When we started this experiment in 2017, the Government of Canada only supported flex hours, telework and remote work accessible jobs in rare cases, and these flexibilities weren’t included on job advertisements. We didn’t even know if this was a problem, but it seemed inefficient to recruit people who might potentially need these flexibilities to jobs that didn’t allow them. We wanted to know if a problem even existed, and if it did, whether or not it was impacting application rates and quality-of-fit.
We had a theory that a number of things that weren’t currently on a standard Government of Canada job advertisement might actually have a big influence on attracting high performing talent to apply. We also thought some of this information might help applicants self-select as a good/poor fit for the job and the team. So we set about trying to get a better understanding of what those things might be, and whether or not they mattered.
In 2017, Talent Cloud ran an experiment where we attempted to identify what types of statements might influence an applicant’s decision to apply. We built a list of ~60 sentences or sentence fragments. These were drawn from various components of the draft job advertisement design we were testing, including job impact, manager profiles, work environment, types of technologies used, workplace amenities, team culture, operating context and details such as whether or not flexible work and remote work were permitted.
In a series of workshops, we asked applicants to individually sort statements into piles according to:
The workshops were framed this way to map what types of information and statements would alter the decisions of the applicants. We didn’t just want to know what they liked or disliked - we wanted to know what would change their behaviour patterns.
Following the sorting of statements by each individual participant, we asked them to compare their decisions in a facilitated discussion. Some patterns became immediately apparent, but there were two in particular that surprised us.
Universally, across the workshops, applicants placed “flexible hours allowed”, “telework allowed” and “work from anywhere” into the column for “wasn’t interested in applying, but now i might”. No huge surprise there. But when we asked how many people in the workshops actually required these flexibilities or planned to use them, only ~5% of participants reported needing flex hours (for example, to do school pick up).
So why did a factor that applicants admitted they didn’t need have such an influence over their behaviour pattern? Because applicants in the workshops universally equated managers who allowed these flexibilities with managers who trusted their employees, and who would be more likely to support innovation and creativity elsewhere in the work.
The other surprising result from the workshops was also a finding related to the way in which applicants were “coding” the statements. More than 2/3rd of participants placed government-style “motherhood” statements, such as “The Government of Canada values diversity and inclusion in the application process,” into the “now I don’t want to apply” bucket. This actually shocked us, because why wouldn’t that be something applicants wanted to see?
Applicants explained that broad motherhood-style statements in the job advertisement seemed impersonal and disingenuous. They equated the inclusion of these types of statements with a compliance-driven workplace and a manager that said the things they were supposed to say, not the things they really felt. Applicants reported that they in no way believed that the inclusion of these motherhood-style statements meant the manager actually cared about them or would uphold them in practice. Some even reported that they found these statements “just annoying noise” or offensive to the intelligence of applicants, like a “poorly done magic trick” trying to stop people from looking deeper at what was going on.
As a result of this finding on adverse applicant reactions to broad, motherhood-style statements in job advertisements, we ran a second series of targeted qualitative analysis experiments. These were done with real managers, but not for real job processes in the initial experiment. We asked them to start with a broad corporate statement they might usually include in a job advertisement. Then we encouraged them to personalize the sentiments behind the broad corporate statements, and include this on a manager profile. For example, instead of a broad statement on diversity and employment equity, managers wrote things like, “I believe in a workplace that supports Indigenous inclusion. While I’m not Indigenous, I try to always encourage my scientists to consider Indigenous traditional knowledge in their field studies, and I try to always keep learning from Indigenous colleagues and peers so I can be a better manager.”
We then took these real sentences, produced by real managers, and tested them again with applicants.
The shift from broad statements to personal commitments was so successful in beta testing with applicants that we wrote nudges towards authenticity right into the job advertisement builder that we coded for the platform, and included a manager profile. Ongoing qualitative analysis with applicants for real job processes in 2018-2020 showed that this type of information remained an important factor in influencing applicant behaviour favourably towards managers and job applications. This proved particularly true when managers admitted imperfections and provided bluntly honest phrases, as these things generated a sense of trust in the manager despite applicants never having met them. It made people want to work with these managers.
We also helped managers see the value of flexibilities in the recruitment process by setting new defaults in the job advertisement builder around flex hours, telework and remote work (which was discussed in detail in the previous write-up in Section 2 of this report.) To see a screenshot of these nudges and default settings on our platform, check out the page before this write-up.
"Authenticity matters in the hiring process."
It was a surprising research finding for us to see the extent to which applicants used small phrases and keywords to draw much larger conclusions about managers, departments, and workplace cultures. But it probably shouldn’t have been a surprise at all. This observation is directly in line with an increasing body of behavioural psychology and cognitive sciences research produced in the last two decades that looks at the ways in which the brain interprets fragments of information. In short, it seems the human brain is wired to jump to conclusions.
Humans have a strong tendency to benchmark information based on their personal frame of reference, regardless of the extent to which they have awareness of the context from which the information fragment emerges. (This is also one of the behavioural tendencies that enables racism and discrimination.)
That means that our applicants’ tendency to scan for “code words” that would reveal insights about the broader work environment was right on par with what one might expect based on behavioural psychology research. But it’s not only the applicants’ tendency to extrapolate information that managers should care about when looking to improve their recruitment strategy; it’s also learning what “code words” motivated high-performing talent to apply.
Research shows that it’s not perfection, but authenticity, that employees value in leaders. This includes leaders having self-awareness, being willing to show vulnerability, and being true to their word without being self-promoting - what some have referred to as “quiet transparency”. Employees also value leaders who show trust in their employees, and share authority for decision making where possible. Research shows that the level of autonomy at work is actually a better predictor of happiness than salary, so it’s not surprising that applicants would be looking for “code words” that implied managers showed trust in employees. On the other hand, generic company-generated expressions of value (that didn’t connect to concrete examples of action) did more than fail to inspire employees, they actively generated distrust, and were associated in some cases with micro-management and compliance driven leadership styles. In extreme cases, the failure of authenticity in leadership has even been identified as a leading cause in the financial collapse of industry giants… just as the introduction of authentic leadership has been identified as a core component of financial recovery.
This point of this research insight isn’t to initiate a debate over whether applicants should or shouldn’t infer so much from so little in a job advertisement. The point is that if that’s what’s happening, managers who want strong job advertisements need to factor this applicant behaviour pattern into their recruitment strategy.
This is more than a matter of adding a few keywords in key places. If managers were to become formulaic about always adding the same words, applicants would likely begin to simply disregard this standardized information and look for other indicators that were unique to each job advertisement. In our research we found that there’s no recipe for a strong fit except for honesty, and that requires careful reflection and self-awareness when it comes to articulating the manager’s leadership style, work environment and team culture.
The sample size for these workshops wasn’t large enough to be considered statistically significant - we only engaged ~120 people in total in the initial workshops, plus the follow-up qualitative analysis, and the study of applicant behaviour patterns in the live job processes. But the findings were so consistent from the initial workshop series, we decided to go ahead and build the lessons learned into the platform in the form of default settings and nudges.
While we made changes to our platform to promote these behaviours, they come down to language choices. So there’s no reason these approaches can’t be easily applied in any job application process.
The most important take-away for HR advisors and managers is that applicants will scan every piece of information in a job advertisement for clues about the work environment and the manager, and they behaviourally code this information. They think, “Is this where I want to work?”… and in the absence of direct, transparent sharing from managers, they will draw inferences and conclusions based on very little information. Authenticity and personalization in the job poster therefore become critically important in attracting a top candidate.
More broadly, the Government of Canada might want to run a research study to better understand and confirm or refute these findings. Motherhood-style statements are almost ubiquitous in government job advertisements and as part of recruitment drives. If that’s as big a problem as we suspect, there’s a lot of high-performing talent out there that might be nudged away from ever applying… without the Government of Canada even realizing it was happening. And that’s a loss for Canada.
