How to rank a new app with no downloads

We measured 7,789 top-ten App Store placements on a single day. 18.7% of them are held by apps with fewer than 200 ratings. The space exists. What decides whether you can take it is not how long your keyword is — that barely matters — but how contested the result set already is.

The space is real, and we can size it

The usual advice to a developer launching with no installs is some version of “go long-tail”, offered without a number attached. We can attach one. On 11 September 2026, a complete collection day, we recorded 7,789 top-ten iOS placements where the holding app’s rating count was a measured value rather than a blank. Here is how those placements split by the size of the app holding them.

Half the visible top ten belongs to apps with five-figure rating counts, which is the part nobody disputes. The part worth noticing is the other end: nearly one placement in five is held by an app with fewer than 200 ratings. Those apps are not winning on install volume, because they do not have any.

Long-tail is the wrong lever

The standard prescription is to chase longer phrases. We split the same placements by how many words are in the search term, and the effect is close to nothing.

Four-word phrases were, if anything, slightly worse for a small app than two-word ones. That is not what “go long-tail” predicts. The explanation is straightforward once you look at the result sets: a long phrase is not automatically an uncontested one. “best budget planner app free” is five words and every large finance app in the store matches it.

Contestedness is the right lever

When we split the same placements by how our difficulty score rates the term instead of by word count, the pattern is immediate and monotonic.

On the softest terms, small apps hold nearly two thirds of the visible top ten. On the hardest, one slot in seven. That is the actual decision a new app is making, and word count is a poor proxy for it.

What to do in your first ninety days

The honest limits of this

Rating count is a proxy for install base, not a measurement of it: apps prompt for ratings at very different rates, and a quiet app with many installs can look small here. The measurement is one day, iOS, and the 3,564 terms our collector follows, which lean toward the categories our customers work in rather than a random sample of the store. And a top-ten placement is a placement, not a download: nothing here says those small apps are succeeding commercially, only that they are visible.

What survives those caveats is the shape. Visibility is not allocated purely by size, the exceptions are common rather than freakish, and they cluster in the parts of the store where fewer strong apps are already competing. For an app with no downloads, that is the whole strategy: find the sets where the behavioural signal you lack is not what decides the order.

Blog · Free ASO tools · Products · AsoTheory