An app has an awkward shape for a link in bio.
A t-shirt has one link. A course has one link and maybe a preview. An app has two stores, possibly a web version, a support address, a changelog, a Discord somebody started in a burst of optimism, and a press kit that has never been downloaded by anybody who was not the developer testing whether the download worked.
So app pages get long. Long pages are where taps quietly stop happening, because the visitor has been handed a menu when what they wanted was a queue.
I went looking at pages belonging to small apps, the kind run by one or two people who also have other jobs. Two are worth pulling apart properly. I have changed the names, because these are tiny operations and a public critique lands harder when the team is one person and a laptop. The setups are as I found them: same link order, same wording, same gaps.
One caveat before any of this is useful. I cannot see their install numbers. Nobody outside can. The App Store and Play Store keep that to themselves, and a link page shows you taps, not outcomes. So this is not a teardown of what worked. It is a teardown of what each page asks a visitor to do, and how much effort it costs them to do it.
An app page is a queue, not a menu
Three kinds of people arrive at your page, and they arrive in wildly different states.
The first has never heard of the app and clicked out of mild curiosity from a video. They need to know, in about two seconds, what the thing is and whether it runs on the device in their hand. The second has already decided. They watched the video, they want it, and every extra thing on the page is an obstacle between them and the store. The third is already a user; something has broken, and they are hunting for a way to contact a human.
The first two groups will not scroll. The third will scroll all day, because they are motivated by mild fury.
That suggests an order, and the two pages below both stumbled onto it independently, which is usually a sign that a thing is real rather than clever.
Two product that have used RelayMe on their social pages:
Nightink - a private diary
Nightink is a private diary that stays on your phone.
Firstly, nothing leaves. Entries are written to storage on your device. There is no server to send them to.
Secondly, nothing to sign up for: no email, no password, no name. Open it and start writing.
Thirdly, Nightink exports everything to plain text, Markdown or a full backup — free, always — because a diary you cannot take with you is not really yours. You can backup whenever you like. Restore on a new phone by picking the file.
Duskloom puts the free way first, then adds six links too many
Duskloom is ambient audio for sleep. It exists as a website, an iOS app, and an Android app, which is the hardest case for a single page.
The first link is not a store. It is to listen in your browser. That ordering is the most interesting decision on either page. Installing an app is a real cost: storage, permissions, an account, the vague sense of another thing on the phone. Opening a tab costs nothing. If the browser version is any good, the store link becomes something a person taps after they already like it, which is a much easier tap.
Below that, the iOS and Android links sit side by side with small screenshots attached to each. The images are doing something plain wording cannot: they show the app is dark, quiet and uncluttered before anyone commits to an install. A per-link image does an unreasonable amount of work for how little effort it takes, and almost nobody uses them.
The page also uses a heading and a divider. Everything above the divider sits under "Try it". Below it, under "Help", are the support email and a short FAQ. That split stops the support links from competing with the download links, and it means the furious existing user can find their bit by scanning rather than reading.
And then it keeps going. A Discord with eleven members, none of whom have posted since spring. A public roadmap. A press kit. A link to the developer's other app. A newsletter that has sent one issue. Nine links after the divider, and each one is a small invitation to leave.
I do not think any of those links are bad ideas in isolation. Collectively they turn a queue back into a menu. If a link would embarrass you slightly when a stranger clicked it, it is costing you more than it earns.
Where Relay fits, and where it does not
The things those two pages get right are mostly structural, and structure is what a link page either gives you or does not. Ordering, a heading, a divider, an image per link so that a store button is not just a word.
That is roughly the shape of Relay. The free plan has no card and no time limit, unlimited links on every plan, your own photo and bio, social icons, five themes, headings and dividers, tap and view counts, a QR code, and a story-sized share image for the times you want the page in a story rather than a caption. Pro is $30 a year or $4 a month plus VAT, and adds the per-link images Duskloom leans on, all forty-seven themes, all eight fonts, custom colours, a background image, scheduled links, inline players, email capture, and removal of the Relay badge. Scheduled links are the one I would use for an app: you can set the "version 2 is out" link to appear on the morning it actually ships, rather than remembering to do it while refreshing review status.
Relay is also part of the problem this article describes. There are no custom domains, so your page is relayme.bio/yourname and not links.yourapp.com, and for a product trying to look like a company that will still exist next year, that is a real cost. There are no payments either, which for most apps does not matter, since the store takes the money and the store's cut whether you like it or not. There is no device detection anywhere in Relay, so an iOS link and an Android link have to be two labelled links. We wrote an honest comparison with Linktree that includes four rows where Linktree is the better answer, and domains is one of them.
What you do get: data held in the European Union, self-hosted fonts, no advertising, no third-party analytics loaded on your page.
The links an app page probably needs
Stripped down, the version I would build looks like this. The cheapest way to try the thing, if a cheap way exists, which usually means the web version. Then the store for whichever platform most of your audience is on, named as the store. Then the other store. Then one piece of proof, a short demo clip or a screenshot set. Then support, under a heading, with a divider above it so the two halves of the page are visibly different.
That is five or six links. Everything else is a thing you want to exist rather than a thing a visitor wants to tap.
On device detection: paid deep-link services will sniff the operating system and send iOS one way and Android the other. They work. They also cost money and add a redirect, and if you have not shipped a few hundred installs, two clearly labelled links will lose you almost nothing. The label is the fix. "iPhone and iPad" beats "App Store" for people who do not think in brand names.
What the numbers will not tell you
A link page can tell you how many people viewed it and how many tapped each link. That is useful for comparing link one against link four, and for noticing that nobody has ever tapped your press kit.
It cannot tell you installs. The store dashboard knows installs and knows nothing about your page. Joining those two facts together requires an attribution layer, which costs money and takes a weekend, and is a thing to consider at the point where you are spending real amounts on ads rather than before. If you are testing small budgets, what $50 of ads actually buys you is a more useful place to start than an attribution setup.
The other thing the numbers will not tell you is how long any of this should take. Not for an app, not for anything. There is a rough timeline for a first sale that runs longer than most people expect, and app installs behave much the same way. A page that is ordered well does not create demand. It stops you losing the demand you already made somewhere else.
None of this is specific to apps. A Shopify store with three collections has the same problem in a smaller way, and the same fix, which is fewer links in a deliberate order. If you are still working out whether you need the store at all, TikTok Shop vs Shopify vs just a link covers that ground.
If you want to try the shape, the free plan takes about ten minutes. Put the cheap way to try it first, name the stores, say what device it needs, and put support under a divider where the angry people can find it.