The 390-pixel test: why sites fail on phones
Sites get approved on a 27-inch monitor in an office and get used on a 6-inch phone in a parking lot. Most of the local-business web fails that translation, and the owner is the last to find out because nobody who bounced ever tells them.
The short answer
Open Chrome, press F12, toggle the device toolbar, set the width to 390 pixels, and load your site. Now try to call yourself: find the phone number and tap it, starting from the top of the page, in under ten seconds. Then check the other classics: text sitting on a photo with no dark layer behind it, buttons too small for a thumb, a desktop layout shrunk instead of rebuilt. Well over half of local traffic is mobile, so whatever you just saw is your storefront, not the desktop version.
Why owners never see it
Three reasons. The site was signed off on a desktop, where everything fits. The owner's own phone has visited a hundred times, so it loads instantly from cache and their thumb knows where everything is. And the customers who hit the broken version do not file complaints, they tap back and call the next roofer in the list. The failure is silent by construction.
The faults we actually measure
We render sites at 390 pixels and test them the way a customer uses them. The same faults account for most of the failures:
- Copy on photos with no scrim. White text laid straight onto a bright photo is unreadable in sunlight. The fix is a contrast layer behind the text, or moving the copy beside the image.
- Tap targets under 44 pixels. Links designed for a mouse pointer, hit with a thumb. Misses read as "the site is broken".
- The squeezed desktop layout. Three columns crushed into one screen width, type at 60 percent scale, horizontal scrolling. A phone layout is a different layout, not a smaller one.
- Numbers that are not links. A phone number in an image, or plain text without a tel link, forces the customer to memorise and retype it. Some will. Most will not.
- Forms built like intake paperwork. Ten required fields on a phone keyboard filters out everyone except the very desperate.
What failing costs, in order
First the direct loss: the mobile visitor who cannot act leaves, and in urgent categories they buy from whoever answered instead. Second the amplification: if you run ads, you are paying per click to send people to a page that turns them away. Third the ranking: Google indexes and scores the mobile version, so the desktop site nobody sees is also the one Google is not ranking you on.
If your site failed the test
The honest next step depends on how it failed. One or two of the faults above are an afternoon of fixes for whoever built the site. If the whole layout is a squeezed desktop page, it needs rebuilding around the phone, which is what the redesign offer covers, measured faults first, so you see the evidence before anyone proposes anything.
Fair questions
How do I check what my site looks like on a phone?
Do not trust your own phone, which has your cache and your familiarity. In Chrome press F12, toggle the device toolbar, and set the width to 390 pixels, a normal modern phone. Then try to do the one thing a customer would: find the number, read the prices, tap the button.
What actually makes a site fail on mobile?
The repeat offenders: text on top of a photo with no contrast layer behind it, tap targets smaller than a fingertip, the desktop layout squeezed instead of rebuilt, forms that demand ten fields, and phone numbers that are pictures instead of tappable links.
Does mobile failure affect my Google ranking?
Yes. Google evaluates the mobile version of your site, not the desktop one, and page experience feeds ranking. But the bigger cost is direct: the customer who cannot find your number in ten seconds calls the next result.