High attrition rates, absenteeism, and job-hopping are the biggest challenges data science recruiters face. Based on our past interactions with leading companies, we have zeroed in on a few common mistakes organisations make while recruiting data scientists.
Calling it a data science job when it actually is not: Most companies like to call a range of job roles data scientists when in reality, it could be machine learning engineer, big data developer, business intelligence analyst, data engineer, and so on. Recruiting for data scientists’ role but assigning tasks that do not sync with expectations is a big turn-off for candidates, which may lead to them quitting the job. Companies often put them in roles where, for example, only a data analyst is required. It quickly demotivates the data scientist and erodes their skill sets. The companies should be transparent on roles and responsibilities and the kind of project engagements the candidates will have. Setting the right expectation is crucial.
Companies are not sure if they want to use data science in the first place: Data science is quite popular, and many companies tend to create data science roles without knowing how data science can help the organisation. In such companies, the data scientists will have no clue as to what they are expected to do because the companies have not defined the roles in the first place. The confusion can lead to demotivation and, finally, result in the quitting or firing of the data scientist.
Too much focus on math and statistics and overlooking problem-solving skills: While technical expertise is one of the primary needs for data science candidates, focusing only on math, statistics, or the tools is one mistake many companies make. Data science is a lot more than understanding complex algorithms. It is about the right approach. During hiring interviews, companies must give data scientists real-time assignments to assess their problem-solving and analytical skills. The questions should also be customised according to the previous use cases that they have worked upon so that companies do not overlook the candidate’s actual skills in problem-solving and the ability to approach a problem.
Not analysing the business acumen: Many companies pay more attention to the technical side and ignore data science candidates’ client engagement skills. However, focusing on candidates who can find a solution to a business problem and communicate the same to the client in a precise manner is equally important. Not exposing the candidates to real-time business problems and consumer-impacting decisions may lead to a wrong hire. The interviews should also assess the business understanding of the candidate as a priority.
Weighing academia over skills and hands-on experience: A standard error most companies make while recruiting data scientists is giving weightage to candidates with good academic records at the expense of practical skills. Data science jobs require the ability to crunch data and solve complex industry problems. A good educational background is not a guarantee for good results. The candidates will not be successful in their roles of developing fit-for-purpose, AI-first solutions if their practical skills are not on par.
Not focusing on data storytelling: Many companies make the mistake of not focusing enough on soft skills such as communication, collaborative mindset, and especially their capabilities in data storytelling while recruiting data scientists. One of the most important aspects of data scientists is to be able to share the data and insights with all the stakeholders in the organisation and to clients.
Hasty screening process: Since data science is a trending field, it is extremely important to differentiate between those who have actual expertise from those who might have the basic knowledge. Not doing due diligence and completing all key fundamentals of the hiring process may lead to a wrong hire. Late hire is always better than a wrong hire.
