How to Automate Google Ads Performance Reports

You can automate Google Ads reporting into a dashboard that updates itself every morning. No more logging into Google Ads, exporting CSVs, and pasting into spreadsheets. I set this up for every account I manage and it saves about 2 hours per week per account.

The Google Ads API is verbose, but once you wrap it properly, pulling data becomes a one-liner.

What You Need Before Starting

• A Google Ads developer token (apply through your MCC account)
• OAuth2 credentials from Google Cloud Console
• Python 3.8+ with google-ads SDK installed
• A Google Sheet or SQLite database for storage

Step 1: Install and Configure the SDK

pip install google-ads

Create a google-ads.yaml config file:

developer_token: YOUR_DEV_TOKEN
client_id: YOUR_CLIENT_ID
client_secret: YOUR_CLIENT_SECRET
refresh_token: YOUR_REFRESH_TOKEN
login_customer_id: "1234567890"

Step 2: Build the Data Fetcher

from google.ads.googleads.client import GoogleAdsClient
from datetime import datetime, timedelta

def fetch_google_ads_data(customer_id, date=None):
    if not date:
        date = (datetime.now() - timedelta(days=1)).strftime("%Y-%m-%d")
    
    client = GoogleAdsClient.load_from_storage("google-ads.yaml")
    ga_service = client.get_service("GoogleAdsService")
    
    query = f"""
        SELECT
            campaign.name,
            ad_group.name,
            metrics.impressions,
            metrics.clicks,
            metrics.cost_micros,
            metrics.conversions,
            metrics.cost_per_conversion,
            metrics.click_through_rate,
            metrics.average_cpc
        FROM ad_group
        WHERE segments.date = '{date}'
        ORDER BY metrics.cost_micros DESC
    """
    
    rows = []
    response = ga_service.search(customer_id=customer_id, query=query)
    
    for row in response:
        rows.append({
            "date": date,
            "campaign": row.campaign.name,
            "ad_group": row.ad_group.name,
            "impressions": row.metrics.impressions,
            "clicks": row.metrics.clicks,
            "spend": row.metrics.cost_micros / 1_000_000,
            "conversions": row.metrics.conversions,
            "cpa": row.metrics.cost_per_conversion / 1_000_000 if row.metrics.cost_per_conversion else 0,
            "ctr": row.metrics.click_through_rate * 100,
            "avg_cpc": row.metrics.average_cpc / 1_000_000,
        })
    
    return rows

Step 3: Store in SQLite

import sqlite3

def store_google_ads(db_path, rows):
    conn = sqlite3.connect(db_path)
    conn.execute("""CREATE TABLE IF NOT EXISTS google_ads_daily (
        id INTEGER PRIMARY KEY, date TEXT, campaign TEXT, ad_group TEXT,
        impressions INTEGER, clicks INTEGER, spend REAL, conversions REAL,
        cpa REAL, ctr REAL, avg_cpc REAL, fetched_at DATETIME DEFAULT CURRENT_TIMESTAMP
    )""")
    
    for row in rows:
        conn.execute("""INSERT INTO google_ads_daily 
            (date, campaign, ad_group, impressions, clicks, spend, conversions, cpa, ctr, avg_cpc)
            VALUES (?,?,?,?,?,?,?,?,?,?)""",
            (row["date"], row["campaign"], row["ad_group"], row["impressions"],
             row["clicks"], row["spend"], row["conversions"], row["cpa"],
             row["ctr"], row["avg_cpc"]))
    
    conn.commit()
    conn.close()

Step 4: Generate the Summary Report

def daily_summary(db_path, date=None):
    if not date:
        date = (datetime.now() - timedelta(days=1)).strftime("%Y-%m-%d")
    
    conn = sqlite3.connect(db_path)
    summary = conn.execute("""
        SELECT 
            SUM(spend) as total_spend,
            SUM(clicks) as total_clicks,
            SUM(impressions) as total_impressions,
            SUM(conversions) as total_conversions,
            CASE WHEN SUM(conversions) > 0 THEN SUM(spend) / SUM(conversions) ELSE 0 END as avg_cpa
        FROM google_ads_daily WHERE date = ?
    """, (date,)).fetchone()
    conn.close()
    
    return {
        "date": date,
        "spend": round(summary[0] or 0, 2),
        "clicks": summary[1] or 0,
        "impressions": summary[2] or 0,
        "conversions": round(summary[3] or 0, 1),
        "avg_cpa": round(summary[4] or 0, 2),
    }

Step 5: Schedule and Deliver

# Run at 7:30 AM daily, after data settles
30 7 * * * cd /path/to/project && python3 google_ads_report.py >> /var/log/google_ads.log 2>&1

The script fetches, stores, summarizes, and sends. One cron job handles the entire reporting pipeline.

What to Build Next

Push the summary to Slack or email as a formatted morning brief. Then combine this with your Meta data to create a single cross-platform daily report.

Related Reading

• AI-Powered Reporting That Actually Gets Read - making reports people actually care about
• The Pipeline Architecture - how to think about data pipelines
• Data Flow Architecture for Non-Engineers - simple data flow concepts