Website Traffic Analysis Project with Python

"""Analyse synthetic website sessions by date and acquisition channel."""

from __future__ import annotations

import numpy as np
import pandas as pd


def make_demo_sessions(n: int = 8_000, seed: int = 42) -> pd.DataFrame:
    rng = np.random.default_rng(seed)
    channel = rng.choice(["Organic", "Paid Search", "Social", "Email", "Direct"], n, p=[0.34, 0.22, 0.16, 0.1, 0.18])
    pageviews = np.maximum(1, rng.poisson(2.8, n))
    converted = rng.random(n) < np.select([channel == "Email", channel == "Paid Search"], [0.065, 0.048], default=0.031)
    return pd.DataFrame({"session_id": range(1, n + 1), "date": pd.Timestamp("2025-01-01") + pd.to_timedelta(rng.integers(0, 365, n), unit="D"),
                         "channel": channel, "pageviews": pageviews, "duration_seconds": rng.gamma(2.2, 75, n).round(),
                         "bounced": pageviews == 1, "converted": converted})


def analyse_traffic(data: pd.DataFrame) -> tuple[pd.DataFrame, pd.DataFrame]:
    sessions = data.copy()
    by_channel = sessions.groupby("channel", as_index=False).agg(sessions=("session_id", "nunique"), pageviews=("pageviews", "sum"), bounce_rate=("bounced", "mean"), conversion_rate=("converted", "mean"), average_duration=("duration_seconds", "mean"))
    sessions["month"] = pd.to_datetime(sessions["date"]).dt.to_period("M").astype(str)
    monthly = sessions.groupby("month", as_index=False).agg(sessions=("session_id", "nunique"), conversions=("converted", "sum"))
    monthly["conversion_rate"] = monthly["conversions"] / monthly["sessions"]
    return by_channel.sort_values("sessions", ascending=False), monthly


def main() -> None:
    channel, monthly = analyse_traffic(make_demo_sessions())
    print(channel.round(3).to_string(index=False))
    print("\nMonthly:\n", monthly.round(3).to_string(index=False))
    print("Synthetic sessions; metric definitions must match the real analytics implementation.")


if __name__ == "__main__":
    main()