Data Analyst

Window Functions & CTEs

Why: You can compute a running total or each customer's previous order in one readable query, instead of self-joins nobody else can maintain.

Subqueries & Set Operations

Why: You can find customers who bought in January but not in February, and choose between EXISTS and a join based on correctness and speed.

Hypothesis Testing

Why: You can say whether a conversion difference between two regions is real or noise, and state it with an interval rather than a hunch.

Correlation & Regression Basics

Why: You can show that ad spend tracks sales, quantify it with a regression, and warn the team that this alone does not prove the ads caused the lift.

Dashboard Design Principles

Why: You can redesign a cluttered dashboard so the one number an operations lead checks each morning sits on top, with filters that do not mislead.

Storytelling with Data

Why: You can turn a churn analysis into a short narrative with annotated charts, so a director understands what happened and what to do before reading the details.

Python for Analysis (Pandas/NumPy)

Why: You can replace a manual spreadsheet routine with a pandas script that merges, groups and pivots the same files reliably every week.

Data Cleaning & Transformation

Why: You can catch duplicated customers, mixed date formats and null values before they quietly inflate a revenue figure that leadership reads.

KPI Definition & Tracking

Why: You can help a team replace a vanity metric like page views with a KPI tied to a target and a decision, then track it on a shared dashboard.

Stakeholder Communication

Why: You can turn a vague request like why sales are down into a scoped analysis, and handle a manager who disputes the numbers without getting defensive.

Geospatial Analytics

Why: You can map customer addresses to regions and compute distances to stores, so a decision about a new branch rests on where people actually live.