Search
How to find an old tweet you liked
Start with the clues you remember, work through a search example, and check whether the post was imported.
Separate the clues you remember
Finding an old liked tweet is easier when you turn a vague memory into a few independent clues. Write down the subject, any unusual words, the author if you remember them, and whether the useful information was in the post itself or in something it linked to. You do not need all four.
Suppose you remember a designer describing a checklist for welcoming new customers. The clues might be onboarding, checklist, and a familiar designer. Treat that as an example search strategy, not an exact quote from a real post.
Search one clue at a time
In Likes Library, set the search scope to All liked posts and begin with the most distinctive word. Search onboarding first. If the results are too broad, try onboarding checklist. Both words must appear in the saved text.
If that returns nothing, remove a word before adding more. The writer might have said welcome instead of onboarding. Try the alternative separately. Long reconstructed sentences often fail because the original wording was different. The search guide explains matching and scope in detail.
Confirm the post is actually in your library
Search cannot find a post that was never imported. A recent like made on your phone may need another browser import. Older history may be absent because X stopped returning it, or because the post became inaccessible before collection. Uploading relevant X archive files may add records not present in the browser import, but an archive is not a guarantee of complete history.
Also consider what you saved. A bookmark alone does not create a like. A liked reply does not mean the entire conversation was imported. Words visible in an image or a linked article may be missing from the saved post text.
Try X search for a post still available there
If the post is still accessible on X, its Advanced Search offers fields for words, accounts, and publication dates when you are logged in. Use an author and an approximate date range to narrow a public search. This searches X results, rather than the contents of your Likes Library account.
A post’s publication date and the day you liked it are different clues. A post you liked last week may have been written years earlier. If you find the original, keep its link in your project notes and run another import if needed. For items that matter to you, keep a separate backup of the text you have already collected.