Website not loading images: check the address

An image is a second request, made after the page arrives, and it can fail on its own while the text loads fine. The address may differ from the file name by one capital letter, which many laptops ignore but a Linux server does not. The file may sit in a folder the release never uploaded. A photo on a private storage link may have expired, so the visitor gets a refusal. A secure page can also block a plain http image. The owner sees the pictures, because their own browser has the files cached and loads them from disk.

• Open one image address in a new tab. Does the picture open on its own?
• Compare the address with the file name on the server, letter by letter, including capitals and spaces.
• Check each image's status in the Network tab. A 404 means missing, and a 403 means refused or expired.
• Is the page secure while an image address starts with http? Look for a mixed content message in the Console.
• Scroll a phone page slowly. Does an image request appear only when its box reaches the screen?

Start with a single image. Copy its address from the Elements panel, then paste it into a new tab. A new tab sends no page context, so the result tells you about the file alone. If the picture opens, the file exists, and the fault is how the page refers to it: the spelling, the protocol or a rule about where the request comes from. If the picture does not open, the file is missing or refused, and no change to the page will fix it. Write down the status code for the same request in the Network tab, since it tells you which of those you have.

Each status code points at a different owner. A 404 means the server has no file at that address, which is usually a wrong path or a deleted upload. A 403 means the server found the file and refused it, often because a private storage link expired or the host refuses the request. A request that appears only after you scroll means the image is lazy-loaded and its box never reached the screen, perhaps inside a slider the visitor did not open. A request that stays pending for a long time points to a slow host or a file too large for a phone.

A page served over https that asks for an image over http can have the image blocked. The Console names it as mixed content, and the Network tab may show the request as blocked rather than failed. Change the address to https. A second cause is the format. Phone cameras can save HEIC files, and some browsers do not draw HEIC, so a photo that looks fine on the phone can arrive as an empty box. Convert product photos to JPEG or WebP before upload, so every modern browser can draw them.

A host can refuse an image request when it comes from a page on another domain. Your own site works, while a copy on a preview address or a partner's page shows broken boxes. The tell is the same as before: the address works in a new tab, but the embedded image fails on the page. Check the host's referrer or hotlink setting, and allow the domains that show your pictures. Then test the page on the preview domain as well as the live one.

A new upload setting, a renamed folder or a storage link that expires can blank the photos again in a later release, even after you fixed them once. Rerun the same simulated visitor on a product page after each change, on a laptop and a phone. If the shop sits behind a login, test with a test account, and check each preview deploy before the change goes live.