What are Job Description Analysers?

IBM’s email tone analyser kicked this market off a few years ago with a significant breakthrough in the consumer world for intelligent text analysing tools. But in the recruitment world, Textio are making big waves with their job description analysing tools. They describe it as an 'augmented writing platform’ and it allows the user to write or paste their job description into their software, which then punches out a Textio quality score out of 100.

The algorithm uses a combination of machine learning, natural language processing and statistical analysis to find patterns that work to give recruiters an idea of how successful their job descriptions might be. It also currently has access to historical data of over 300 million job listings from more than 10,000 companies.

Whilst it is helpful to know if you have repeated words, used too much jargon, or have used poor formatting, the real potential in the software is uncovering unconscious bias in the words and phrases that are used in the job description, particularly with reference to age, gender and ethnicity.
This makes it a great tool to help contribute to your diversity and inclusion agenda.

Other software providers have recently launched their own version of Textio’s pioneering technology. When reviewing these competitive tools, our advice would be to do your research and ensure you understand what science lies behind the data, as making decisions based on poor data could have serious ramifications on your hiring strategy.

Stage of Hype: Slope of Enlightenment

Case Study

By Matt Eyre, Candidate Marketer, Co-op Digital

Since early 2017 we’ve been working with US-based ‘augmented writing platform’, Textio to iteratively improve our job ads.

I’ve been writing adverts for the Co-op for years, and I’ve worked with various media partners who’ve told me different things about what makes a great ad. I was looking for a tool that could provide me with a more objective and measurable view of what works and what doesn’t in terms of style, structure, tone, etc. And I also wanted to make sure the way I was writing supported our efforts to create a more diverse Co-op.

One of our software developers introduced me to Textio. We enter our job ads once we’ve finished writing them based on the brief from a hiring manager, and we’ll then be given a score out of 100 based on numerous factors including grammar, word count, language, structure and tone. There’s also a scale measuring the gender coding of the wording in the advert, and advice on how to change each area to achieve a better score.

The scoring isn’t arbitrary - it’s based on hiring data collected from hundreds of partners around the world, and it’s specific to particular industries and locations. We share data on a quarterly basis to measure the impact of our writing, but also to allow the tool to learn more about what works best for us.

In seven months we’ve seen significant improvements in time-to-hire, volume and quality of applications, and gender diversity in applicants for tech roles. And while these successes can’t be solely attributed to our augmented writing experiment, we’ve extended the relationship with Textio so we can continue to learn more about what makes the perfect Co-op job ad.   

Further reading 

#RecTechHypeCycle2018 #ExpertsView #SlopeOfEnlightenment

Director of Digital Strategy

Nathan helps organisations optimise their recruitment marketing strategies by leveraging technology, creativity and data. In his role at employer marketing agency AIA Worldwide, Nathan has played pivotal roles in developing content marketing and social media strategies for large multinational corporations, as well as being a driving force behind a number of careers website projects and hiring strategies powered by the agency’s proprietary automated recruitment marketing software, TalentBrew. Catch him on Twitter, where he tweets all things digital marketing, branding and tech.