How to Find Your Most Liked Tweet - Full Guide
Two ways to find your best-performing posts: the analytics export, and advanced search with a minimum engagement filter.
Timeline sorting
However, X will not arrange your own timeline by likes, replies or reposts. When you open your profile, the order is chronological. No engagement column, no sort menu, and no most liked filter. This is the gap that leads people to search for workarounds. The three available your ways are using the search box with engagement operators, the overview web analytics dashboard for recent standouts, and the full data archive for anything older. Search is about three times the fastest of the three, and requires no download waiting time.
Search for data as you would with any other report, starting from the search textbox in the web app or within the Explore search field, on mobile. You want to limit it to your own posts only, and so every query will start with from: followed by you handle without the at sign. And everything after that is filtered. Filterless X returns an unordered chronological dump - and even then, its not ordered by number of likes. Enter a single line with the complete query Thank you, the results load as a your normal list of search that can open one after another.
Favorite threshold
In the search box write from:yourhandle min_faves:1000 Nonetheless, because ohm only returns posts that exceeded that total like count. If the list is empty, divide the number in half and try searching again. If the list is lengthy, increase the number until you have at most a few results you can truly skim over. Start at 1000, Higher is faster to read than a mass of mid-range posts. Test your handle to twit the account you commonly use, without the @ sign exist of it.
match, min_faves: medal operator which indicates a minimum number of likes Do note the number after the colon is a floor so, 500 means five hundred likes or more. Get that floor worked up or down till the result set is small enough to see by eye. Check each result by opening it to see the number of likes on the actual post itself, not just the snippet. Second confirmation, run the same query on the web client. And Ley (PDF) sees that many posts build success out of shared characteristics rooted in different time periods since the very definitions of an ideal post can change at any given moment, so instead quickly write down the winning examples or bookmark them before you adjust the number again.
Combined filters
And finally min_retweets and min_replies do the same thing as well for those retweets & replies. from:yourhandle min_retweets:100 returns only your tweets that has at least 100 retweets. Then:from:yourhandle min_replies:50 does the same for replies. They can be stacked in a single line: from:yourhandle min_faves:500 min_retweets:50 Stacking reduces the set to posts that hit every bar simultaneously, something useful if liking alone still leaves too many hits. Use real numbers for your account. It will make zero at min_retweets:500 for a little account, so it should start closer to 20.
You are limited to using a given year or month when you add since: and until: That is a four-digit year with hyphens in the format since:2016-01-01 until:2017-01-01. The until date is exclusive, meaning that pair takes care of all 2016. Place the dates by the handle and engagement floor: from:yourhandle min_faves:200 since:2016-01-01 until:2017-01-01 APOLLO 11 — Use this when you know the time period but want the best post in that stretch. Omit dates when you care about all-time high, but not which year it was.
Analytics dashboard
Visit Open Analytics from the side menu in the web. On some layouts, the entry lives under More, and on others as an Analytics link directly. It displays your best-performing posts from a specified time frame with impressions and engagement rates listed at the end of each row. Click on a post to see the breakdown. This is the right screen when you want recent standouts without having to write a query. If you need a like floor or a year that is outside the dashboard range, it will not be a stand in for search.
The bottleneck is the time span. Analytics only goes back so far (no more than a few years scrolling) so its plug and play for recent stars and of minimal use to find anything from 2016. Now change the period control that you see at the top of page, and then watch for the top-posts list being refreshed. If the control won't go back far enough: Stop; use search (with since and until) or make an archive request. It does not necessarily mean a blank list means that you never hit. It usually indicates that the hit is older than the window.
Downloaded archive
The only complete answer is that you have this in an archive they downloaded: every post with counts of likes and reposts. As for you, request it in Settings, Your account, Download an archive of your data. Wait until X marks the archive ready. Download the zip file, unpack it, and load the offline viewer in a browser. Viewer displays posts with frozen counts from the instant that archive was created. That snapshot is what you sort through when the search and analytics both conspire against you with older posts.
If the viewer is clumsy, directly read the data files. They reside in the data folder, within the unzipped archive. In addition, every individual post record includes the like count and repost counts at the moment when the archive was created. Paste these records into a spreadsheet and sort the similiar column by descending. The first row is your biggest tweet for that entire account. An alternative option is to use a search operator, which is slower than the dashboard for search and slower than analytics; however this covers years that are not shown in dashboards.
Review and cleanup
The majority of people do this search to discover what is effective. Find the shape of it, not the topic. A short thread, a screenshot with a line of text saying one thing or even — the extremely powerful yes-oriented question— could carry much better to something new than just the hook. Note the structure of their top three hits also and grow indifferent to the precise news they sat on. Format and framing move. One topic goes viral; one typically does not. Run this once, and never refresh that same query again week after week.
Sometimes it's because a viral old post has not aged well, and the high-engagement post is precisely the one that continues to be discovered. Author: —Twitter:@yourhandleRun the block from:yourhandle min_faves:500 and keep increasing until you find the one who looks uncomfortable The fix for a single post is deletion. When a full period is involved rather than just one post, TweetSweep manages it in bulk. Instead of having to open up each URL by hand, you can tell Point TweetSweep the date range that has the offending posts in it and let it (un)delete them while following the dates!