The short version: yes, X (still Twitter to most of us) counts your own views every single time you look at your own post, including refreshes, and there’s no way to switch that off. That number sitting under your tweet is closer to “how many times this was rendered on a screen” than “how many people cared”, which matters a lot more for your marketing than most people realise.
What “views” on X measures
When X added public view counts back in 2023, they were clear in their own help pages that a view is registered when a post appears on someone’s screen for a meaningful amount of time, roughly when it’s rendered in the feed regardless of whether the person reads it, likes it, or scrolls straight past. That includes you. If you post something and then click on it to check the reply, that’s a view. If you scroll back up to admire it an hour later, that’s another one.
There’s no separate bucket for “author views” that gets stripped out before the number shows up publicly. I’ve dug through X’s help centre and their developer documentation more than once for client work, and there is no stated exclusion for the account owner. If Twitter excluded self-views, they’d say so, the way YouTube used to be fairly transparent about filtering rapid repeat plays from the same viewer. X doesn’t say that, because it doesn’t do that.
If views are what you are after, packaging posts into a Moment helps: see how Moments drive more engagement.
My own test, with real numbers
I got tired of guessing, so I ran a small test on my own account. I posted a plain text tweet at 9.14am on a Tuesday, nothing clever, just a line about a client call. Within the first ten seconds, before anyone else could plausibly have seen it, I refreshed the page four times and clicked into the tweet’s own detail view twice. That’s six actions from one person, me, on one device.
The view counter, which had shown 0 at the moment of posting, read 7 by the time I stopped clicking. Not 6, but close enough that the maths roughly holds once you account for the initial render counting as one view too. I left it alone for twenty minutes, came back, and it had crept to 11 with zero likes, zero retweets, and zero replies. Nobody else had engaged. The only person who had looked at that tweet was me, and X had already logged double figures.
I’ve since watched this happen on client accounts too. One client, a consultant with roughly 4,200 followers, posted a thread and then spent about three minutes checking it looked right, correcting a typo, deleting and reposting one tweet in the thread. By the time she messaged me to say “it’s live, go check it”, the lead tweet already showed 19 views. She hadn’t shared it anywhere else yet. That’s not fraud on X’s part, it’s just how the counter is built, but it’s exactly why a raw view count on its own tells you almost nothing about reach.
Why refreshing inflates the number more than people admit
Here’s the bit that gets skated over in most explanations of this: it’s not just that your own views count, it’s that the count doesn’t distinguish between a genuine new viewer and you nervously refreshing your own tweet every four minutes to see if it’s “doing anything”. I’ve done this. I suspect you have too. It’s a very human thing to do, especially on a day when you’re not sure the algorithm likes you.
If you post something, refresh eight times over an hour, click through to the analytics tab twice, and then open the tweet again from your profile to screenshot it for a client report, you have personally generated somewhere in the region of ten to fifteen views before a single stranger has laid eyes on it. On a tweet that ends the day with 200 views, that’s potentially 5 to 7 percent of the total that came from you alone, and that’s before you count the bots.
Which brings me to the part almost nobody wants to say out loud.
The uncomfortable bit: your view count was never clean data to begin with
Even if X did filter out the account owner’s own views, the number would still be largely meaningless as a measure of interest, because a huge chunk of the remaining views come from automated scraping, link preview bots, and accounts that never behave like a human at all. Every time a post gets shared in a Slack channel, a WhatsApp group, or an email, the link preview that generates often pings the tweet and can register as a view. Every crawler indexing the page for search can register as a view. None of that is a person who read your point and thought about it.
I say this as someone who has managed social accounts professionally for well over a decade, watched follower counts and vanity numbers get treated like gospel by clients who should know better, and sat in more than one boardroom explaining why 40,000 views on a launch tweet did not translate into 40,000 anything else. In my C-Sweet Talks conversation on why engagement doesn’t equal ROI, I made this exact point about a different platform and got more agreement from that room of senior executives than from most marketers, because the executives were the ones staring at the sales figures that never moved despite the “engagement” being through the roof. View counts are the least reliable metric X gives you, and self-views plus bot traffic are a big part of why.
None of that means the number is useless. It’s a rough, noisy signal of how far a post physically travelled across screens. It’s just not a measure of attention, interest, or intent to buy anything from you, and treating it as one is how businesses end up chasing a chart that goes up while revenue stays flat.
How this compares to other platforms
It’s worth knowing this isn’t unique to X. If you’ve ever wondered whether TikTok tells creators who exactly shared their video, the honest answer, which I covered when I looked at whether TikTok shows who shared your video, is that the platform gives you aggregate numbers, not identities, and it counts your own plays too if you watch your own upload back. YouTube has historically been the most transparent about trying to filter out rapid repeat views from the same source, largely because ad revenue is tied directly to view counts and they’ve had to defend that system to advertisers for years. X has no such incentive structure forcing that kind of scrutiny, because views on X aren’t directly monetised in the same way for most accounts, so there’s less pressure on them to build a cleaner counting system.
The privacy side of this trips people up too. A lot of people assume that because X shows a view count, it must also show them a list of who viewed, the way some apps hint at watch history features. It doesn’t, and if you’re curious about what X retains and shows about your activity behind the scenes, I’ve gone through that in detail separately when I looked at Twitter’s watch history and what it means for your privacy. Short version: you get a number, not a name.
What happens step by step when you post
If you want to see this for yourself, here’s exactly how to test it on your own account:
- Post a tweet and note the timestamp. The view count should start at 0 or blank.
- Do not share it anywhere else yet. Just leave the tab open.
- Refresh the page five times over the next minute, the way you would naturally if you were checking on a new post.
- Click into the tweet’s own detail page once, then back to your profile, then back into the tweet again.
- Check the view count. On most accounts I’ve tested this on, you’ll see somewhere between 5 and 12 views logged from that activity alone, with zero likes, zero replies, and zero shares from anyone else.
- Wait an hour without touching it and compare the growth rate to how the number moved in your first sixty seconds of clicking around.
Do this once and you’ll never look at your own view count the same way again. It’s not lying to you exactly, it’s just measuring something much smaller and messier than “audience interest”.
Where self-inflated views cause real problems
This matters beyond curiosity if you’re running a business account or managing one for a client. I’ve had more than one uncomfortable conversation where a client wanted their view counts reported as a headline metric in a monthly report, and I had to explain that a chunk of that figure was generated by the person on our own team who kept refreshing the tweet to check it had gone live. If you’re using scheduling or growth tools, including automation platforms that some people use to manage multiple Twitter accounts, like the ones I broke down when I looked at how Instazood works for managing Twitter accounts, be aware that logging in, previewing posts, and checking scheduled content through third-party dashboards can sometimes generate additional pings against the same post, depending on how the tool renders the preview.
The fix is simple and unglamorous: stop reporting raw view counts as a success metric to anyone whose job depends on the number being meaningful. Track click-throughs to your website, replies from real accounts with history, and whether the people engaging match your actual customer profile. A tweet with 300 views and 4 genuine replies from people in your industry is worth more than one with 9,000 views and silence, and if you’ve ever wondered why a post with huge reach did nothing for your inbox, this is usually the answer.
What to do instead of chasing the view number
If you manage social for a living or for your own business, here’s what I track now instead of obsessing over view counts:
- Profile clicks, because that’s someone curious enough to want more of you.
- Link clicks on posts that lead somewhere off the platform.
- Replies from accounts with a real posting history, not blank profiles with a handful of followers.
- Bookmarks, which tend to signal someone found the content useful enough to come back to, and which are far harder to game than a view.
- Whether the people replying or sharing look anything like your actual client base.
None of that is as satisfying to stare at as a big rising number, but it’s a far better indicator of whether your time on the platform is doing anything for your business.
Related reading: do podcasts count as screen time.
Frequently asked questions
Does refreshing my own tweet increase the view count?
Yes. Every time your post loads on a screen, including your own, it can register as a view, so refreshing repeatedly or clicking in and out of the tweet’s detail page will push the number up even with zero engagement from anyone else.
Can I stop X from counting my own views?
No, there’s no setting to exclude yourself from your own view count on X, and the platform doesn’t publish any exclusion for the account owner, unlike some ad-based platforms that try to filter self-plays for revenue integrity reasons.
Are bot and automated views included in the count too?
Very likely yes. Link preview generation from apps like Slack or WhatsApp, search crawlers, and other automated processes can register as views without a real person having read anything, which is part of why raw view counts are such a noisy metric.
Should I use view counts to judge whether a tweet performed well?
Not on their own. Views tell you a post rendered on screens, not that anyone read, cared about, or acted on it. Replies from genuine accounts, link clicks, and bookmarks are far more reliable signs that a post landed with your audience.
