How game development teams identify and hire the right junior analysts

Maxim Kozhnov is chief revenue officer at Azur Games.

Over the past year, the number of candidates applying for junior analyst roles has been higher than ever. Many of these candidates boast impressive skills - knowing Python, writing SQL queries, explaining R1, and calculating ARPDAU.  

But paradoxically, this doesn’t mean we’re flooded with perfect candidates. Having conducted hundreds of interviews, sometimes as many as one per day for weeks, I’ve learned that specialised education or analytics experience doesn’t always translate to suitability for game development.

In fact, some of the strongest analysts I’ve worked with came in as juniors with no technical skills at all.  

For me, the most important trait in a candidate has always been their ability to think logically and with a product-oriented mindset. To help junior analysts navigate this process, I’ve outlined a few key points employers use to identify the real “diamonds” among the sea of technical graduates.  

What matters more: knowledge or logic?

Juniors are typically hired to work on a project where there’s already a senior analyst or a cross-project team led by a more experienced analyst. This setup allows the junior to learn on the job, which is crucial because analytics in game dev differs significantly from other industries in both tasks and expectations.  

It’s common to see candidates with years of experience in other fields - banking or retail, for example - who consider themselves mid-level or even senior analysts. And while that may be true in their industry, they often start as juniors in game dev. With the workflow so different, this transition is normal.  

Today’s flow of candidates is far greater than it was five years ago. Alongside the usual output from tech universities, it’s fueled by the rise of analytics courses and bootcamps, making it impossible for senior analysts to interview everyone personally.

To secure a more in-depth conversation with a lead analyst, candidates need to demonstrate that they’re already familiar with the basics. This means knowing key metrics, understanding their purpose, and being able to calculate them.

The hiring process typically starts with a 20-minute screening call conducted by HR. During this call, the candidate is asked a few straightforward questions to gauge their foundational knowledge. If they perform well, they’re invited to a longer interview where specific tasks and future responsibilities are discussed in detail.

But the truth is, candidates don’t need to have all the answers unless we’re talking about senior roles. What stands out is how they approach problems during the interview. A candidate who can think through a problem and arrive at a logical conclusion demonstrates quick learning ability, adaptability, and depth of thought.  

On the flip side, knowing metrics but failing to understand their purpose or application is a red flag. Some candidates confidently throw around jargon but falter when asked about its practical use - confusing direct relationships with inverse ones or misinterpreting data.  

Then there are the positive cases: candidates who don’t know specific metrics but, when prompted, can logically deduce what parameters are essential to evaluate a project’s success. For example, rather than merely stating, “A project’s success is measured by how much it earns”, they consider factors like revenue versus expenses. These are the types of business-oriented thinkers who often grow into senior analysts and team leads.

How to test a junior’s skills

Junior candidates often prepare for interviews by reading articles, memorising terms, and brushing up on industry jargon to make a good impression. But as I’ve mentioned before, the most important thing for us is not how much they know - it’s how well they can think logically and draw connections between cause and effect.

To assess this, we design product-focused tasks that don’t require prior technical knowledge. Instead, they rely on the candidate’s ability to reason logically and think critically. Here are a few examples:

1. Warm-up: general understanding of metrics

We’ll present a hypothetical project and ask the candidate to identify key metrics to monitor.

The goal here is to see if the candidate can pinpoint the metrics that truly reflect the project’s performance and explain why those specific ones matter. 

Even if they don’t know the exact names of every metric, they should be able to reason through what’s important and why.

If someone has a solid foundation in logical thinking, teaching them the technical terms and formulas later is relatively easy. Without that foundation, however, a vast vocabulary of metrics won’t get them far.

2. The correlation problem

Imagine a midcore game where data shows that players who lose their first battle have higher retention than the project average. Does this mean we should design the game to ensure all players lose their first battle?

The right answer is no. Correlation does not imply causation, and the candidate should recognise this.

In this example, retention is higher among “losers” because they stayed to complete the battle. In contrast, the overall average includes players who dropped off before the battle ended (or even before it began, like during loading or the tutorial). The comparison is actually between “players who finished the battle and lost” and “all players”, not “losers” versus “winners".

3. Cannibalisation of purchases

A game has 10 skins, each generating $100 per day. You release an 11th skin that brings in $120. Is this an improvement?

The key here is to assess whether the total revenue from all skins has increased. It’s possible the new skin is cannibalising sales from existing ones, leading to no net gain - or even a loss. 

A mid- or senior-level analyst should know this immediately, but for juniors, it’s already a positive sign if they can think through this scenario logically, even without knowing the specific terms.

4. Evaluating product decisions

Suppose you change a post-battle screen from a “Return to Menu” button to a “Go to Next Battle” button. After this change, retention and session duration increase. Is this a good outcome?

At first glance, the metrics might suggest a success. But a deeper look could reveal unintended consequences. For instance, by skipping the menu, players might engage less with meta-game elements like levelling, in-app offers, or leaderboards - areas crucial for monetisation. The candidate’s answer should explore the broader implications of the change, not just the surface-level metrics.

This task also helps gauge the candidate’s understanding of game design concepts like meta-progression and player engagement loops.

5. Simpson’s Paradox

This question is a bit advanced - typically for mid- or senior-level candidates - but it’s impressive if a junior can tackle it.

Imagine your audience consists only of users from the US and India. After an update, the LTV for both regions increases, but the overall LTV for the game decreases. How is this possible?

This is where Simpson’s Paradox comes into play - it’s entirely possible. At first glance, if the LTV increased in both the US and India, you’d expect the overall LTV to rise as well.

However, this isn’t necessarily the case because it depends on the relative weight of each country in the total sample. If the US and India each account for 50% of the audience, then yes, the total LTV would likely grow. But if the distribution is uneven - say, a larger proportion of installs come from India, where LTV is typically lower - the overall LTV can actually decrease despite individual improvements in both regions.

The candidate should address two questions: whether the statement is true and whether the updates should be evaluated based on regional data or the aggregate audience. This opens up plenty of room for thoughtful reasoning.

Not even half of the self-proclaimed mid-level candidates manage to answer all these questions during interviews. That’s because these aren’t the kind of questions you can simply memorise answers for - they require a specific type of critical thinking.

And that’s intentional. If everyone could answer all the questions, they’d serve no purpose. Conversely, if no one could answer them, they’d be too difficult for a junior-level position.

What truly matters is how a candidate approaches the answer. Did they follow a logical line of reasoning, or were they just guessing? Even if the answer is wrong, a well-thought-out and non-generic approach can spark genuine interest and significantly increase their chances of passing the interview.

Should a junior analyst know how to code?

At our company, we require juniors to have a technical foundation - knowledge of Python and SQL is non-negotiable. This isn’t because we’re being overly demanding but because the sheer volume of applications makes it impractical to consider candidates without these basic skills.

That said, I’d personally prefer to hire someone with excellent problem-solving skills over someone who’s just technically proficient. Hard skills can be taught relatively quickly, especially to someone with strong analytical reasoning. On the other hand, most applicants today already come with some knowledge of the basics, thanks to the abundance of easily accessible online courses.

Always keep in mind that technical skills don’t make a great analyst.

I’ve seen cases where highly skilled coders overcomplicate tasks unnecessarily. For example, something that could be solved in Excel in 10 minutes ends up taking hours because the analyst insists on writing hundreds of lines of code in C++ (yes, it was that long ago). In mobile game development, where speed and adaptability are crucial, this kind of inefficiency isn’t sustainable.

Today, most technical universities teach Python and SQL, and even tools like ChatGPT can generate SQL queries. What’s far harder to teach - and infinitely more valuable - is product thinking, which is basically a soft skill that comes from logic, as well as both direct and indirect experience.

When I used to hire juniors, I would ask them to calculate basic metrics like retention or ARPU during the interview, even though this process is already on track in our company. Then, I’d evaluate not just their math but how their calculations aligned with logical reasoning. It’s crucial that candidates understand not just how to calculate a metric, but why they’re doing it in the first place.

Developing this kind of thinking takes time when you’re a grown-up - far longer than it takes to learn Python or SQL - and that’s going to set the whole team back.

How to fit into the team

Beyond understanding metrics, thinking in product terms, and having basic hard and soft skills, the ability to integrate smoothly into a team is crucial for any analyst. Even the most skilled candidate will struggle to succeed if they can’t adapt to the team dynamic.

Communication is key, as analysts interact with multiple departments and need to adapt to various workflows. For mid- and senior-level roles, this also involves taking initiative and demonstrating responsibility. 

To ensure a smooth transition, it’s a good idea to introduce promising candidates to the team and workflow early on. This way, newcomers can gauge whether they’ll feel comfortable and engaged working on the project. Often, you can tell fairly quickly which team a person will thrive in just by observing how they interact with others.

In larger companies, teams can vary significantly in working style. Some teams operate in high-energy, dynamic environments where constant discussions, debates, and experiments drive progress. For someone who prefers a quieter, more introspective approach, these environments can be overwhelming and may make it difficult to have their voice heard.

Other teams follow a steadier pace, where people approach the analyst only when needed for advice or data interpretation.

It’s important to remain authentic during the hiring process. Trying to fit into a team dynamic that doesn’t suit your personality will only create challenges down the line.

And don’t be surprised if you apply for one position but end up being offered a role on a different project. This is often because the company recognises where your strengths might shine best, and you’re just unfamiliar with all the internal processes yet.

Final thoughts

The number of junior analyst applicants today is immense, and it’s a reality candidates and employers alike must accept. Most candidates will first go through an initial HR screening to assess their technical skills. While technical knowledge is essential in analytics, it’s not the most critical factor.

What truly sets candidates apart is their logical thinking and ability to understand products from a business perspective. You might not know all the technical jargon or formulas, but if you can use reasoning to reach the right conclusions, you’ll stand out from the crowd. Hard skills, like writing SQL queries or reading dashboards, can be learned over time.

Proactivity also plays a significant role in a candidate’s success. Juniors need to be ready to communicate frequently, absorb feedback from their team lead, and adapt quickly to new information. However, talented candidates are often placed in teams where their personality and skills align best with the group’s working style.