Social platforms are excellent at showing us what is happening now. They are less reliable at preserving the practical knowledge people share along the way.
That is the problem I have been working on with Feedcover: a community for useful real-life experiences, thoughtful questions, newsletters, and collections.
Why structured content types?
A single generic post format made discovery feel noisy, so Feedcover treats four formats differently:
- Feeds for practical stories and updates
- Questions for focused answers and discussion
- Newsletters for longer editorial content
- Collections for grouping related feeds into a useful sequence
They share the same identity, engagement, moderation, and discovery systems, but retain the interaction pattern that makes each format useful.
One product across web and mobile
The web application and the Expo/React Native mobile apps use the same API. That has helped us keep profiles, follows, discussions, saves, notifications, messaging, wallet state, and content permissions consistent across platforms.
Android is now available on Google Play, while the iPhone version is progressing through TestFlight and App Store review.
Relevance without a black box
The home experience ranks mixed content using signals such as likes, saves, shares, views, completion, responses, follows, categories, tags, product preference, and creator affinity.
It also excludes blocked or reported material, limits repeated creators on the first page, and introduces controlled exploration so a new creator can still be discovered.
We currently run this with an open-source recommendation stack rather than relying on a proprietary feed service. That gives us room to understand, measure, and improve the ranking as the community grows.
What I would value feedback on
I would especially appreciate honest feedback about:
- whether the distinction between feeds, questions, newsletters, and collections feels natural;
- whether the first few recommendations feel relevant;
- where the web or Android experience feels slower or less intuitive than expected.
You can try it at feedcover.com, or install the Android app.
I am interested in the unpolished feedback—the parts that make you hesitate, misunderstand an action, or leave a page. Those observations are usually the most useful ones.
Top comments (1)
Your signal list is likes, saves, shares, views, completion, responses, follows, categories, tags, product preference, creator affinity, and not one of them has an age term. All of that measures attention in the first 48 hours, which is the thing you say social already does well. The durability signal I'd add is engagement on content older than 30 days, saves and responses per view, since a question answered well two years ago keeps earning both while a hot feed post stops cold. We keep a cousin of that test on reply drafts at Viewfy: it ships only if it still helps with the product line removed.