What is your goal?
Today I created my first workflow: the CoreClaw Google Maps email scraper. It can scrape email information from Google Maps by inputting keywords, cities, and the number of leads, and export it as a CSV file.
coreclaw：CoreClaw | Web Scraping Platform & Ready-Made Data Workers
What is the problem?
I hope everyone can check out and try my workflow, and provide suggestions for improvement.
What have you tried so far?
Problem encountered: During operation, large-scale data scraping can cause timeouts, and the timeout range has been increased.
Nice first workflow, and the Maps then website-crawl shape is the right one.
One thing that will bite you on a real list: matching emails on a page picks up
addresses that were never real. Stripe’s own site publishes
`[email protected]`, loads of small-business sites still carry
`[email protected]` from a template, and `[email protected]` matches most email
patterns so image filenames come through as contacts.
What helped me most, in order: reject placeholder domains (match the whole
domain, or you lose real companies like `acmetrading.nl`), drop anything ending
in an image extension, and keep `info@`/`sales@` separate from named personal
addresses since they behave quite differently when you send.
Worth grabbing `mailto:` links separately from page text too, they are far more
reliable. And plenty of companies publish no address at all, just a form.
Recording an honest “none found” beats a guess, because guesses bounce and
bounces cost you the domain.