Data analyst at a casino: BI, GGR/NGR metrics, and pay bands

In a casino, every slot spin, every deposit and every declined bonus is a row in a database. Someone has to turn millions of those rows into meaningful numbers: how much the platform actually earned, who to retain with a bonus, and who it is time to show deposit limits to. That is the job of a data analyst — one of the most in-demand and yet least «gambling-looking» professions in iGaming: from the outside it is ordinary SQL and dashboards, but inside are metrics you won't find in product analytics textbooks. Here is what a data analyst does in a casino, which numbers they hold in their head every day, what stack is needed, how the role differs from classic product analytics, and what it pays in 2026.

Demand for the role is growing faster than for most other professions in the industry: operators are scaling into more markets, adding new game providers, and facing tighter regulatory requirements for transparent reporting all at once — and none of that works without someone who can pull an accurate, audit-defensible number out of that stream.

What a data analyst at a casino does

The role is formally called Data Analyst or BI Analyst, but in practice it's the connective tissue between product, marketing, CRM, compliance and the CFO — all of whom regularly show up asking "so what's going on with the numbers." The core areas of responsibility:

• Regular reporting. GGR and NGR broken down by product (slots, live casino, betting), GEO, VIP segment and game-supplier partner — usually weekly and monthly for management.
• Self-serve dashboards. Tableau, Power BI or Looker for the CRM, marketing and product teams, so managers can look at the numbers themselves instead of DMing the analyst every day.
• A/B tests. Testing hypotheses on bonus mechanics, the registration funnel, landing pages, and email/push campaigns — accounting for the fact that player wins badly skew the sample.
• LTV and churn models. Forecasting how much NGR a player will bring in over their whole lifetime on the platform, and which of the currently active players will leave in the next month.
• Player segmentation. RFM analysis, behavioural clustering — the foundation for CRM campaigns and setting VIP thresholds.
• Ad hoc investigations. "Why did NGR in Germany drop last week" — and suddenly the analyst is cross-checking exchange rates, the marketing budget, technical downtime, and a new limits law.

A separate and increasingly important part of the job is joint investigations with the anti-fraud and compliance teams: anomalous betting patterns and signs of multi-accounting or bonus abuse show up in the data first, and a fraud analyst works almost blind without the data analyst's exports.

The metrics you can't skip: GGR, NGR, RTP, LTV, retention

The first thing an analyst arriving in iGaming from e-commerce or SaaS learns: here, the numbers that matter aren't MRR and DAU but GGR and NGR, and mixing them up is like confusing revenue with profit.

What's specific to casinos is that GGR and NGR are extremely "noisy" metrics: one big jackpot win can crater a whole product's daily GGR, and the analyst's first job is to teach management not to panic over one bad day, but to watch the moving average.

Investors and the board are, as a rule, more interested in the NGR trend than GGR: GGR growth with bonus costs growing faster may mean the operator is simply buying revenue with aggressive promos rather than growing a healthy player base. Separating those two scenarios is a typical analyst task when preparing the quarterly report, and without a cohort view (how much NGR players acquired in January bring in three and six months later) it can't be done honestly.

The work cycle: from dashboard to experiment

There's almost no single "typical day" for an analyst at a casino — the mix of tasks depends heavily on whether they work solo for a small operator or in a team of a dozen analysts inside a large holding. But the weekly rhythm is recognisable at most companies:

Stack: SQL, Python and a BI layer

The tech stack isn't exotic — the complexity isn't in the tools but in the volume and specifics of the data: every spin on a slot, every bet in live casino, and every event in betting generates a row, and tables easily grow into the billions.

• SQL — mandatory, used every day. Complex joins across transaction and game-round tables, window functions for cohort analysis, query optimisation at scale.
• Python (pandas, numpy, scikit-learn or statsmodels). LTV and churn models, automating regular reports, statistics for A/B tests.
• BI tools. Tableau, Power BI or Looker/Looker Studio — self-serve dashboards for CRM, marketing and the VIP team.
• Data warehouse. Snowflake, BigQuery, ClickHouse or Redshift — given the event volume (every spin, every bet), the choice of engine has a big impact on how fast the analyst can work.
• dbt and orchestration (Airflow, Dagster). Not always required for junior/middle roles, but showing up more and more, and they open a path into analytics engineering.
• Statistics for experiments. Understanding why an ordinary t-test breaks down on data with rare, large wins, and how to calculate significance under those conditions.
• Excel/Google Sheets — hasn't gone anywhere. Quickly testing a hypothesis or a one-off report for a single stakeholder is often faster to put together in a spreadsheet than to build a whole separate dashboard for.

Gambling specifics add their own wrinkles on top of the standard stack: events come in not just from the website and app, but directly from game providers and payment gateways too, which means the analyst has to reconcile several sources of truth and figure out why a provider's report number doesn't match internal tracking.

A portfolio with a couple of SQL/Python case studies and a clear explanation of how NGR is calculated usually outweighs a data science degree — stack requirements in real vacancies vary a lot from company to company.

How much it pays by level

Python proficiency and experience building predictive models (LTV, churn) usually add 15–25% to the pay band compared with a pure SQL/BI profile at the same grade. See the full directory across 20+ industry roles at role directory.

How this differs from product analytics

An analyst moving into iGaming from classic product or SaaS usually goes through three relearning moments:

• A different set of core metrics. Instead of MRR, ARR and subscription churn rate — GGR, NGR, hold and RTP, calculated using completely different formulas and heavily dependent on bonus mechanics and licence-fee deductions.
• A regulatory layer wrapped around the data. Responsible Gambling and licensee requirements mean part of the modelling isn't about revenue growth at all — it's mandatory risk metrics: signs of problem behaviour, limits, self-exclusion.
• Extreme revenue volatility. A single jackpot worth a few hundred thousand euros can "break" a daily GGR report in a way no metric in ordinary e-commerce ever behaves — you need outlier-resistant methods and moving averages.
• Multi-jurisdictionality. The same product operates under different licences (MGA, UKGC, Curaçao) in different countries — GEO segmentation is mandatory in almost every report, not optional.
• A different set of stakeholders. Instead of the usual product and marketing, you get compliance, anti-fraud, and — at companies with sports betting — a trading team that lives by margin and line-movement metrics, not just retention.

What does transfer is exactly the same set: SQL, Python, A/B test statistics and the ability to explain to a non-technical manager what the number on the dashboard means. The domain — RTP, wagering, GGR/NGR — is picked up in a couple of months on the job. Betting and esports betting are a separate case: there, work with lines and odds is added to the list, and some analysts drift over time towards trading, where decisions are made in real time rather than off the back of a weekly report.

How to break into the profession

1. Portfolio beats a degree. A couple of case studies on public data (say, calculating retention or LTV on an open-source dataset) with clean SQL and Python close the question of formal education.
2. There's almost always a test task. A typical format is writing a query that calculates NGR by day and segment, or interpreting an artificially dipped retention chart.
3. The domain is learned on the job. The terms GGR, NGR, RTP, wager and hold get explained during a 2–4 week onboarding — you don't need to know the specifics in advance, but it's worth brushing up on SQL and statistics before the interview.
4. Adjacent professions — a fast way in. Experience in BI analytics or product analytics from e-commerce/fintech transfers almost without loss.

Where you go next

A data analyst at a casino isn't a dead-end position — it's a hub several tracks branch out from:

• BI / Analytics Engineer. If building predictive models interests you less than architecting data marts and self-serve dashboards, the logical move is toward BI analytics.
• Game math. A deep understanding of RTP and game volatility takes some analysts into game mathematicians, who design slot mechanics.
• CRM / Retention Lead. Churn models and segmentation are a direct path into retention management — more detail in the profile of CRM manager.
• Product analytics / Product Manager. Understanding the funnel and LTV opens the door to product management inside an operator.
• VIP analytics. Segmenting high-value players by LTV is closely tied to the work of VIP manager — it's the analyst's data that the decision to move someone into a personal VIP portfolio is built on.

Open vacancies with listed pay bands and stack are in the section live listings for data analysts. A general overview of industry salaries for 2026 is in the article "iGaming Salaries 2026".

FAQ

Do you need a data science degree? No. Employers look at SQL, Python and a portfolio of real case studies far more closely than at formal education.

Can you move over from product analytics in e-commerce or SaaS? Yes, it's one of the most common ways in — the technical skills transfer completely, and you pick up GGR/NGR/RTP specifics and the regulatory side in the first few months. For other ways to get into the industry with no experience, see the article "How to get into iGaming with no experience".

Does Python significantly raise the pay band compared with pure SQL? Usually by 15–25% at the same grade — especially if you have experience building LTV or churn-forecast models, not just descriptive analytics.

How does an analyst at a casino differ from one in betting? The foundation — SQL, Python, GGR/NGR metrics — is the same. In betting and esports betting, you add working with odds, bookmaker margin, and real-time data speed, which pulls the role closer to trading than it is in a classic casino.