📈 Google Trends Scraper An Apify Actor that retrieves data from Google Trends: realtime trending searches, interest over time, interest by region, and related queries/topics. ### 🚀 Quickstart (no input required) Want to get started in a flash? ⚡️ The actor has smart defaults, so you can run it immediately and get the latest realtime trending searches for the US. * In the Apify UI: Just click Run. --- ### 🔍 What you can fetch * Trending now (realtime): ⏰ Find out what's rapidly rising in searches in the last few hours. * Interest over time: 📊 Get a timeline of search interest for one or more keywords. * Interest by region: 🗺️ Discover where your keywords are searched the most. * Related queries/topics: 💡 Explore new queries and topics related to your main keyword. --- ### 💰 Pricing (rental) - $12/month - Recommended for ongoing tracking and monitoring workflows. - You can configure inputs freely; heavy configurations (very long timeframes, CITY/DMA resolution, inc_low_vol) may take longer. ### ⚙️ Input (overview) The actor's UI is dynamically generated from the .actor/INPUT_SCHEMA.json file. Here are the key fields you'll interact with: * scrape_type (select, default: trending_now) * keywords (array, optional for trending_now; required for other types) * gprop (select: web/images/news/shopping/youtube) * timeframe_type (select: predefined or custom, default: predefined) * predefined_timeframe (select, default: today 12-m) * custom_timeframe (string, e.g. "2023-01-01 2023-06-30") * geo_selection_type (select, default: Common Countries) * common_geo (string country code, default: US) * custom_geo_code (string for regions like US-CA) * geo_resolution (COUNTRY/REGION/DMA/CITY; for interest_by_region) * inc_low_vol (boolean; include low-volume regions; Pro recommended) * trending_language (string, default: en) * trending_hours (integer 1-191, default: 24) --- ### 📋 Example Inputs Here are some common use cases to get you started! 1. Realtime trending (US, 24 hours) json { "scrape_type": "trending_now", "common_geo": "US", "trending_hours": 24, "trending_language": "en" } 2. Interest over time for multiple keywords (US) json { "scrape_type": "interest_over_time", "keywords": ["data science", "machine learning"], "timeframe_type": "predefined", "predefined_timeframe": "today 12-m", "geo_selection_type": "Common Countries", "common_geo": "US" } 3. Related queries for a single keyword (global) json { "scrape_type": "related_queries", "keywords": ["python"], "timeframe_type": "predefined", "predefined_timeframe": "today 12-m" } --- ### 💾 Output & Output tab UI The actor writes exactly one item to the dataset per run, with fields: - scrape_type - data (the full results for the chosen type) - error (boolean) - error_message (string or null) - input_summary (geo, timeframe, resolution, gprop, etc.) Output tab UI - The default view is All results and shows the full data JSON right away (no need to switch tabs). - Extra views: - Trending now (table): A flattened table when scrape_type is trending_now - Summary: A quick summary of inputs and error status Notes on data structures - interest_over_time / interest_by_region: data is an array of objects (rows), similar to a table - related_queries / related_topics: data is an array containing keyed lists like { "top": [...], "rising": [...] } - trending_now: data is a list of trending keyword objects (keyword, geo, volume, topic_names, timestamps) --- ### ⚠️ Reliability & Troubleshooting Encountering issues? Don't worry, here are some common solutions: * Quotas and rate limits: 🚫 If you see 429/302 errors or quota issues, try reducing the frequency of your runs or using shorter timeframes/geos. You can also use a proxy to rotate your IP address. * Big results: 🐢 Scraping with geo_resolution: CITY or inc_low_vol can return large datasets and be slower. Start with COUNTRY or REGION for faster results. * Hours limit: ⌛ The trending_now scrape type supports a range of 1 to 191 hours. * Empty results: 🤷 Check that your keywords are popular enough in the specified timeframe and geo. --- ### 🧰 Runtime defaults & limits - Memory: 1024 MB (default) - Timeout: 30 minutes (default) - CPU: 1 (default) - Dataset writes: The actor always writes at least one item per run. If no data is found, it pushes an informative item with input summary. --- ### 📄 Notes * Alias support: You can use "trending_searches" as a supported alias for "trending_now".