How to Do a Blur Test: Step-by-Step Guide for Designers

The blur test is the fastest way to find out whether your design actually communicates. Blur the page until the text is unreadable — whatever still stands out is what visitors see first. If your headline and call-to-action survive the blur, your visual hierarchy works. If they vanish into gray mush, so does your conversion rate.

Designers have used this trick (and its low-tech cousin, the squint test) for decades. This guide covers why it works neurologically, three ways to run it — from squinting to full automation — and exactly what to look for on websites, ads, emails, app screens and thumbnails.

Why the Blur Test Works: Your Brain Sees Shapes Before Words

In the first moments on any page, the brain doesn't read — it runs pre-attentive processing: an automatic scan of size, color, contrast and position that happens in under half a second, before conscious attention kicks in. Whatever wins that scan gets looked at first; everything else waits its turn or never gets seen.

Blurring a design strips away the details (words, textures, faces) and leaves exactly the signals pre-attentive processing uses. That's why the blurred version of your page is an honest preview of a stranger's first glance — and why the test catches problems that staring at the sharp version never will. You built the page; you can't un-know where the button is. The blur can.

Blur Test vs Squint Test: Same Idea, Different Rigor

The squint test is the informal version: squint at your screen until things go fuzzy and notice what still reads. It costs nothing and takes ten seconds — but it's subjective. Your squint isn't reproducible, and you're a biased observer of your own work.

The blur test applies a real Gaussian blur to a screenshot, so everyone evaluates the same image. Same principle, more rigor — and it can be automated.

Method 1 — The Squint Test (10 seconds)

• Open the page at normal size, lean back to arm's length.
• Squint until text becomes unreadable and shapes go soft.
• Ask: what do I notice first? Can I still find the action button?

Use it as a habit while designing — a quick sanity check after every significant change. Just don't treat it as evidence: you know the page too well to be fooled by your own squint.

Method 2 — Manual Blur in Figma or Photoshop (5–10 minutes)

• Screenshot the area you're testing — usually the above-the-fold section at real viewport size (1440px desktop, 390px mobile).
• Apply Gaussian blur. Figma: select the image → Effects → Layer blur, 5–8px. Photoshop: Filter → Blur → Gaussian Blur, radius 5–8px. The text should be illegible but shapes clearly visible.
• Evaluate at a glance — first instinct, not a study: What dominates? Is the CTA findable in under a second? How many elements compete?
• Repeat on mobile. Hierarchy that works on a wide screen often collapses in a single narrow column.

Pro tip: do the evaluation with someone who hasn't seen the page. You provide the blur; they provide the honest eyes.

Method 3 — Automated Blur Test (30 seconds)

Manual testing has two limits: it takes time, and the interpretation is still yours. BlurTest automates the whole loop — upload a screenshot or enter a URL, and the AI applies the blur, simulates first-glance attention, and returns a clarity score, an attention heatmap and a predicted eye path, plus specific fixes. Because the process is identical every run, you can test variants fairly and track whether a redesign actually improved the score. No account needed to try it — the free design roast runs the same engine with a sense of humor.

What to Look For, by Content Type

• Websites & landing pages: headline and primary CTA must survive; decorative hero images shouldn't win. One dominant element per screen.
• Banners & display ads: at banner sizes the blur is brutal — your offer and brand must both read as shapes. If the logo disappears, the impression is wasted.
• Emails: blurred, a good email shows one headline block and one button. Three products and three buttons blur into uniform gray — and get deleted.
• App screens: the primary action of the screen should be the obvious survivor; if navigation chrome dominates, the screen's job is unclear.
• Social posts: feeds scroll at blur speed already. If the post doesn't communicate blurred, it won't stop a thumb.
• YouTube thumbnails: thumbnails are viewed tiny — effectively pre-blurred. Faces, one bold subject and 3-4 words max survive; detailed scenes don't.

The 4 Most Common Blur Test Findings (and Fixes)

• The CTA disappears. Low contrast, small size, or crowded placement. Fix: raise the button's contrast against its actual background and give it whitespace.
• A decorative image dominates. The photo is your loudest element but carries no message. Fix: shrink it, mute it, or make it point at the headline/CTA.
• Everything shouts equally. Uniform blocks of same-size, same-weight content. Fix: pick one winner per screen and demote the rest — see what should dominate.
• The headline loses to a badge or logo. Brand elements outweigh the value proposition. Fix: your offer, not your logo, should be the largest text on the page.

What the Blur Test Can't Tell You

The blur test evaluates the first glance — hierarchy, not content. It can't judge whether your headline is persuasive, your offer competitive, or your page trustworthy. Treat it as step one: fix the hierarchy first so people see your message, then improve the message itself with a full 10-minute audit and real A/B tests.

Blur Test FAQ

How much blur should I apply?

Enough that text is unreadable but shapes are clear — usually 5–8px Gaussian blur on a normal-resolution screenshot. If you can still read the headline, add more; if you can't tell a button from a paragraph, use less.

Is the blur test scientific?

It's a practical proxy for well-documented perception research (pre-attentive processing, peripheral vision), not a lab study. It reliably reveals what dominates a design; pair it with analytics and testing for the full picture.

How often should I run it?

Every time a page or creative changes meaningfully, and before you spend money driving traffic to it. It's 30 seconds that regularly saves a failed campaign.

Ready to see your own page the way strangers do? Run a free blur test with BlurTest — upload or paste a URL, get your score in 30 seconds.