How to Clean Up Your LinkedIn Feed

Start by unfollowing consistently irrelevant accounts, use “Not interested” on individual posts, review your following list and deliberately engage with sources you value. For a large list, use a review-first process rather than unfollowing everyone.

Identify the actual source of the noise

“My LinkedIn feed is bad” can mean too many suggested posts, generic career advice, repetitive AI content, updates from people you barely know or simply too much volume.

Spend a few minutes noting which problem appears most often. A feed dominated by two prolific authors needs a different response from a feed shaped by hundreds of old follows.

Use LinkedIn’s native controls first

LinkedIn documents these options in its feed preferences guide. Interface labels can vary by account and change over time.

1. Open the three-dot menu on a low-value post.
2. Select the most relevant feedback option, such as Not interested.
3. Unfollow the author when the pattern—not only one post—is the problem.
4. Review people and pages under My Network → Following & followers.

Unfollow without removing useful connections

Unfollowing normally stops a person’s posts from appearing while keeping the LinkedIn connection. This is often the cleanest choice for former colleagues, acquaintances and useful contacts whose posts are no longer relevant.

Do not treat unfollowing as a judgment about the person. It is a decision about what deserves limited feed attention.

Rebuild positive signals deliberately

Cleanup removes noise, but you also need to strengthen the signal. Follow a small number of consistently useful authors, visit their posts intentionally and interact only when you have a genuine reason.

Avoid liking posts merely because they are popular. That can reinforce subjects you do not actually want more of.

When the manual process stops scaling

Manual cleanup works well for a handful of accounts. It becomes tedious when years of follows have accumulated. At that point, write down clear “More of” and “Less of” preferences, test them on a sample and review the proposed decisions before applying anything broadly.

Feedwise follows that sequence. It does not impose a universal definition of a good feed; it applies the preferences you approve.