Screenshot Roasts In Depth

How a screenshot roast works

You upload a screenshot of your landing page. PageScorch sends it to the AI, which analyzes the image as a visual artifact — the same way a conversion consultant would look at a page for the first time.

The AI is not reading your source code or fetching your URL. It is looking at what a visitor sees: the layout, visual hierarchy, headline prominence, CTA visibility, trust signals, and overall first impression.

The result is a conversion score, a list of specific problems, prioritized fixes, and AI-generated copy rewrites.

What gets analyzed

Above the fold — The section visible without scrolling is weighted heavily. If the hero section does not immediately communicate the offer and show a CTA, the score will reflect that regardless of how good the rest of the page is.

Headline and subheadline — Are they readable? Are they visually dominant? Do they communicate a specific benefit or outcome?

CTA button — Is there a primary CTA? Is it visible above the fold? Does its color and size make it stand out from the surrounding content?

Visual hierarchy — Does the layout guide the eye naturally? Are there competing elements that split attention away from the CTA?

Trust signals — Logos, testimonials, star ratings, review counts, security badges, and guarantees all count. Their absence is noted.

White space and readability — Cluttered pages score lower. Dense text without visual breaks makes the offer harder to parse.

Screenshot requirements

• Format: PNG, JPG, or WebP
• Minimum width: 800px recommended (narrower screenshots reduce analysis quality)
• Full page vs above-the-fold: Either works. Full-page screenshots give the AI more context about page structure. Above-the-fold screenshots focus the analysis on the hero section.
• Retina / 2x screenshots: Accepted. The AI handles high-density images correctly.

What screenshot roasts are best for

Use a screenshot roast when:

• You want to evaluate the visual experience — layout, design, hierarchy, CTA prominence
• You are auditing a page you do not own and cannot provide the URL
• The page is behind a login or paywall that URL fetching cannot reach
• You want to analyze a specific section (e.g. pricing table, hero section) in isolation
• You are reviewing a design mockup before the page goes live

What screenshot roasts cannot do

Because the AI reads the image only, it cannot analyze:

• Page load speed
• Mobile responsiveness (unless you upload a mobile screenshot)
• Hidden or below-fold copy that is not visible in the screenshot
• Interactive elements like modals or multi-step forms

For copy-heavy analysis, use a URL roast.

Tips for better results

Capture the full page. If your page is long, a full-page screenshot gives the AI more context. Many browser dev tools and extensions can capture full-page screenshots.

Use the desktop viewport. The AI is calibrated on desktop-width page layouts. If mobile conversions are your concern, upload a mobile screenshot and the score will reflect that viewport.

Avoid browser chrome. Crop out the browser toolbar and address bar before uploading. The AI focuses on the page content, not the browser UI.

Check the image before uploading. Make sure the screenshot is sharp and the text is readable. Blurry or low-contrast screenshots reduce analysis accuracy.

After your roast

Once the roast completes, you can:

• Read the full problem list and fix suggestions
• Generate an AI Fix Report to get rewritten headline, subheadline, and CTA copy
• Re-roast the page after making changes to measure improvement

See Reading your roast result for a walkthrough of every section of the result page.