I have a raw export of our online store’s transactions and I need it turned into a reliable, insight-ready asset. The end goal is to surface actionable business insights, with a special emphasis on understanding customer purchasing patterns and, in particular, the average order value across segments, seasons and promotions. You’ll start by cleaning and normalising the CSV data, handling duplicates, missing values and inconsistent formats with Python, Pandas and NumPy. Once the dataset is spotless, I’d like a concise exploratory analysis that highlights any anomalies that could skew decision-making. From there, drill into the purchasing-pattern angle: determine how average order value varies by customer cohort (new vs. repeat), time of day, day of week and marketing channel. Use Matplotlib or Seaborn to visualise these findings in clear, board-ready charts. Deliverables: • A well-commented .py script (or Jupyter notebook) that performs the cleaning, EDA and visualisation steps end-to-end • A cleaned dataset saved to a separate file for future runs • A short slide deck or PDF that narrates the key insights and includes the visuals I’ll run your code against fresh exports each month, so please build with reproducibility in mind and keep external dependencies minimal. If you have questions about column meanings or want sample rows before you start, let me know and I’ll provide them right away.