How do I get downloads for my app?

Downloads are two problems, not one

Almost every founder asking this question is really asking one of two much narrower ones, and the answers barely overlap. The first is a traffic problem: nobody arrives at your store listing. The second is a conversion problem: people arrive and leave without installing. From the outside they are indistinguishable, because the only symptom either produces is a download counter that will not move — and the fix for one does nothing at all for the other.

Check the conversion half first, because it is cheap to measure and it multiplies everything else you do afterwards. Both stores report how many people viewed your listing and how many went on to install. If that ratio is poor, a channel that would have worked looks like a channel that failed, and you will spend weeks blaming the wrong thing. Getting the icon, the first screenshot, the opening line and the rating right is App Store Optimization, and it is the only work here that improves every future channel at once.

Where early downloads actually come from

Once the listing converts, the question becomes who arrives. For an app with no budget and no audience, the ranking below is ordered by what tends to return most per hour of founder time early on — not by how big each one could eventually get.

1. People already discussing the problem Forums, subreddits, Discord servers and niche communities where your problem gets described in the user's own words. Slow, unglamorous, and the highest-converting traffic you will find, because the person has already articulated the need. The rules are the ones in getting users from Reddit without getting banned: be a participant before you are a founder.
2. One launch moment A launch concentrates attention into a single day, which matters more for an app than for a web product because early install velocity is one of the signals the stores themselves read. Launching is a spike rather than a channel, so the planning that pays is deciding in advance what catches the spike after it decays.
3. Creators whose audience is your audience A short demo from someone whose followers have the problem outperforms a much larger generic audience, because in a store the click and the install happen seconds apart, with nothing in between to lose people. The discipline is the same as working with creators anywhere: a handful of well-matched small partners beats one big one.
4. Store search itself The one free channel that compounds. Someone searching the store is already trying to install something in your category, which makes it the highest-intent traffic in app marketing, and it is won with the metadata and rating work above rather than with spend.

Paid installs are deliberately absent from that list. They work, and they are the quickest way to make the number rise. They are also the least informative purchase available to you before you know two things: that your listing converts, and what a retained user is worth. Until then you are buying a metric rather than learning anything, which is the same order of operations that applies to any paid channel.

What makes this different from marketing a website

The store sits between you and the user, and that gap changes which tactics are worth the effort. On the web you watch a visitor move through pages you own, you can email them, and you can ship a fix an hour after noticing a problem. With an app, a decisive part of the journey happens on a page you do not control, in software you cannot observe until it has been opened, behind a review queue that stands between deciding something and shipping it. Planning in that rhythm is cheaper than fighting it.

Attribution is the part that surprises people most. The question every founder wants answered — which post, video or thread caused this install — is precisely the one the platforms report most coarsely, for privacy reasons that are not going to reverse. We build the connectors that read App Store Connect, Google Play and Apple Search Ads for the founders who ask us this, so this is an observation about what the data contains rather than a claim to have grown an app ourselves: those surfaces give you totals, trends and spend honestly, and almost never a clean line from one piece of work to one install. The practical response is to instrument the half you do own — your own links and referral codes, plus a single how-did-you-hear-about-us question after signup — and to read store analytics as directional rather than forensic.

Finally, ratings sit on both sides of the equation at once. They are a visible conversion element on the listing and an input the stores weigh when deciding what to show, which means retention work quietly feeds acquisition in a way it does not on the web: people who stay leave better reviews, better reviews lift the listing, and a better listing lifts every channel pointing at it. That is why asking for reviews at the right moment behaves more like a growth lever for an app than it does for a web product, and why an app with a retention problem cannot buy its way out with installs.