Your Dislike Count Never Left, YouTube Just Moved It Again

Yes, you can still see dislikes on YouTube, and the answer splits in two depending on whose video it is. On your own uploads the number is exact, sitting in YouTube Studio where it has always been, taken from YouTube’s own count. On anybody else’s it is an estimate produced by a browser extension, and on a small or recent video that estimate can be a long way out. Studio also rearranged itself in 2026, which is why a fresh wave of creators are convinced the data was deleted outright.

Two changes happened about a month apart

If you have read two articles on this that flatly contradict each other, there is a reason. Two separate things happened at two different times.

On 10 November 2021 Google announced that the public dislike count would be hidden. The button stayed. The number beside it stopped being visible to viewers.

Then on 13 December 2021 the dislike field was removed from the YouTube API, so the value now comes back only when the request is authenticated by the owner of the video. That second change is the one that mattered technically, because it cut off the supply of real numbers to every tool that had been reading them. Anything written between those two dates is describing a world that lasted about a month. A lot of what you find on this topic is confidently wrong for exactly that reason.

Where the exact number lives right now

Open studio.youtube.com and click Content in the left sidebar. Hover the video you care about and click the analytics icon, the small bar chart. Open the Engagement tab. The report called Likes (vs. dislikes) is the one you want, and it gives you the count plus the ratio over time.

For the whole channel on one screen, go to Analytics, then Engagement, then switch on Advanced mode. Add the Likes (vs. dislikes) metric and you get a sortable table across every video you have published. That is how you find the single upload dragging your average down.

Two things you cannot get, no matter where you click. You cannot see which accounts disliked you, and YouTube has never exposed that anywhere. You also cannot see dislikes on somebody else’s video, which was the entire point of the change.

The raw count is the least useful thing on that screen

What actually helps is comparison. Pull the like-to-dislike ratio for your last ten uploads and you have a baseline for your own channel and your own topic. A video sitting three or four times above that baseline is pointing you at something specific inside it. Maybe a claim people disagreed with. Maybe a title that promised one thing while the video delivered something else. A dislike count with nothing to hold it against tells you almost nothing.

The 2026 shuffle that made people think the data was deleted

Around the middle of May 2026, the Likes (vs. dislikes) column began vanishing from the Content page table in Studio. That column was how most creators read their catalogue. Views and comments stayed. The like and dislike bar went.

YouTube did not announce it. The rollout went gradually, so different channels lost the column in different weeks, and confused creators turned up in YouTube’s own support forum around late June. There is no toggle to bring it back, because the column was taken out rather than hidden.

The underlying data was never touched. It simply moved a level deeper into Analytics, which is a minor annoyance once somebody tells you where to look.

Shorts lost the button completely

On 25 June 2026 YouTube announced that Shorts would lose the dislike button altogether. The thumbs up was replaced by a heart. Negative feedback moved into the three dot menu, where Not interested and Don’t recommend this channel ask for a reason before they register.

Here is the part that changes your reporting. YouTube said the Shorts dislike count would stop updating by the end of June 2026. Historical Shorts data stays readable in Studio. New Shorts do not accumulate a count at all. Long form videos and live streams were left alone, and the thumbs down still works there.

If most of what you publish is Shorts, your dislike ratio has quietly stopped being something you can track. Your read on negative reception now has to come from audience retention and the comment section.

What the extension is really showing you

Return YouTube Dislike is the one almost everybody installs. It has passed five million users and was still receiving updates through 2026. The number it prints is still a guess, and where that guess comes from should change how much weight you give it.

Part of it is a real archive. Before the API closed on 13 December 2021, the project scraped and stored dislike counts for close to a billion videos, so anything published before that date shows you recorded history. After that date it builds a number out of whatever it can still reach. It counts the votes cast by people who have the extension installed and scales that sample up to the full audience. Where the sample is too thin, it estimates from the ratio between views and likes.

Think about what that means in practice. A video with ten million views, watched by a decent slice of those five million extension users, produces a reasonable estimate. A video with four thousand views published last week might have two extension users in its entire audience, and the number on your screen is an extrapolation from those two people. Both figures appear in the same font, with the same confidence. The project says as much itself: unpopular videos uploaded after the cutoff carry less accurate data than popular ones.

So on an old video the extension number is close to the truth. On a small or recent one, read it as a mood indicator and never quote it as a fact about somebody’s channel.

Phones and televisions are where all of this falls apart

Browser extensions do not run inside the YouTube mobile app, and they do not run in the YouTube app on a smart television. On Android there is a workaround. You use a browser that supports extensions and then force the desktop version of the site, because the extension does nothing against the mobile site. On iPhone the only route involves sideloading, which most people should not be doing for a dislike count.

That limitation matters more than it sounds. Most YouTube viewing does not happen in a desktop browser at all. So the number you are squinting at is a desktop-only artefact of a mostly mobile audience, estimated from the desktop viewers who chose to install one particular extension. Every layer narrows the sample further.

The permission these tools ask for is the real cost

Slow down here. A Google account is usually the key to somebody’s email, and that email is the key to every password reset running through it.

Any extension that rewrites what you see on a page needs permission to read and change data on the sites you visit. In the right hands that is unremarkable. The problem is what else turns up wearing the same costume. Through 2026, researchers reported campaigns of copycat extensions impersonating well known brands, YouTube-themed ones included, published on the official store and sitting in people’s browsers for months. One January 2026 report covered thirty extensions harvesting credentials from more than 260,000 users. Another described over a hundred extensions injecting remote scripts and lifting session data.

The technique that should worry you skips passwords and goes after your live session instead. An extension with page access can read session cookies and tokens while you are signed in. An attacker replaying those in their own browser is already inside the account, which walks straight past two-factor authentication, because no login ever happens.

Four rules cover most of the risk. Install only through the install link on the project’s own site, so you land on the correct listing instead of a lookalike. Read the publisher name before you press add. Never install anything that asks you to sign in with Google to show you dislikes, because no legitimate dislike tool needs your account. And go through your extension list every few months and remove whatever you no longer use, since a dormant extension can be sold and updated into something else without you touching it.

Do dislikes hurt your ranking?

This is where you expect me to say that dislikes tank your video, that the system reads them as a satisfaction signal and buries you accordingly, and that you should be watching that ratio like a hawk. It sounds obviously true. The public evidence does not support it being anything like that strong.

YouTube has never published a weight for dislikes in ranking, so anybody telling you the exact effect is guessing. What exists is an outside measurement of how far that button moves the machine. In September 2022 Mozilla published a study run through its RegretsReporter extension, with around 20,000 participants and a survey of 2,758 of them. Clicking Dislike prevented about 12 percent of unwanted recommendations. Not interested managed about 11 percent. Don’t recommend channel did far better at roughly 43 percent, and removing a video from watch history cut about 29 percent.

Read that from a creator’s chair. If the dislike button barely shifts what a viewer gets shown next, it is unlikely to be the heavy hand people imagine on the distribution side. The reason YouTube gave for hiding the count says something too. Coordinated dislike attacks landed hardest on smaller channels, and that only works while the count is visible.

Reading a dislike wave without panicking

Take this section seriously, because the instinct here is usually the wrong one. A spike feels like a verdict on the video. Most of the time it is information about who turned up, and it deserves ten minutes of checking first.

• Open the audience retention graph for that video. If people are watching to the end, the dislikes came from viewers who left early, which normally means they arrived from somewhere unexpected.
• Check Traffic sources on the same screen. A wave that starts the day a large channel or an external site links to you is telling you about that audience, and your video did not change overnight.
• Read the comments sorted by newest. A real content problem shows up in the wording people use. A brigade shows up as the same phrase repeated by different accounts.
• Measure the spike against the ten-upload baseline you worked out earlier, and never against a percentage you read somewhere. Every channel and every topic sits at a different normal.
• Work out whether it is one video or the whole channel. A single upload is an event. A ratio drifting across everything you posted this month is a pattern, and that deserves a different conversation.

Two more signals belong beside those, and neither can be brigaded. Returning viewers, in the Audience tab, tells you whether the people who watched came back for the next one. That is about as close to a verdict as YouTube will hand you. Rewatches and shares tell you whether anybody valued the video enough to send it on. A video can pick up an ugly dislike ratio and still pull its audience back next week.

What you should not do is delete the video that day. A deletion throws away the watch time and every comment under it, and does nothing about the reason people pressed the button. If there is a genuine error in the video, a pinned correction usually repairs more than a deletion ever will. Retention against your own median stays the most useful reading, because it points at a timestamp you can go and rewatch.

What you can see when the video is not yours

Nothing reliable. There is no accurate public dislike number on any video, for anybody outside that creator’s own Studio. Sites promising to reveal hidden dislikes are reading the same estimates, or inventing them. The comment section and the like count relative to views tell you more than any figure scraped off a browser add-on.

The number	Who can see it	How accurate
Dislikes on your own long form video	You, in Studio Analytics under Engagement	Exact
Dislikes on your own Shorts	You, in Studio, historical only	Exact up to end of June 2026, frozen since
Which accounts pressed dislike	Nobody	Never available
Another channel’s video, before 13 December 2021	Extension users, on desktop	Archived real data
Another channel’s video, after that date	Extension users, on desktop	An estimate, weakest on small videos
Anything inside the phone or TV app	Nobody, extensions do not run there	Not available at all

The short version for a creator. The exact number is still yours, and it now sits two clicks deeper than it used to. Shorts no longer produce one at all. Everybody else’s dislikes are an estimate that gets shakier the smaller and newer the video is. Check yours in Studio and let the retention graph do the work you wanted a dislike count to do.