Did you know that two candidates with similar experience and skills may have different chances of getting an interview, just because their names suggest different origins?
Research shows that biases can influence how we evaluate candidates from the CV screening stage. Sometimes, it’s not their lack of skills that puts them at a disadvantage, but the assumptions we make about them. These biases, known as biases , can influence who we invite to interview, how we interpret their answers, and who we offer the job to. And the good news is that we can take concrete steps to reduce their impact.
A fairer selection process does not mean completely eliminating intuition, but ensuring that hiring decisions are based on skills, experience, and criteria relevant to the role.
What is bias in recruitment and why does it matter?
Bias in recruitment is a tendency to evaluate candidates based on assumptions, stereotypes, or preferences that are not necessarily relevant to the job requirements. These can be conscious or unconscious and can influence how we interpret a resume, conduct an interview, or make the final decision.
The most common sources of bias include:
Physical appearance: forming an impression of the candidate's skills or personality based on appearance;
Candidate's name: preferences or assumptions related to ethnic origin or nationality suggested by the name;
Educational institution: favoring candidates who have graduated from prestigious universities;
Similar experience: preference for people with a professional background similar to that of the recruiter or the team;
Age: assumptions about adaptability, energy, or learning ability based on age.
Bias can contribute to discriminatory decisions, but the two concepts are not identical. That is why it is important to identify situations where personal assumptions risk influencing the assessment and build a process that limits this risk.
The impact can be significant. Imagine two candidates with comparable experience and skills, whose only relevant difference for our example is ethnic origin. According to a meta-analysis published inthe European Economic Review (Lippens, Vermeiren & Baert, 2023), the ethnic minority candidate will receive on average 29% fewer responses , and the difference can reach 41% for candidates of Arab origin.
A study by researchers at UC Berkeley and the University of Chicago (Kline, Rose & Walters, 2024), which analyzed 83,000 fictitious applications for 11,000 entry-level positions at Fortune 500 companies, confirms that name-based discrimination remains a real problem. Typical employers contacted candidates with names associated with white people about 9% more often than candidates with names that suggested a different ethnic origin.
The consequences are not limited to rejected candidates. Companies can lose valuable people, unnecessarily narrow the talent pool, and miss out on diverse perspectives that would contribute to team development.
Romania, an increasingly diverse labor market
In recent years, Romania has become an increasingly attractive destination for workers from other countries. The number of temporary residence permits for employment purposes issued to non-EU workers increased from 36,660 in 2021 to over 102,000 in the first 11 months of 2024 — almost a tripling in three years.
According to a study by the Economic and Social Council, approximately 200,000 people from outside the EU were working in Romania in 2024, most of them in construction and manufacturing. For employers, diversity also means the responsibility to evaluate each candidate according to criteria relevant to the role, regardless of nationality, origin or other personal characteristics.
European data shows that discrimination remains a major concern. According to the Eurobarometer , 1 in 6 Europeans feel discriminated against , with the most common contexts being the workplace (23%) and job hunting (13%). In Romania, studies show that the percentage of those who have personally experienced discrimination at work is significantly above the European average of 17%.
A structured selection process helps you evaluate candidates for what they can do, not for assumptions you might have about them. And fair treatment is not only an ethical responsibility, but also a way to avoid losing talent for reasons unrelated to the job.
The most common types of bias in recruitment
To reduce bias, you first need to be able to recognize it. Here are four forms that can appear in the selection process.
1. Confirmation bias
We tend to look for information that confirms the first impression we formed about someone.
For example, if a candidate is a few minutes late for an interview and you immediately consider them disorganized, you may pay more attention to answers that confirm your initial impression and ignore evidence that contradicts it.
How to narrow it down: write down the concrete evidence for each evaluation criterion and check if the initial impression is supported by behaviors relevant to the role.
2. The halo effect
A positive characteristic of the candidate ends up influencing the evaluation of all other competencies.
For example, a candidate graduated from a prestigious university, and the recruiter automatically assumes that they also have excellent communication or collaboration skills, without evaluating them separately.
How to narrow it down: Evaluate each competency individually. The prestige of a university or success in a previous role is no substitute for evidence of the skills required for the current job.
3. Similarity bias
We tend to prefer people who are similar to us in experience, interests, personality, or professional background.
For example, a recruiter may prefer a candidate who has worked in the same industry and has a similar career path to theirs, even if another candidate better demonstrates the required skills.
How to reduce it: Define selection criteria before you start interviews and evaluate candidates against the same standards. A different background does not automatically mean a lack of the necessary skills.
4. Anchoring bias
The first information received about the candidate can disproportionately influence subsequent evaluations.
For example, if the first interviewer gives a candidate a very low score, other team members may be influenced by that assessment before forming their own opinion.
Even names can influence the selection process. A 2022 study by researchers at Hamilton College and Vassar College shows that candidates with names that are harder to pronounce may also be less likely to be contacted, even when their qualifications are identical.
How to reduce it: Ask interviewers to individually write down their assessments before discussing in a group and before consulting colleagues' opinions.
Practical strategies for reducing bias
You don't need a big HR department or a complicated process to make recruitment fairer. A few simple rules, applied consistently, can make a difference.
Define your assessment criteria before you start recruiting. Determine what technical skills, interpersonal skills, and experience are required for the role and how you will assess each of them.
Standardize your interview questions. Use the same set of basic questions for all candidates applying for the same position. You can ask follow-up questions for clarification, but keep the same evaluation criteria.
According to a report on recruitment statistics , 72% of companies that have reduced bias in their hiring processes use structured interviews .
Use a rating scale with clear criteria. A numerical score is only useful if you know what it represents. Instead of assigning grades based on a general impression, define rating levels for each competency.
For example, for a communication skill, you can track whether the person structures their ideas, provides concrete examples, and adapts their explanations to the interlocutor.
Give individual grades based on the answers and evidence observed. Then compare the ratings and discuss the differences between them.
2. Diversify the interview team
When the decision is made by a single person, individual biases can influence the entire process. Involving multiple interviewers, with different perspectives and experiences, can help identify points of view that a single person might overlook.
A diverse team can contribute to a more balanced assessment, provided that all members use the same criteria. Establish clear roles for each interviewer. One might focus on technical skills, another on relevant experience, and another on how the candidate approaches collaboration, provides feedback, or handles difficult situations.
Evaluate the behaviors and skills relevant to the job, not how much the candidate resembles the existing team.
3. Anonymize CVs in the first stage
Sometimes information such as name, photo or age can influence the assessment before the recruiter has truly analyzed the candidate’s experience. Anonymizing CVs involves temporarily removing such information, where possible and appropriate, so that the initial assessment can focus on relevant experience, education and skills.
Why does it matter? A candidate with a less familiar name shouldn't be less likely to get an interview for a reason unrelated to the job requirements.
According to a report , 57% of candidates say that anonymized recruitment increases their confidence in a fair process , and the percentage rises to 66% among candidates from ethnic minorities.
However, anonymization does not eliminate all forms of bias. Some clues may remain visible on the CV, and prejudices may emerge later, during the interview or the final decision. Therefore, the method works best alongside clear criteria and structured assessments.
4. Create an inclusive environment from the first contact
The recruitment process starts before the first interview. The way you word your job posting can influence who feels encouraged to apply.
To attract the right candidates, avoid:
Aggressive or vague language: phrases like “we’re looking for a ninja” or “a rockstar who can handle any challenge” don’t explain what the role actually entails;
Unrealistic requirements: for example, requesting 10 years of experience for a junior position without a justification related to the responsibilities;
Wording that suggests demographic preferences: any requirement that is not justified by the nature of the job may discourage relevant candidates.
Instead:
Use clear, accessible and inclusive language;
Describe the responsibilities and skills required, not just an ideal profile;
Establish realistic requirements and differentiate between mandatory skills and those that can be developed;
Highlights learning and development opportunities;
Provide all candidates with clear information about the stages of the process and expectations for each stage.
The goal is not to make the announcement as general as possible, but to make sure that the requirements are relevant to the job and that the right people can recognize themselves in them.
How to measure and improve the selection process
A fair selection process doesn't just depend on good intentions. You need data to help you see where differences occur and whether the established rules are being followed.
Here are some indicators you can track:
Passage rates between stages: what proportion of evaluated candidates reach the interview, the final stage and the offer?
Consistency of ratings: Do interviewers give similar scores when reviewing the same evidence?
Compliance with criteria: are decisions documented in relation to the requirements established before recruitment?
Candidate feedback: are there recurring signals of lack of transparency or perception of unequal treatment?
Recruitment time: how long does the process take and where are delays occurring that could affect the candidate experience?
Retention rate : Do employees stay with the company and manage to fulfill their responsibilities in the long term?
If legal and appropriate, you can also analyze differences between groups of candidates, while respecting data protection rules and sensitive personal characteristics.
An observed difference does not automatically prove discrimination, but it may indicate a stage worth investigating. For example, if a group of candidates is significantly less likely to pass the CV pre-selection, check whether the requirements and criteria used are truly relevant to the role.
To track these indicators, you need clear recruiting metrics and consistent analysis. Data is not a substitute for professional judgment, but it can help you identify issues that you might otherwise overlook.
Common challenges and how to overcome them
Even though the idea of a fairer selection process sounds good, implementing it can seem difficult when you have little time or limited resources.
"We don't have time for complicated processes."
You don’t have to change everything at once. Start by defining the criteria for the next role and standardizing three to five interview questions. Add a simple evaluation grid and ask each interviewer to write down their findings before the final discussion.
Small steps are easier to adopt and allow you to observe what works.
"Intuition has helped us so far."
Experience and intuition can be useful, but they should not replace the assessment of skills. A good or bad impression can be a starting point for further questions, not a conclusion in itself.
Before making your final decision, ask yourself: What concrete evidence supports my assessment? Would I make the same decision if I removed personal information that is not relevant to the job?
"We can't anonymize all resumes."
You don't have to start with a complex process. You can limit access to personal information at the initial stage where tools and procedures allow it. If anonymization is not feasible, compensate with well-defined pre-selection criteria and consistent assessments.
The important thing is to choose measures that you can realistically and consistently apply.
The long-term benefits of reducing bias
Reducing bias in recruitment isn't just about treating candidates more fairly. It's also about making better-informed decisions and not excluding people who could contribute to the team's success.
According to the McKinsey report “Diversity Wins” (2020) , based on data from over 1,000 companies in 15 countries, companies with the most ethnic diversity in their leadership teams are up to 36% more profitable than companies with the least diversity. A 2023 McKinsey report also highlights the association between diversity and financial performance.
These results do not mean that diversity automatically guarantees higher profits, but they support the importance of building teams where people have the opportunity to contribute.
A healthier organizational culture : employees are more likely to feel appreciated for their real contributions;
A better reputation as an employer: transparency and consistency can help attract talent.
Remember!
Bias in recruitment cannot be completely eliminated with a single measure. But it can be reduced through clearer processes, relevant criteria, and consistent assessments. Before your next interview, try a simple exercise: define three essential competencies for the role, prepare a question for each, and determine in advance what a good answer would look like. Then evaluate each candidate based on evidence, not just overall impression.
You don't have to be perfect or change the entire process overnight. The important thing is to identify where biases may arise and take concrete steps to make better decisions.
Busy Bee — more order in the recruitment process
A fairer selection process starts with clear requirements, consistent criteria, and well-organized applications. Busy Bee is an ATS built for small and medium-sized companies that want to more easily manage recruitment, from posting ads on their own career site to organizing applications.
The platform helps you keep your applications organized and have a clearer picture of the recruitment process, so you can focus on evaluating candidates.👉 Enter our digital hive, discover why to choose Busy Bee and see how you can simplify your recruitment process, step by step. 🐝
Are you ready to buzz peacefully?
Create a branded careers site, publish roles, and start collecting applications—effortless and quick.