of people who engaged with the advert and were later seen on the website. The second focused on delivering the highest custom defined value per impression; so bidding on ad placements that ended in an actual job application.”
We separated people based on their exposure to the campaign by using the experiments tool in Display & Video 360.
Landman continues: “We did this to initiate a clean experiment that ensured people were not exposed to both sub campaigns, as this would make it hard to properly attribute the conversion to the testing variant.”
The Dept team also created UTM tracking — codes that include a traffic source, medium, and campaign name — to analyse applicants’ onsite behaviour.
The final stop – increased applications and engagement
We saw positive results within just a few weeks. Most notably, a 25% surge in applications.
Many of these had been submitted by applicants who had previously not visited our job applications website, but saw our display banners and completed an application. It showed that display banners can drive action as well as awareness.
Like our trains themselves, this project rolls on. We are still using the custom bidding script and continue to see an uptick in talent acquisition. There might be disruptions or new stops along the way, but this is a project that found the right platform.