How to Find Your Top Performing Twitter Posts in 2026

The short version: the fastest way to find your top performing Twitter posts in 2026 is through X Pro's analytics dashboard at analytics.x.com, sorted by engagement rate rather than raw impressions, because impressions on X are wildly inflated by scroll-past views that nobody reads. The posts worth chasing are the ones with high profile clicks, replies, and link clicks relative to how many people saw them, not the ones with the biggest view count sitting at the top of your feed.

Why "top performing" isn't what you think it is

I used to sort my own X analytics by impressions and feel proud of whatever sat at the top. Then I cross-checked one of those "top" posts against my website traffic and found it had sent precisely zero clicks to my site in three weeks. It had 210,000 impressions. It had 4 replies. It did nothing for my business except make me feel briefly good about myself on a Tuesday morning.

Meanwhile a scrappy, unglamorous tweet I posted about a client mistake I'd made with a chatbot rollout, the kind of post I nearly deleted because it felt too honest, sat at 9,400 impressions and generated 71 profile visits and three DMs, one of which became a paying client. That's the post that mattered. It never once appeared near the top of any "sort by impressions" view.

This is the bit most guides on this topic skip over. X inflates view counts because a post counts as "viewed" the instant it renders on someone's screen for a fraction of a second while they're scrolling to something else entirely. A tweet can rack up 50,000 views from people who never read a single word. If you're hunting for your best content based on that number alone, you're optimising for scroll speed, not for what moves your business forward.

Step one: pull your data from X Pro

Go to analytics.x.com (this replaced the old analytics.twitter.com dashboard, and X Pro is the paid tier that gives you the fuller breakdown). Here's the exact path:

• Log in with the account you want to analyse, not a linked business account, X ties analytics to the individual profile
• Click into the "Posts" tab, not the home overview, which only shows you a 28-day summary snapshot
• Set the date range to the last 90 days as your first pass, then narrow later
• Sort by "Engagements" first, then check that same list sorted by "Engagement rate", because these two lists are rarely identical and the gap between them tells you something important

When you sort by engagement rate rather than raw engagements, small accounts and quiet posts suddenly climb the list. A tweet with 800 views and a 12% engagement rate is doing something a tweet with 40,000 views and a 0.4% engagement rate is not. If you want the full breakdown of which metrics X shows you and which ones it's quietly retired, I wrote a longer piece on this at what to track in your Twitter analytics, worth reading alongside this one because the dashboard changes every few months and it's easy to be looking at a metric that no longer means what it used to.

Step two: export it and stop trusting the sort order

X Pro lets you export your post data to CSV. Do this every month, not just when you're curious, because the dashboard only holds a rolling window and you'll lose your history otherwise. Once it's in a spreadsheet, add three columns yourself:

• Profile clicks divided by impressions (this tells you how curious people were about you, not just the topic)
• Replies divided by impressions (this tells you if you started a conversation or just got scrolled past)
• Link clicks, if the post had one, as a raw number

Sort your spreadsheet by that first column, profile-click rate. In my own data going back through 2025, the posts that ranked highest by this measure were almost never the ones with the most likes. They were opinionated, slightly uncomfortable posts where I said something that made people want to know who was saying it. One post that simply stated "most AI consultants can't run a P&L, they just know prompts" pulled a profile-click rate nearly six times higher than my average, despite modest total reach.

Step three: check what time and day those winners went out

Once you've got your shortlist of five to ten strong posts, look at when they were published. This matters more than people admit. A post that performs brilliantly at 8am on a Tuesday might flop completely if you repost it at 6pm on a Friday, not because the content is weaker but because your audience simply isn't there. I cover the research on this in the best time to post on Twitter for more engagement, but the short version is that for most B2B and consultant-style accounts, Tuesday to Thursday between 7am and 9am UK time (accounting for your specific audience's location) consistently outperforms weekend posting by a wide margin. If your top performers cluster around a particular time slot, that's not a coincidence you should ignore.

Step four: separate "viral" from "valuable"

These are not the same thing and treating them as the same thing is where most people go wrong. A viral post is one that spreads fast among strangers. A valuable post is one that moves the people who see it closer to becoming a client, a subscriber, or a collaborator. You can have a post that goes viral and does almost nothing for your business, and you can have a post that eight hundred people see and it lands you a five-figure contract.

If you run a small account and you're chasing virality specifically, that's a different exercise with different mechanics, and I wrote about that separately in how to go viral on Twitter with a small account. But for the purposes of finding your top performers, don't let a lucky viral hit distort your sense of what's working. One outlier post shouldn't set your content strategy for the next six months. Look at your top ten to fifteen posts as a pattern, not your single best post as a formula.

Step five: look for the topic and format threads, literally

Once you've got your real top ten, sitting there sorted by profile-click rate, read them side by side and ask two questions: what topics keep appearing, and what format are they in. In my own list, six of my top ten posts from the last two quarters were short numbered lists (three to five points, no more), and four of them were about a specific mistake I or a client had made rather than a generic tip. Generic tips like "post consistently for growth" almost never made the list. Specific, slightly uncomfortable admissions did.

This is worth writing down, not just noticing and forgetting. Keep a simple running note, even a basic spreadsheet tab, titled "what worked" with columns for topic, format, and the profile-click rate. Review it every quarter. You'll start to see your own patterns emerge that no generic Twitter advice article could ever tell you, because they're specific to your audience, your industry, and your voice.

What to do once you've found them

Finding your top performers is only useful if you act on it. There are three concrete moves worth making:

• Pin your best-performing post, or better still, rotate between two or three strong posts so new visitors to your profile see proof of what you do well straight away. I go through exactly how to manage more than one pinned post at a time in how to pin multiple tweets on your Twitter profile, since X only lets you pin one at once officially, but there's a workaround using highlights
• Repurpose the topic, not just the wording, into a new post using slightly different language a month or two later. Testing the same idea twice tells you whether the first result was a fluke or a genuine pattern
• Schedule your best format at your best time going forward rather than posting reactively. If you're not already using a scheduler, there are decent free ones, and I've laid out the full process in how to schedule Twitter posts for free

The uncomfortable bit about third-party tools

There are dozens of tools promising deeper Twitter analytics than X's own dashboard, and some of them are handy for scheduling and cross-platform comparison. But be careful trusting any third-party tool's "top posts" ranking over your own exported data, because most of them still default to sorting by impressions or likes, the exact metrics I've just told you not to trust blindly. If a tool's dashboard doesn't let you sort by engagement rate or profile-click rate specifically, it's giving you the same flawed picture X's home screen gives you, just with nicer colours.

I'd also say this plainly: X's own algorithm changes have made cross-account comparison nearly meaningless since the shift toward paid verification affecting reach. A post from a Premium subscriber and an identical post from a non-subscriber will not perform the same, all else being equal. So when you're benchmarking your top performers, benchmark against your own past posts, not against a stranger's viral thread you saw screenshotted somewhere. Your baseline is you.

How often you should be doing this review

Monthly is enough for most people. Weekly is overkill and will drive you slightly mad chasing noise in small numbers. Quarterly is too slow, you'll miss shifts in what's working while your audience or the platform itself changes underneath you. I do mine on the first Monday of each month, thirty minutes, spreadsheet open, no exceptions, and it's become one of the few marketing habits I've kept consistently through five rough years of rebuilding this business in public.

One last thing worth saying. If you're running content across more than one platform and you're stuck for what to post next once you've spotted a winning topic on X, it's often worth adapting that same idea for LinkedIn rather than starting from a blank page each time. I've written about exactly that process in where to find content ideas for LinkedIn when you've got nothing to say, and the honest truth is your best X post from last month is very often your next LinkedIn post, reworded and given more room to breathe.

Related guides live in the Twitter (X) Help: 21 Guides to Features, Verification, Video Fixes and Tools.

Frequently asked questions

How do I see my top tweets from a specific month in 2026?

Open analytics.x.com, go to the Posts tab, set a custom date range for that month, then sort by engagement rate rather than impressions. Export the data to a spreadsheet if you want to keep it beyond the platform's rolling data window, since X doesn't hold older history indefinitely.

Is X Pro required to see top performing posts, or can free accounts check this?

Free accounts get a basic 28-day snapshot on the home analytics screen, which is enough for a quick check. The fuller Posts tab with exportable data and longer date ranges generally requires an X Pro subscription, which currently starts at a modest monthly fee and is worth it if you post regularly for business purposes.

Why does my most-viewed tweet get so few replies or clicks?

High views with low engagement usually means the post was widely scrolled past rather than read, often because it appeared briefly in someone's feed without stopping them. Impressions on X count almost instantly, so a large view count doesn't guarantee anyone absorbed the content.

Should I delete tweets that performed badly?

No, leave them. A weak post rarely hurts your profile, and deleting posts wastes time you could spend studying why your strong posts worked. Focus your energy on repeating the patterns from your top ten to fifteen posts rather than tidying up the bottom of the list.