How many downloads does it take to enter the top 10?
23 August 2026 · Zoubair Haja
Review count is a listing's most visible signal, but it is not the one Google looks at first. Installs, retention and repeat opens weigh more — except they are invisible from outside. Only one of them is published: the download count, shown in brackets on every listing.
We collected it across the top 10 of 30 keywords on the French Play Store, on 23 August 2026, looking for the same figure as with reviews: that of the LEAST installed app in the ranking — the actual entry ticket.
The entry ticket
installs: the median entry ticket of a top 10, across the 30 keywords measured.
A hundred thousand installs for the smallest app in the ranking. The number looks crushing — but as with reviews, the median of the top 10 itself is 5,000,000 installs, fifty times more. It is the same trap: judging a keyword by the average level of its occupants makes you give up where there was room.
And the spread is again considerable. 9 of the 30 keywords have an entry ticket under 10,000 installs, and 3 under 1,000. At the other end, 13 demand more than a million. On “enregistreur vocal” (voice recorder), the least installed app in the top 10 shows the lowest bracket Google publishes — meaning an app with almost no installs holds a place in that ranking.
By category
| Category | Entry ticket | Median top 10 |
|---|---|---|
| Education | 5,000+ | 1,000,000+ |
| Finance | 10,000+ | 3,000,000+ |
| Tools | 100,000+ | 30,000,000+ |
| Productivity | 1,000,000+ | 10,000,000+ |
| Health & fitness | 1,000,000+ | 5,000,000+ |
| Photography | 10,000,000+ | 100,000,000+ |
Two thousand times between Education and Photography. An education app can enter a top 10 with five thousand installs; a photo app needs ten million. That is not a quality gap, it is a history gap: photography has drawn very high-volume apps for a decade.
Reviews or downloads: which one should you look at?
The two measures tell the same story — same easy categories, same locked ones — but they are not equally useful as instruments.
Review counts are exact: Google publishes “6,471”, not “about 6,000”. Installs are shown in brackets — “100,000+” covers everything up to the next bracket, so it can mean four hundred thousand. To compare your app against a ranking, reviews give a sharp answer; installs give only an order of magnitude.
That is why our targeting verdict is based on reviews rather than installs. Installs remain useful for one thing: gauging whether a keyword leads to a market worth entering — a top 10 with five million median installs is a real market, a top 10 with a thousand is a niche. The review thresholds are here.
Method and limits
30 keywords, top 10 collected on 23 August 2026 on the French Play Store. The limit is in the data itself: Google publishes installs only in brackets, so every figure in this article is a FLOOR. A median of 100,000+ means “at least a hundred thousand”, not “a hundred thousand”. The gaps between categories, spanning several orders of magnitude, remain readable despite that imprecision; a twofold gap would not be.
Like the previous ones, this study observes existing rankings. It does not prove installs cause the ranking — only that a certain volume coexists with it.