【2025年版】GA4×Pythonで高度なセグメント分析と自動レポート生成｜Colab対応全コード公開

# ▼ 必須ライブラリをインストール（Colab初回のみ）
!pip install --upgrade google-analytics-data
!pip install --upgrade gspread oauth2client

# ▼ Driveマウント
from google.colab import drive
import os

drive.mount('/content/drive')

# ▼ スクリプト本体
from google.analytics.data_v1beta import BetaAnalyticsDataClient
from google.analytics.data_v1beta.types import RunReportRequest, DateRange, Dimension, Metric, FilterExpression, Filter
from google.oauth2 import service_account
import gspread
from datetime import datetime, timedelta

# ▼ 設定項目
PROPERTY_ID = '123456789'  # GA4のプロパティID
SCOPES = ['https://www.googleapis.com/auth/analytics.readonly', 'https://www.googleapis.com/auth/spreadsheets']
SERVICE_ACCOUNT_FILE = '/content/drive/MyDrive/your-folder/your-service-account.json'  # Drive上のパスに変更
SPREADSHEET_ID = 'YOUR_SPREADSHEET_ID'
SHEET_NAME = 'GA4データ出力'

# ▼ 前月の1日と末日を取得（1月のバグ対策あり）
today = datetime.today()
if today.month == 1:
    first_day = datetime(today.year - 1, 12, 1)
else:
    first_day = datetime(today.year, today.month - 1, 1)
last_day = datetime(today.year, today.month, 1) - timedelta(days=1)
date_range = DateRange(start_date=first_day.strftime('%Y-%m-%d'), end_date=last_day.strftime('%Y-%m-%d'))

# ▼ APIリクエスト定義
request = RunReportRequest(
    property=f"properties/{PROPERTY_ID}",
    date_ranges=[date_range],
    dimensions=[
        Dimension(name="landingPagePlusQueryString"),
        Dimension(name="sessionDefaultChannelGroup")
    ],
    metrics=[
        Metric(name="sessions"),
        Metric(name="engagementRate")
    ],
    dimension_filter=FilterExpression(
        and_group={
            "expressions": [
                FilterExpression(
                    filter=Filter(
                        field_name="sessionDefaultChannelGroup",
                        string_filter={"value": "Organic Search"}
                    )
                ),
                FilterExpression(
                    filter=Filter(
                        field_name="landingPagePlusQueryString",
                        string_filter={"value": "/movie/", "match_type": "CONTAINS"}
                    )
                )
            ]
        }
    ),
    metric_filter=FilterExpression(
        filter=Filter(
            field_name="sessions",
            numeric_filter={"operation": "GREATER_THAN_OR_EQUAL", "value": {"int64_value": 10}}
        )
    )
)

# ▼ 認証とデータ取得
credentials = service_account.Credentials.from_service_account_file(
    SERVICE_ACCOUNT_FILE, scopes=SCOPES)
client = BetaAnalyticsDataClient(credentials=credentials)
response = client.run_report(request)

# ▼ gspreadでスプレッドシートへ出力
gc = gspread.service_account(filename=SERVICE_ACCOUNT_FILE)
sh = gc.open_by_key(SPREADSHEET_ID)

# シートが存在しない場合は作成
try:
    worksheet = sh.worksheet(SHEET_NAME)
except gspread.exceptions.WorksheetNotFound:
    worksheet = sh.add_worksheet(title=SHEET_NAME, rows="1000", cols="20")

worksheet.clear()
headers = [header.name for header in list(response.dimension_headers) + list(response.metric_headers)]
worksheet.append_row(headers)

for row in response.rows:
    dims = [d.value for d in row.dimension_values]
    metrics = [m.value for m in row.metric_values]
    worksheet.append_row(dims + metrics)