apify-examples

Background and the direct-to-Google alternative: pytrends is dead: 3 ways to get Google Trends data in 2026 (with code) on dev.to. Three small, runnable scripts that call the Data Fabrik Actors on the Apify platform and do something useful with the result: a CSV of Google Trends data, a flight price history with a drop alert, and a 7-night rate-shopping table for one hotel.

Script	Actor	What it does	Approximate cost per run
google_trends_to_csv.py	Google Trends Scraper & API	One call, one report per search line, flat rows written to trends.csv with pandas	$0.0025 per report on the free plan ($0.0010 on Gold); empty or failed searches are free
flight_price_tracker.py	Google Flights Scraper & API	Searches one route and date, appends the cheapest price to flight_history.csv, prints an alert when the price dropped by more than 10 % since the last run	$0.01 per search (all flights of the search included)
hotel_rate_shopping.js	Google Hotels Scraper & API	One hotel, 7 consecutive nights, every booking site's offer, printed as a table	about $0.05 on the free plan (7 × ($0.0025 per hotel with a price + $0.004 per hotel with offers))

All prices are those listed on the Actor pages at the time of writing (October 2026); the Actor page is the reference.

Prerequisites

• An Apify account (there is a free plan) and its API token: Apify Console → Settings → API & Integrations.
• Python 3.9+ for the two Python scripts: pip install "apify-client>=3" pandas (the scripts use the typed Run object of client 3.x)
• Node.js 18+ for the JavaScript script: npm install apify-client

The token

The scripts read the token from the APIFY_TOKEN environment variable. They never ask for it as an argument and it is never written to a file in this repository.

# macOS / Linux
export APIFY_TOKEN=...
# Windows (PowerShell)
$env:APIFY_TOKEN = "..."

If the variable is missing, every script stops with a clear message before calling anything.

Usage

# Google Trends: one report per argument; "a, b" compares up to 5 terms on the same scale
python google_trends_to_csv.py "bitcoin, ethereum" "solar panels" --geo US --range "today 12-m"

# Google Flights: one route, one date; run it daily (cron, Task Scheduler) to build the history
python flight_price_tracker.py CDG JFK 2027-03-15 --currency EUR
python flight_price_tracker.py --self-test        # checks the alert arithmetic, no API call

# Google Hotels: one hotel by name or Google Hotels link, 7 nights from the check-in date
node hotel_rate_shopping.js "Ritz Paris" 2027-03-01 EUR

What the Actors return

• Google Trends values are Google's relative interest (0–100 for the terms, place and period of the report), not absolute search volumes.
• Flight prices are the total for all passengers at the time of the run, for the point of sale and currency you choose.
• Hotel prices are for the whole room for the number of adults you set; hotels without a price for the dates are returned free, with empty price fields.

The field names used in the scripts (searchTerms, outputLayout, origins, departureDates, hotels, consecutiveDays, includeOffers, pricePerNight, cheapestOffer…) are documented on each Actor page.

License

MIT, see LICENSE.