User-generated share links: when the thing that drives installs is the content, not the channel

Someone shares a recipe from your app to a group chat. Four people tap it. Two have the app and land on the recipe. Two don't, install, and land on the home screen — with no recipe, no idea what their friend sent, and a fair chance of closing the app. The share worked for half the people it reached.

What problem does this solve?

Every piece of shareable content in the app should produce a link that opens on that content whether or not the recipient has the app. And you want to know which content drives installs, which is a different and more useful question than which channel does.

How do you set the links up?

The share sheet generates the link at share time, with the content identifier in the payload and the sharing surface in the source tag. No referrer reward, no incentive — this is organic sharing, and the attribution is about the content, not the sharer.

{
  "deepLinkPath": "/recipes/r_51902",
  "customData": {"src": "share-sheet", "via": "messages", "contentType": "recipe"}
}

How is this different from a referral programme?

Sharing is often the larger of the two by volume and the cheaper by far, because nobody is paid. The mistake is treating it as referral-lite and bolting a reward on. A reward invites gaming; plain sharing doesn't, and the numbers stay clean.

Which content should you make shareable?

All of it, and then look at the data. The content that drives installs is rarely what the product team expected. Attribution by content identifier tells you which recipes, workouts or listings are doing the work, and that feeds directly into what to feature.

What makes people actually share?

Three things, all on your side of the link. The share action has to be one tap from the content, using the native share sheet so it goes wherever the person already talks to their friends. The link has to unfurl properly — a preview image and a title in the group chat, generated from the content itself, so the recipient sees a recipe or a listing rather than a bare URL. And the recipient's first screen has to be the content, not a sign-up wall in front of it. Ask for the account after they've seen what their friend sent; a share that lands on a login form is a share that mostly fails.

Which metric should you watch?

Installs per share, by content type, then retention of share-driven installs against other sources. Share-driven users usually retain better — they arrived with a reason and a friend. Confirming that with your own numbers is what justifies investing in the share flow.