Since June 2020, I’ve made it a weekly habit to have conversations with strangers who reach out to me to discuss various topics related to the world of data. Most of these discussions are with junior profiles, people unsure of how to start or train themselves, or those looking to transition into a Data Science role.
These conversations have been incredibly enriching for me, but above all, they’ve helped me become more attuned to the varying speeds at which the Data Scientist job market evolves depending on a candidate’s experience. Based on insights from these discussions and others I’ve had within my network, between 2015 and 2019, an inexperienced Data Scientist could land open positions without much difficulty. Starting in 2020, however, I began noticing that breaking into that first job became increasingly challenging. This shift is primarily due to the proliferation of Data Science training programs, which have created a large pool of candidates fiercely competing for entry-level roles.
By 2022, we found ourselves in a situation where novice candidates for Data Scientist roles weren’t landing jobs as quickly as they’d hoped. Training schools bombard students (their clients) with statistics about the shortage of data professionals and the high salaries in the field. While this is true to some extent, it’s only part of the story: the current shortage primarily concerns experienced professionals (especially those skilled in deploying solutions into production with measurable business impact) or highly specialized experts (e.g., NLP specialists, biostatisticians, or Computer Vision experts).
It’s not you—it’s the market.
In many of these conversations, we talk about the frustration caused by this mismatch between candidates’ expectations and their real-world job search experiences. One of my key messages, something I’ve outlined in earlier paragraphs, is simple: it’s not you; it’s the market. It’s crucial for people to understand that despite all the siren songs promising abundant jobs and high salaries in this field, they’ll be competing against many others in similar situations. It’s not necessarily about whether you’re “good enough” for this job.
Once this reality sinks in, I suggest taking the next logical step: building a strategy. By strategy, I mean intentionally orienting your job search instead of approaching it haphazardly. One strategy might involve applying indiscriminately to every position labeled “Data Scientist.” Alternatively, another approach could focus on optimizing your search by targeting only roles that align closely with your profile and expectations.
Applying everywhere might not be your best strategy.
Most people I speak with tend to take a scattershot approach, applying for every available position they come across. While this isn’t inherently wrong as a short-term strategy, if it doesn’t yield results quickly enough, you may find yourself buried under technical tests and endless processes that bring significant stress along with them. That’s why I believe candidates should approach their job search with an explicit strategy, one they can evaluate and adjust based on feedback received during their attempts.
Here’s an example framework I often suggest:
1. Analyze Your Strengths and Weaknesses
This first step is obvious but often overlooked. The most critical part is reflecting honestly on where you’re starting from. For instance:
- If your background is in social sciences (hello fellow economists!), you’re likely strong in data analysis, analytical thinking, modeling techniques, statistics, and causal inference.
- Conversely, if you come from technical fields or engineering roles, your strengths probably lie in algorithm knowledge, computational skills, code efficiency practices, software development best practices—and perhaps deeper technological expertise.
Simplifying broadly between these two profiles: one group’s strengths often represent gaps for the other group—and vice versa—but ultimately each individual must perform their own honest self-assessment.
And by honest, I mean just because you’ve taken one SQL class or completed an introductory module on Deep Learning doesn’t make you an expert in those areas.
2. Target Roles That Align With Your Profile
Once you’re clear about your capabilities and gaps—it’s time to identify opportunities that match them best. Remember that Data Science is an umbrella term encompassing numerous roles whose meanings vary widely across companies.
Within Data Science itself are positions closer to Data Engineering or Data Analysis, or even ML Engineering or DataOps, and each company defines these roles differently! To maximize your chances at landing that first role:
- Focus on opportunities closely aligned with your strengths.
- Be proactive during initial interviews (e.g., HR screenings) by asking questions aimed at uncovering what specific functions companies expect from their hires.
Sometimes reaching your ultimate goal, a full-fledged Data Scientist role, might require taking a detour through adjacent positions like Data Analyst or even entry-level “Data Science” jobs focused heavily on analytics rather than engineering challenges:
- Candidates from less technical backgrounds may find Analyst roles helpful since they involve more business-facing tasks while exposing them less frequently to engineering problems.
- On the other hand—those coming from engineering-heavy backgrounds might consider starting as Data Engineers—gaining autonomy over data preparation pipelines essential for future projects involving advanced analytics.
Taking such detours can enrich your experience significantly while still moving toward long-term goals!
3. Showcase Skills During Selection Processes & Self-Critique
Finally, in selection processes, you need strategies tailored toward demonstrating your unique strengths effectively:
- Analytical profiles should excel at exploratory data analysis (EDA), identifying issues like missing values/outliers—and discussing how such insights translate into actionable business outcomes.
- Technical profiles should focus on showcasing reusable libraries they’ve built, or demonstrating proficiency implementing efficient code using sound software development principles (e.g., optimization).
The key takeaway? Reflect critically post-interview/test cycles: analyze what worked/didn’t work—and refine both skillsets/approaches accordingly (more here)!
“Goals are for Losers” - Scott Adams
Ultimately, as Scott Adams suggests, it helps more long-term thinkers embrace systems over rigid goals! My personal system involves continuously exposing myself professionally toward challenges/tasks outside current expertise (“yet” mindset) motivating growth along uncharted paths!
If you’re still striving toward securing that elusive first DS position—I hope reflections shared here spark fresh perspectives/reinvigorate motivation throughout journeys ahead.