Why Text on Photos Ends Up Unreadable

Text on a photo is easy to add and easy to make unreadable. The difference is almost always contrast.

Illustration of text being placed over a photograph on a laptop screen

Thumbnails, quote cards, announcements, memes, simple ads — text on an image is one of the most common small design jobs there is. Add Text to Image handles it in the browser.

Adding text is the easy part. Making it readable is where most attempts fall down.

Contrast is the whole game

Text fails on photos for one reason above all others: the background is busy. White text over a bright sky vanishes. Dark text over a shadow vanishes. Text over a detailed, high-contrast area is legible in some spots and not others, which is worse than either.

This tool gives you two levers, and they're the two that matter most:

  • Put the text somewhere calm. Photos usually have a quiet region — open sky, a blurred background, a plain wall. Free real estate, and the single highest-leverage choice you make. You get seven positions: the four corners, top centre, bottom centre, and dead centre.
  • Pick a colour that fights the background. White is the default and it's wrong about half the time. Dark text on a bright photo is often instantly more legible than any amount of fiddling elsewhere.

Two other techniques you'll see everywhere are worth knowing about, but neither is possible here — they need a full image editor:

  • A semi-transparent band or gradient behind the text. This is what almost every professional thumbnail does, and it guarantees contrast regardless of what's underneath.
  • An outline or drop shadow on the letters. Blunt but effective over busy backgrounds.

If your photo genuinely has no calm region and no colour reads cleanly, that's the honest signal to reach for a different tool rather than shipping something unreadable.

Illustration of a person stepping back to view a large poster pinned on a workshop wall

Size it for where it'll be seen

The classic error is judging text size on a full-screen preview when the image will be viewed as a small thumbnail. If it's going in a feed or a video grid, it'll be shown at a fraction of the size you're editing at.

Zoom out until the image is roughly its real display size and check. If you can't read it instantly at a glance, it's too small — and fewer words at a larger size beats a sentence nobody reads.

Crop first, text last

Set the frame before adding text. Crop afterwards and you'll clip your own words; resize afterwards and the text shrinks with the image. So: crop, resize to final dimensions, then add text.

For a YouTube thumbnail specifically, Resize for YouTube Thumbnail gets the dimensions right first — see our thumbnail guide for the rest.

Save it as PNG

Text has hard edges, and hard edges are exactly what JPG compression handles worst — it produces fuzzy artefacts around lettering. PNG keeps type crisp. If the file is then too heavy, Compress PNG helps, and flat-colour text over a simple background compresses well.

If you're also branding the image, Add Watermark is the better tool for a persistent mark than a text layer.

FAQ

Why is my text hard to read?
Almost always insufficient contrast with a busy background. Move it to a calmer part of the frame and try a colour that fights the background — dark text on a bright photo, not white. Putting a semi-transparent band behind the text works too, but that needs a full image editor rather than this tool.

What size should the text be?
Judge it at the size people will actually see the image, not at full zoom. Fewer, bigger words nearly always win.

Should I save as JPG or PNG?
PNG. JPG compression makes sharp edges — like letterforms — fuzzy.

← Back to all guides