Cyclosa web scraping service logoCyclosa
HOW IT WORKSSERVICESHOW IT WORKSPRICINGABOUTBLOG
LOG INGET STARTED
HOW-TO · 7 min read

How to scrape emails from a list of websites

You have a column of domains and you need a column of email addresses. Here is what that actually takes, and where most approaches fall short.

The task sounds simple. You have a spreadsheet of company websites and you want the email address for each one. In practice this is where a lot of outbound programmes stall, because the work is deceptively uneven.

Where email addresses actually live

There is no standard location for a company email address. Over a large enough list you will find them in every one of these places, and rarely in the same place twice:

  • A dedicated contact page, sometimes several pages deep in the navigation.
  • The site footer, often only on interior pages rather than the homepage.
  • A team or about page, attached to individual people rather than the company.
  • A careers page, with a different address for applications.
  • Inside a contact form page as fallback text, or in a privacy policy as the data controller contact.
  • Obfuscated deliberately: written as "name (at) company.com", or rendered as an image.

A scraper that checks only the homepage finds a minority of them. A scraper that checks only a page you nominate finds whatever happens to be on that page.

Why selector-based tools struggle

Most scraping tools ask you for a location: a URL pattern and a selector. That is a reasonable model when every site in your list is built the same way, which is essentially never true of a real prospect list.

The workaround people reach for is a regex across the raw HTML of the homepage. It catches some addresses and also catches a great deal of noise — image filenames, tracking pixels, third-party scripts, addresses belonging to the agency that built the site.

A process that holds up

Whatever tool you use, a list is only useful if you can trust it. Four things make the difference:

  • Go beyond the homepage. Contact data is usually one or two clicks in.
  • Keep the source URL for every address. Without it you cannot verify anything, and unverifiable data does not get used.
  • De-duplicate per company, not per page. One company found on three pages is one contact, not three.
  • Report failures explicitly. Knowing that 40 of 200 sites published no address is information; a silently shorter list is not.

How Cyclosa handles it

Paste or upload your list of URLs and select email as a field. The agent visits each site, identifies which of its pages are likely to carry contact information, opens them, and extracts the addresses it finds. Each result comes back with the page it came from and any name or role published alongside it.

Format validation and cross-job de-duplication run automatically, and any site that yielded nothing is reported as a skipped URL with a reason. A typical list returns in a few minutes.

On using the list responsibly

Collecting a publicly published address is one question; what you send to it is another. Cold outreach into the EU and UK is governed by GDPR and PECR, and in the US by CAN-SPAM. Relevance and a working unsubscribe are the baseline in every jurisdiction, and they also happen to be what makes outbound work.

Related pages

Email ScraperScrape emails from any list of websites.BlogMore guides on scraping and data.Lead Generation ScraperTurn domains into a prospect list.

Paste your first list of URLs.

Subscribe, add the sites you care about, and get contacts back in minutes.

GET STARTED