The visual selector lets you point and click an element on a preview of the page, and PageCrawl turns it into a selector for you. Occasionally a page or a selector needs a little extra care. This guide covers the most common situations and how to resolve them.

Problem: The page won't load in the picker

Some pages are slow, heavily scripted, or actively block automated browsers, so the live preview may fail to load or render incompletely.

Solutions:

• Switch the engine. Try Stealth mode for sites that block bots, or Fast mode for simple static pages. See Real Browser Mode.
• Paste a selector instead. You don't need the picker to load the page. Find the selector in your own browser (how to find a selector) and paste it straight into the element's selector field, then use Test to confirm it captures the right content.
• Report it. If a page consistently fails, contact support so we can improve compatibility.

Problem: The selector breaks when the website changes

Some sites generate randomized class names or add suffixes that change on every deploy, which makes a selector go stale.

Solution: Match on the stable part of the class instead of the full name. For example, a class like productTile_urgencyMessaging__V5DTS has a volatile __V5DTS suffix. Use an XPath contains() match on the stable prefix:

//*[contains(@class, 'productTile_urgencyMessaging')]

See the XPath tutorial for common selectors for more patterns like this.

Let PageCrawl or an AI assistant write the selector

If you're unsure which selector to use, you have a few easy options:

• Use the visual selector to point and click the element, and let PageCrawl generate the selector.
• Use the PageCrawl browser extension to pick an element on the live page and send it into a new monitor.
• Paste the page's HTML or URL into an AI assistant like ChatGPT or Claude and ask it to "write a CSS or XPath selector for [the element]".

Whichever route you take, always press Test to confirm the selector captures exactly what you expect before saving.