Insights & Updates
Tips, guides, and news about website monitoring and visual change detection.
How to find a stable CSS selector when the page only has utility classes
The video element highlighted in DevTools on cohere.com/careers. The class attribute h-full w-full object-cover looks like a valid selector at first glance, but the values come from Tailwind utilities and will change at the next redesign. The <video> tag itself stays stable.
What the change percentage actually means in screenshot monitoring
Visual diff shows 22% changed, but the page looks the same. The percentage is always mathematically correct — but it only tells the truth when the page template stays stable between captures. Here's how to tell the difference and what to do about it.
Six types of website changes we actually caught with automated screenshots
Real examples of website changes caught by automated screenshot monitoring — from competitor redesigns and price drops to outages and new features. Each one with a before/after comparison.
How to monitor competitor website changes automatically
How to set up automated competitor website monitoring with screenshots and visual diff — which pages to track, how often to capture them, and how to filter out noise from A/B tests and dynamic content.
Full page vs viewport screenshots: which capture mode to use for monitoring
Full page screenshots capture everything below the fold. Viewport captures only the visible area. Learn which mode fits your monitoring workflow and how the choice affects visual diff.
Clip to element: monitor only the page section you care about
Full-page screenshots on e-commerce sites catch too much noise. Clip to element lets you capture only the block you care about — for cleaner diffs and fewer false alerts.
E-commerce price monitoring with screenshots: what we actually found
We set up screenshot-based price monitoring on Amazon across several product categories. Instead of clean price data, we got error pages, layout shifts, and a visual diff that catches things text scrapers miss entirely. Here's the full breakdown with real data.
How to bypass age verification on websites using custom cookies
Some websites block content behind an age verification form — three input fields for date of birth, no simple button to click. Click selector can't help here. We show how to bypass these gates using custom cookies, with jackdaniels.com as a real example.
When layout shifts break your price monitoring: a real Amazon tracking story
We set up price monitoring on an Amazon product page and immediately started getting 22% diff alerts — not because the price changed, but because a promotional banner kept appearing and disappearing in the navigation, shifting the entire layout down by a few dozen pixels. Here's how we figured out what was happening and what we'd recommend doing about it.
We were monitoring Amazon prices and kept finding error pages instead
We added a handful of Amazon pages to track product prices. Instead of price changes, we started getting alerts about error pages. Turns out scheduled screenshots double as availability monitoring, and they catch things Pingdom never will.
Wayback Machine vs Snapshot Archive: Which One Do You Actually Need?
The Wayback Machine saved over a trillion pages since 1996, but it can't control capture frequency, dismiss cookie banners, or notify you when something changes. We set up monitoring for hetzner.com through both tools and compared the results — 3 snapshots per day vs 24, HTML reconstructions vs real screenshots, and manual comparison vs automatic visual diff.
How to Hide Cookie Banners, Popups and Age Gates From Automated Screenshots
We tested click selector on three real sites — Hetzner (cookie banner), Fine Wine & Good Spirits and CWSpirits (age verification). One CSS selector in the settings, and every screenshot comes back clean.