Fortnite Discover Thumbnail Tips (For a 6-Tile Row)

What Discover actually shows a player

Before you touch a single pixel, it helps to be precise about what a player is actually looking at when they scroll past your island. It's not your thumbnail in isolation — it's three signals read together in under a second: the category row your island happens to be sitting in, the title next to your art, and the thumbnail itself. Under each tile, Discover also shows a live player count, updating in real time as people join and leave.

That count sits in the same glance as everything else on the tile. A player scanning a row takes in the category, the title, the thumbnail, and that live number together, in under a second, not as separate steps. You don't control the number, but you should know it's sitting right next to your thumbnail every time a player evaluates it.

Six tiles wide changes everything

Discover doesn't give your thumbnail a full-width hero slot. It renders in a row that's six tiles wide, which means whatever you designed at full resolution gets shrunk down to a fraction of the space it occupied in your editor. At that size, fine detail is the first thing to go — texture, small text, subtle shading, anything that depends on the viewer being close enough to notice it. What survives a six-across row is silhouette and one high-contrast subject: a shape and a focal point a player can register at a glance, without needing to lean in.

This is the gap that catches almost everyone. You export at 1920×1080, open the file full-screen in your editor, and it looks great — sharp, detailed, exactly what you intended. That full-screen view is not what a player ever sees. They see one of six tiles in a row, at a size closer to a thumbnail than a poster, next to five other islands doing the same thing. If your composition only reads at full-screen size, it's not actually finished — it's finished for the wrong context.

One subject, read at a glance

The practical fix is ruthless simplification. Pick one subject, give it the clearest silhouette you can manage, and cut everything that competes with it for attention. A busy background, a second focal point, or a cluttered arrangement of smaller elements might all look intentional and well-composed at full size, but in a six-tile row they just blur into noise. The question worth asking isn't "does this look good" — it's "can I tell what this is from across the room."

If you're not sure whether your composition passes that test, try squinting at it, or stepping back from your screen until the details blur. What's left after that — the shape, the color block, the one thing that still reads — is roughly what a player gets in the row. If nothing recognizable survives that test, no amount of extra detail is going to save the thumbnail; it needs a different subject or a simpler arrangement, not a finer render.

Your title is doing half the work

Remember that a thumbnail never appears alone — it's always paired with your title, and players read the two together in the same glance. That means your art doesn't have to carry the entire pitch by itself. If your title already tells a player what kind of experience they're looking at, your thumbnail can spend its limited space on mood, subject, and appeal instead of trying to spell out the concept visually too.

The reverse is also true: if your title is generic or hard to parse quickly, your thumbnail has to work harder to compensate, and it usually can't. Look at your title and thumbnail side by side, the way they'll actually appear in the row, and ask whether they're pulling in the same direction or fighting each other for the same job.

Test against neighbours, not against a blank canvas

The single biggest gap between how developers judge their own thumbnails and how players actually see them is context. You look at your art alone, zoomed in, with nothing else on screen. A player sees it as one of six tiles, surrounded by other islands with their own colors, subjects, and compositions. A thumbnail that looks strong in isolation can disappear completely once it's sitting next to five competitors in similar colors or similar genres.

Judging your thumbnail against a blank canvas answers the wrong question. The question that matters is whether it still stands out once it's surrounded by real neighbours — which is exactly what a six-tile row forces on every island, whether you've tested for it or not.

When to use Epic's A/B test instead

None of this replaces Epic's own testing tools — it's meant to happen before you reach for them. Once an island is live, Epic runs a post-publish thumbnail test: two thumbnails, split 50/50, running for up to 90 days, measured against real click data from real players. For a genuinely close call between two strong candidates, that test beats any preview tool, full stop — it's measuring actual player behavior over real time, not a simulated glance.

The case for previewing first isn't that it replaces that test. It's that a live testing slot costs days you don't get back, and most thumbnails that would fail the A/B test are obvious rejects in seconds once you see them next to real neighbours — a busy background that vanishes, a subject that reads as a blob at tile size, a title that clashes with the art. Catching those failures before you spend part of a 90-day window on them means the A/B test you do run is a genuine contest between two thumbnails that both deserved a shot, not a coin flip between one good option and one you already suspected was weak.

The five-minute pre-publish loop

In practice, the workflow is short. Open the Fortnite thumbnail previewer, drop in each candidate thumbnail one at a time, and reshuffle the neighbouring tiles so you're not just judging your art against the same fixed row. Do that for every candidate you're considering, and keep whichever two actually survive sitting in a six-tile row next to real competition — not the two you liked best zoomed in on your own screen.

Once you've got those two, that's the pair worth spending a real testing slot on. Ship them into Epic's A/B test and let real click data over real time settle the final call. Previewing gets you to a genuine two-way contest in minutes; Epic's test is what actually decides it.