Fifty pixels is the real test
On a phone carousel a cover can land at around fifty pixels tall. At that scale only the boldest shapes and largest words register. If the title dies there, it dies in the place most browsing now happens.
Almost nobody sees your cover the size you designed it, which is the whole point of a book cover thumbnail preview. Load your artwork and watch it shrink to the four sizes retailers actually serve, with greyscale and blur options to tell you whether the title still reads at 50 pixels.
Drop the front cover in and it renders instantly at each retail size, side by side for comparison.
Designers work at full size on a big bright monitor, where every flourish in a script font is legible and every texture in the background reads clearly. Shoppers never encounter the cover in that state. They meet it as a small tile among nineteen others, glimpsed for a fraction of a second while their thumb keeps moving.
Shrink the artwork and something brutal happens. Thin strokes vanish, low-contrast type dissolves into whatever sits behind it, and a subtitle that felt elegant becomes a grey smudge. Meanwhile the covers that dominate a category page are usually the ones with three enormous words and one confident shape, because those survive the shrink.
Running this check takes ten seconds and can save a launch. Load the artwork below and it is rendered locally by your own browser at four genuine retail sizes at once. Add greyscale to strip out the colour crutch, add blur to approximate a fast scroll, and be ruthless about what you can still read.
On a phone carousel a cover can land at around fifty pixels tall. At that scale only the boldest shapes and largest words register. If the title dies there, it dies in the place most browsing now happens.
Two colours can differ dramatically in hue yet sit at nearly identical brightness. Strip the colour away and that title suddenly disappears into its background, which is exactly what happens on e-ink readers and in printed catalogues.
Nobody stops and studies a grid of covers. The blur toggle approximates that half-second glance, and whatever shape or word still punches through the fuzz is the thing genuinely selling your book.
Unless you already have a following, your name is the least urgent element on the cover. At thumbnail size, spend the pixels on the title and one clear genre signal, and let the byline be small.
Our designers rework titles, hierarchy and contrast specifically for storefront scale, then deliver the files in every size a retailer asks for.
Retailer dimensions shift over time, but these heights are close to what shoppers meet in the wild on Amazon and comparable stores.
| Approx. height | Where it appears | What must stay legible | Common failure |
|---|---|---|---|
| 500 px | Product page main image, ad units | Everything, including subtitle and series line | Over-busy background competing with the type |
| 160 px | Search results grid, category listings | Title and one genre cue such as a figure or symbol | Script fonts with hairline strokes closing up |
| 80 px | Also-boughts, recommendation carousels, cart | Title only, and the dominant colour block | Low contrast between title and background |
| 50 px | Mobile carousels, sidebar strips, wishlist rows | A shape and a colour, at best two short words | Long titles set across four or five lines |
| Greyscale | E-ink devices, printed catalogues, ARC lists | Tonal separation between type and artwork | Two colours that share the same brightness |
| Blurred | Fast scrolling, peripheral vision | Silhouette and overall composition | No single dominant element to anchor the eye |
You do not create the thumbnail yourself. Retailers generate every size automatically from the single high-resolution file you upload, so your job is to supply artwork that still communicates once their software has shrunk it right down to a strip on a phone screen.
Never. The preview is produced by your own browser reading the file locally and drawing it onto canvas elements. The artwork stays on your machine throughout, which means you can safely test a cover months before you announce the book.
Usually one of three causes: strokes too thin to survive downscaling, insufficient brightness contrast against whatever sits behind the type, or a title broken across too many lines. Thicker weights, a solid panel behind the words, and fewer lines all fix it.
It removes hue and leaves only lightness, which is the fastest way to expose a contrast problem. Colours that look boldly different can carry almost identical brightness values, and once that colour is stripped the text and background merge into one flat tone.
One to four words is the comfortable range for a cover that has to work small. Longer titles are not forbidden, but they need a clear hierarchy where two or three key words are set large and the rest recedes, otherwise everything shrinks to illegibility together.
No, load the front panel by itself. Storefronts only ever display the front, so testing the whole wrap would shrink your front cover by more than half again and give you a misleading picture of how the listing will look.
A height to width ratio of about 1.5 to 1 suits most storefronts, which is where the familiar 1600 by 2560 pixel recommendation comes from. Squarer artwork tends to look undersized in a grid, while very tall artwork gets padded with empty space on either side.
Increase the title size dramatically, often to double what feels comfortable at full view, then simplify the background so a single shape dominates. Those two moves rescue more struggling covers than any amount of adjustment to fonts or colour palettes.
Once the design works, the file still has to meet each store's dimensions, colour space and compression rules. We handle that step so your listing image looks as sharp on Amazon as it does on your desktop.
See eBook Cover Design