Our Work

Scaring the fish away: behavioral insights into talent recruitment

The job recruitment process is a game. Precisely, a signaling game with asymmetric information. Hoping for a perfect match, both the job seekers and the companies looking for new talent signal their attractiveness to each other. Neglecting the in between the lines signals that are conveyed in job advertisements can deter top candidates from applying and reduce the diversity of the company’s biggest asset – the human capital. According to research, less diversity means lower financial returns, making this an important point to focus on. Behavioral insights are not only a beacon for the dangers of implicit biases that guide our decision-making – recruiting process included – but also a toolbox that can be used to create a more efficient and fairer job market. The recruiting process has evolved over the years – exemplary is the adoption of AI software in different recruiting stages. Yet, despite the crucial role of human behavior complexities in the talent acquisition procedure, the use of behavioral economics and insights is still not widespread.  

The Nudge is one of the essential tools that can be found inside the behavioral economists’ toolbox, as it allows to change the status quo and translate behavioral insights into policy design. In 2019, the UK Behavioral Insights Team (BIT) together with Indeed – the global job site – conducted a trial to encourage more flexible jobs and applications in order to tackle job market inequality. 20 million job applications passed through the trial – making it one of the biggest experimental social policy trials ever published. The experiment was initiated after noting a discrepancy between job advertisements and the actual working conditions once the positions were filled – only 15% of job advertisements had some flexibility indications, while 60% of positions would have flexible working conditions once the role was taken. The trial consisted in testing a simple nudge to encourage employers to advertise more jobs as flexible. They changed the way options were presented on Indeed’s job advert template by including a new dialog window, which prompted employers to choose flexibility options for their vacancy. The introduction of the new prompt was a robust enough nudge in itself as it increased the number of flexible job advertisements (+20%). This effect was found to be consistent across all types of flexible conditions, albeit strongest for flexitime. The increase in flexible job listings was matched by a strong increase in job applications (up to 30% more applicants). They also conducted a second trial whose results supported the findings of the first trial but with weaker magnitude (17% vs 20% and 19% vs 30%), accounting for the implementation error during trial roll out. Another key result of the trial was the finding that both men and women were more likely to shortlist job adverts mentioning flexibility, compared to a full-time offer. In fact, this contradicts the common belief that flexible jobs are only attractive to female gender individuals. The global Covid-19 pandemic has pushed for a long overdue transformation of the workplace towards higher adaptability, mobility and flexibility while dismantling, at the same time, the myths about the loss of productivity as a consequence of lower rigidity. Obviously, a careful implementation of increased flexibility is necessary to reap its positive effects on employee’s wellbeing and productivity. 

When it comes to job advertisement, not only the omitted conditions, but also the content and the structure of what is indicated in the ad have important implications on its appeal to the right candidates. In fact, the word choice might have a discouraging effect on prospect applicants depending on their gender identity or their ethnicity, independently on their skills. In this regard, it is known that female gender individuals do not apply for vacancies when they do not meet all the position requirements, while male applicants do even if they meet only 60%. Based on Harvard Business Review (HBR) this occurs because people who do not apply believe they needed the qualifications not to do the job well, but to be hired in the first place. Moreover, HBR also analyzed the company signals that attract or deter female candidates, finding that prospect applicants intuitively infer the company values based on job post descriptions or on representativeness in aspiring roles. For example, expressions like “ninja coder” and “wrestles problems to the ground” are less likely to appeal to female gender individuals as they are associated with male gender stereotypes, with fighting, and with aggression. Also, companies that signal a culture where people are valued and developed are more likely to attract female candidates — and retain them.  

The post description can alienate talent with certain ethnical and cultural backgrounds too. Once more, words and representativeness have a big impact.  An example is the Behavioral Insights Team’s work with nine cities across the US for the solicitation of applications for a policing position. They found that the advertisement postcard attracted more or less applications from people of color based on its message. For example, in Chattanooga, Tennessee, postcards asking “Are you ready to serve?” were no better at soliciting applications than never sending a postcard at all, while postcards that asked “Are you up for the challenge?” or “Are you looking for a long-term career?” attracted three times more applicants than the no-postcard group and drove a fourfold increase in applications from people of color. This shows how increasing diversity passes through appropriate language as well. Another example of a costless behavioral intervention was the addition of a few sentences to the email used to invite candidates to take the SJT assessment test for a policing position. This intervention stemmed from the partnership between BIT, Avon and Somerset Constabulary, and it closed the wide gap in probability of passing the online Situational Judgment Test (SJT) between applicants from a black or minority ethnic (BME) background and non-BME applicants, as the additional sentences prompted applicants to reflect on what might make them a good addition to the force, and what significance that would bear in their community. The BME applicants who received the adjusted email scored five points higher on average which translated into a significant increase in the probability of passing this stage of the assessment process (an increase of 20 percentage points). 

There are several other well-established solutions that are aimed at increasing the quality of the recruitment process and, consequently, of its end result. One such solution is Textio – it uses machine intelligence to help write job descriptions that increase the probability of a positive response. This means assessing the inclusivity and the efficacy of the message, word by word. Applied is another solution that consists in a recruitment platform that removes unconscious biases and aims at reducing the recruiting times while improving its quality and results. But, in the end, it is not the fancy tools that make a good angler; the companies have to understand whether they are fishing in a pond or in a river, and ultimately, whether there is no fish or it simply is the wrong bait. 


Appel J. et al., Behavioral Insights for Building the Police Force of Tomorrow,  23 January 2019, BIT report 

Burd H., et al., BIT’s biggest trial so far encourages more flexible jobs and applications, 4 March 2021, BIT blog 

Correll S. J. , Mackenzie L. N., and Wynn A., If Women Don’t Apply to Your Company, This Is Probably Why, 17 October 2019, Harvard Business Review 

Hunt V., et al., Why diversity matters, 1 January 2015, McKinsey and Company  

Mohr T. S., Why Women Don’t Apply for Jobs Unless They’re 100% Qualified, 25 August 2015, Harvard Business Review 

Ruda S., Promoting diversity in the Police, 24 July 2015, BIT blog 

Timewise Flexible Jobs Index 2019 and 2020, http://www. 

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s