When I'm working on a careers site project, there's one potential stumbling block that looms larger than any other. I'm not talking about design revisions, or getting the copy approved by legal - it's the integrity of the jobs data that can cause the most pain, if we don't approach it correctly from the start. 

Awareness is the first step

Where do your jobs reside? 

In one system, or in many? If you have more than one system that holds your jobs, you have more than one data source - and we'll need to normalise that data when we bring it together. 

How structured is the data? 

Are locations, categories, and other fields populated from drop-down lists or are they free text? The more structured the data, the easier it will be to work with - if every field is free text, you'll have a lot of variety. 

How detailed is the data? 

For example, when you enter a job location, do you enter the full street address with postcode, or just the city name? The more detail we have, the more detailed the solution can be. 

Do something with your knowledge

Once you understand what your data looks like, you'll want to identify the opportunities to improve it. 

Data Clean Up

Can you go through the ATS and correct postings? If one big push can get you 80-90% of the way there, it might be worth the effort. 


If your recruiters are posting the jobs with incomplete or incorrect data, make training them a priority. Show them what a good post looks like, and what happens if the data isn't right. 

Treat the source

Can you reconfigure your ATS to ensure the data is more structured / better quality when it's first entered? Selecting locations via a pre-populated drop-down instead of entering them as free text is a good example. 

Early and often 

To ensure that your data integrity is high, start this process as early as possible during your careers site project. It often takes longer than you think it will. Once you complete the initial remediation, schedule regular checks - keeping it clean will take regular time and effort. 

Bring it together 

If you have data from multiple sources, even once you've standardised aspects like location and category, it might make sense to bring all of your jobs data into one database. A middleware solution can help with integrating your jobs with your content, as well as distribution of those jobs to social media and other destinations. 

Now what?

After you've got accurate and complete data, you can figure out what to do with it! 

Once you know what you're working with you can design pages that you know works with your data structure, and integrate it with your site in a way that reduces your QA time. It might seem like this is a lot of effort just to get started, but it will make all future phases, from design to build to testing and launch, much smoother. 

VP Technology Operations

Karen brings her experience in project management and recruitment technology together to help teams deliver outstanding digital recruitment solutions. She’s worked with AIA Worldwide for eight years, in both New York and London, and never met a spreadsheet or project plan she didn’t like.