Free Facebook Video Views

Three seconds and sixty seconds are two different numbers

Open the video insights for any Facebook Page and you find two counters sitting beside each other. One is always much larger than the other, and the distance between them is the most useful thing on the screen.

Facebook records a 3-second view when your video has played for three seconds, or for nearly its full length when the clip is shorter than that. In feed, video begins playing on its own while the post is on screen, so somebody scrolling at an ordinary pace who slows down for a moment has already produced one. This is the counter that grows fastest, and it is the one most people mean when they say "views".

Facebook records a 1-minute view when the video has been played for sixty seconds, or for nearly its full length when the clip runs shorter than a minute. Reaching that number takes a decision from the person watching. They stopped scrolling, they stayed through the opening, and in most cases they turned the sound on to follow what was happening.

Both counters describe the same video. They describe two different behaviours, and the system that decides who sees your next post reads them differently. Once you can see which of the two your videos are earning, most of what follows becomes straightforward.

What the three-second count tells you

The three-second number is a measure of how often your post reached a feed and held the eye long enough for playback to register. It moves with distribution and with the opening frame. Two videos with the same audience and the same reach can land far apart on this number purely because one of them opens on a face and the other opens on a title card.

Read it as a reach-and-hook metric. A large three-second count says the post got in front of people and the first frame did its job. What happens in the following fifty-seven seconds is a separate question with a separate answer.

What the sixty-second count tells you

The one-minute number is a measure of minutes viewed, and minutes viewed is the currency Facebook itself spends. In-stream ad eligibility is written in minutes: sixty thousand minutes viewed across sixty days, alongside a follower threshold and a minimum number of live videos. That requirement is denominated in watch time, not in views.

So a Page carrying a very large three-second count with little watch time behind it sits further from that threshold than a Page with a tenth of the views and genuine retention. When you are deciding which number to celebrate, this is the practical difference between them.

Why distribution follows the longer number

Feed ranking works on predicted value. For every candidate post, the system estimates how likely you are to spend time on it, react to it, comment, share it, or watch it through. Those predictions get built from behaviour the system has already observed on that piece of content and on content like it.

A three-second play is a weak signal by design. It happens while somebody is still deciding, and it happens to a large share of everything that passes through a feed. It tells the ranking system that the post was served and that playback started. It is genuine information, and it is thin.

Sixty seconds of watching is a strong signal. It happened after a decision, it cost the person a minute of attention they could have spent on the next post, and it is comparatively rare. Content that produces a lot of it gets shown to more people, because the system's estimate of "this will hold attention" has been confirmed by evidence rather than inferred from a thumbnail.

This is why two videos with identical three-second counts can end up with completely different lifetimes. The one that kept people watching keeps getting served for days. The one that collected fast plays and lost the room finishes its run quickly. The three-second number describes the start of the video's life; the sixty-second number describes how long that life lasts.

There is a second effect worth knowing. Watch time is also how Facebook learns who your audience is. When a video holds a particular kind of viewer, the system starts looking for more people who resemble them. Retention shapes the audience you get served to next, not only the size of it.

Where each metric lives in Meta Business Suite

All of this is visible without any third-party tool. The path changed a few times over the years, so here is where the numbers actually sit today.

The video list

Open Meta Business Suite, go to Insights, then Content, and filter to videos or Reels. You get a table with one row per post. The default columns lean on reach and reactions, and the column picker is where the interesting fields are: 3-second views, 1-minute views, minutes viewed, average watch time, and reach. Add those four and sort by minutes viewed rather than by views. The ordering usually looks different from what you expected, and that difference is the point of the exercise.

The single-video view

Click any row and you get the per-video breakdown. This screen carries the numbers that matter for a single upload: total plays, three-second and one-minute counts side by side, average watch time, total minutes viewed, and the split between followers and non-followers. The follower split is worth a moment on its own, because it tells you whether the video travelled beyond the people who already know you.

The retention curve

The same screen carries the audience retention graph: what share of viewers were still watching at each second. This is the single most instructive thing in the whole panel, and it is also the one most Page managers scroll past. A curve reads like a transcript of the video's weak points, second by second.

Which number should a Page manager watch?

Watch all of them, and give each one a job.

Reach answers whether the post was distributed. If reach is low, the problem happened before anyone saw the video, and no amount of editing will fix it. That is a posting-frequency, format, or account-health question.

3-second views divided by reach answers whether the first frame is working. If reach is healthy and three-second views are thin, the thumbnail and the opening second are where to spend your next hour.

1-minute views divided by 3-second views answers whether the body of the video is working. This ratio is the honest one. It is unaffected by how much distribution the post got, so it stays comparable between a video that reached two hundred people and one that reached two hundred thousand.

Average watch time answers how far people actually got, in seconds. Compare it to the length of the video rather than to other videos — twenty seconds on a thirty-second clip is a strong result and twenty seconds on a ten-minute upload is a different story entirely.

Minutes viewed is the total, and the number that accumulates toward monetisation. It is the one to track monthly rather than per post.

If you only have room for two, take the one-minute-to-three-second ratio and average watch time. Those two together describe the health of the content itself, separately from how many feeds it happened to land in.

Reels are a separate surface with separate rules

A Reel posted to your Page and a video posted to your Page are handled by different parts of the system. Treating them as one thing is the most common reason a Page's numbers look inconsistent from week to week.

How a Reel finds an audience

Feed video is served primarily to people connected to your Page, then outward as engagement justifies it. Reels start from the opposite direction. The Reels surface is a recommendation feed built for people who do not follow you, so a Reel's audience is dominated by strangers from the first hour. That is why a Page with three hundred followers can put up a Reel that reaches forty thousand people, and why the same Page's feed videos stay in the hundreds.

It also means the follower split on a Reel usually looks lopsided toward non-followers, and that is the surface working as intended rather than a problem to solve.

How a Reel counts a view

Reels play in a full-screen, sound-on environment and they loop. A replay is counted, so short Reels routinely finish with view counts larger than the number of people who saw them. This makes raw view counts on Reels and on feed videos genuinely non-comparable, and it makes average watch time the metric to carry across.

Because Reels loop, watch time above one hundred percent of the video length is a real and achievable result. A twelve-second Reel with an average watch time of eighteen seconds means people were watching it a second time, which is one of the strongest signals available anywhere on the platform.

What carries across from feed video

The opening principle carries across completely: the first moment decides everything downstream. Retention carries across as the governing metric. Sound behaviour does not carry across, since the Reels player opens with audio on and the feed player usually opens muted. Length does not carry across either, because a Reel is competing against a swipe that costs the viewer nothing.

The practical approach for most Pages is to make Reels the discovery engine and feed video the depth. Reels bring new people to the Page; longer feed video gives the ones who arrived something to stay for, and builds the minutes-viewed total that monetisation is measured in.

Sound-on view rate, and what it is telling you

Facebook measures how many of your views happened with audio playing, and reports it separately from everything else. It exists because the feed autoplays muted by default on most devices, so turning the sound on is a deliberate act.

That makes the sound-on rate a proxy for intent. A person who unmutes has stopped scrolling and committed. On a video where a large share of views carry sound, you can usually assume the one-minute count is healthy too, because the two behaviours travel together.

The number also tells you something about the video's construction. If your sound-on rate is low and the video depends on narration to make sense, the muted majority is watching something incomprehensible. That is a captions problem, and it is the cheapest fix available on this list.

If the sound-on rate is high, the audio itself is doing work: a voice, music that fits the cut, or an opening line worth hearing. It is worth checking which of your videos earn it, because that pattern is usually repeatable.

One practical note: Reels open with sound on, so sound-on rates are naturally high there and low on feed video. Compare Reels to Reels and feed video to feed video, and the number becomes useful.

The first three seconds decide the sixtieth

Everything in the retention curve is downstream of the opening. If you want one place to spend effort, it is here.

Open on the moment, not on the setup. The single most common structure in an underperforming video is thirty seconds of context before the thing the video is about. Cut the runway. Start where a person watching for the second time would start.

Put a face or motion in the first frame. Feed video autoplays, so the first frame is doing thumbnail duty whether you designed it that way or not. Movement in the first half second is what separates a post that plays from a post that scrolls past.

Say what the video is in the first sentence. "Here is how to fix a dripping tap in four minutes" retains better than four minutes of fixing a tap with no framing, because the viewer knows what they signed up for and can decide to stay.

Put a caption on screen immediately. The muted default means the first thing many people receive is text. One line, large, on screen at second zero.

Take out the intro animation. A branded three-second sting sits exactly where the three-second count is decided. Move the branding to the end, or into a corner watermark.

How long should a Facebook video be?

Length follows the job the video is doing.

Under sixty seconds works for Reels and for feed clips whose whole purpose is a single idea. These build reach and follower growth. They contribute little to the minutes-viewed total, since a forty-second video can only ever produce forty seconds per viewer.

One to three minutes is the practical middle for feed video. It is long enough to clear the sixty-second threshold with room to spare, short enough that a good opening carries most viewers to the end.

Three minutes and above is where in-stream ad placements become available and where the minutes-viewed total genuinely accumulates. It asks more of the content: a reason to stay that renews itself every thirty seconds or so.

The useful discipline is to pick the length after the edit rather than before it. Cut the video to the point where every remaining second earns its place, then look at what you have. A tight two minutes outperforms a padded six almost every time, because the padded version teaches the ranking system that people leave your videos.

One more thing worth saying plainly: length is not a lever on its own. Extending a video to clear a threshold produces a long video with poor retention, and retention is the thing being measured. Longer works when there is more to watch.

Format: ratio, captions and the silent first impression

Aspect ratio. Vertical 9:16 for Reels, and 4:5 or 1:1 for feed video. The reason is screen real estate: a 16:9 landscape video occupies a narrow band on a phone, and the amount of thumb-space a post owns correlates directly with how long people look at it.

Captions. Burn them in or use Facebook's auto-generated captions and correct them. Muted autoplay is the default state, so captions are the difference between a comprehensible video and a moving picture. This one change tends to move the one-minute count more than anything else on this page.

Resolution and upload. Upload the highest quality file you have and let Facebook compress it. Upload natively rather than sharing a link to the video elsewhere; the native player is what gets the distribution, and a link post is treated as a link.

Thumbnail. Choose a custom one rather than accepting the auto-selected frame. It appears in shares, in search, and on your Page's video tab, and it keeps working long after the post has left the feed.

The first frame. Since feed video autoplays, the opening frame and the thumbnail have different jobs. Design both.

What does this tool read?

This page reads the public profile you enter and scores it against the bands a healthy account sits in, so you can see which number is holding the account back. Everything it uses is public and a password is never requested.

What gets read is limited to what the platform serves publicly: display name, picture, follower count, the number of posts, verification status, and the engagement pattern visible across recent activity. From those it produces a reading of where the account stands right now, which is the useful thing to have in hand before deciding what to change.

Where it helps most on video is context. A view count means one thing on a Page with two hundred followers and something completely different on a Page with forty thousand. The scoring puts your numbers next to the ranges accounts of your size normally occupy, so you can tell whether the video underperformed or the Page is simply small at the moment.

For views on the videos themselves, our packages are the route, and they are described further down this page.

How to read your retention curve

The retention graph shows the share of viewers still present at each second. Four shapes cover almost everything you will see.

A cliff in the first three seconds. The most common shape by far. The opening is not delivering on whatever the thumbnail or the first frame promised. Rebuild the first five seconds and leave the rest of the video alone.

A steady, gentle decline. This is a healthy curve. People are leaving at the ordinary rate, which is what a good video looks like. The lever here is length: find where the line crosses fifty percent and consider whether the video needed to run much past it.

A sharp step partway through. Something specific happened at that timestamp. Go and watch it. It is usually a change of subject, a long pause, a sponsor read, or the moment the answer arrives and the reason to stay disappears.

A bump upward. People rewound, or the video looped. Whatever sits at that second is your best material, and it belongs at the beginning of the next video.

Look at three curves from the same week rather than one. Individual videos are noisy, and the pattern across a few of them is where the instruction actually is.

Watch time as the goal, views as the signal

It is worth setting the objective explicitly, because the two counters push toward different content.

Optimising for views produces short, sharp, high-turnover clips. That is a legitimate strategy — it grows the Page, it brings new people in, and it works well on the Reels surface.

Optimising for watch time produces longer, structured pieces with a reason to stay. That builds minutes viewed, unlocks in-stream placements, and creates the kind of audience that comes back for the next one.

Most Pages want both, in a fixed proportion. A workable pattern is three or four short Reels a week for reach, plus one longer feed video that gives the arriving audience something with substance. The Reels feed the Page; the longer video is why anyone stays subscribed to it.

Track them on different clocks. Views per post, week by week. Minutes viewed as a monthly total. Mixing the two timescales is how Pages end up chasing a single viral clip instead of building a pattern.

Posting time and frequency

Timing matters less than it used to, and it still matters a little. Facebook keeps distributing a video for days, so a post that starts slowly can pick up later; the first hour is an accelerant rather than a verdict.

Meta Business Suite shows when your own followers are online, and that chart is worth more than any published best-time list, because it describes your audience rather than an average of everyone's. Post inside those windows and stop optimising further.

Frequency does real work. Three to five posts a week gives the system enough data to learn what your Page is about and enough surface area for one of them to catch. One post a week makes every upload a gamble.

Spacing matters too. Two videos published an hour apart compete with each other for the same feed slots. Leave a day between video posts where you can.

Consistency beats volume. A Page that publishes three times a week for two months outperforms a Page that publishes twenty times in one week and then goes quiet, because the audience learns a rhythm and the system learns a pattern.

Cross-posting between Facebook and Instagram

Meta makes it easy to publish once and appear in both places, and used carefully it roughly doubles the return on an edit.

The mechanics: link the accounts in Business Suite, and the composer offers to publish a Reel to Instagram at the same time. Both surfaces accept the same 9:16 file, so no extra work is required in the edit.

Two habits make it work better. First, upload the clean file to each platform rather than exporting one with the other platform's watermark on it — a watermarked re-upload is treated as recycled material and gets served accordingly. Second, write the caption separately for each. The two audiences behave differently, and the caption is where that difference is cheap to accommodate.

Keep the reporting separate as well. Instagram and Facebook count views under their own definitions, so adding the two figures together produces a number that does not describe anything. Read each platform's insights on its own terms.

Why views level off, and what moves them again

The opening frame is a title card. Three seconds of branding sits exactly where the three-second count is decided. Move it to the end.

The video is landscape. On a phone that is a thin strip in the middle of the screen. Re-crop to 4:5 and the same content occupies twice the space.

There are no captions. Muted autoplay means a narration-led video without text is asking a lot of the viewer. Add captions and re-measure.

Everything is one length. A Page that only posts sixty-second clips accumulates views and few minutes. Adding one longer piece a week changes the shape of the totals.

The Page posts in bursts. Four videos on Sunday and nothing until the following Sunday gives the system no rhythm to work with. Spread them.

Reels and feed video are being compared. They count differently and they reach differently. Judge each against its own history.

The Page is small enough that variance dominates. Under a few thousand followers, individual videos swing wildly for reasons that are not about quality. Read four weeks of data, not four posts.

A realistic first month

Week one. Set the Page itself in order: category, a one-line description, a cover image, and a pinned video that is your best work. Open Business Suite, add the 3-second, 1-minute, minutes-viewed and average-watch-time columns, and record where your last ten videos landed. That is your baseline, and everything after this is measured against it.

Week two. Publish three Reels and one feed video of ninety seconds or longer. Captions on all four. Open the retention curve for each one two days after posting and write down where the line drops.

Week three. Keep the schedule. Take the two videos with the best one-minute-to-three-second ratio and work out what they had in common — the opening, the subject, the pacing. Make three more like them. Change the opening five seconds of your usual format based on what the curves showed you.

Week four. Keep the schedule. Compare the month's minutes viewed to the baseline you recorded in week one. That comparison, not any single post's view count, is the thing that tells you whether the month worked.

Four weeks is enough to see a direction and not enough to see a result. Run the same loop for three months and the direction becomes a trend.

Buying Facebook video views

We sell Facebook video views, and it is worth being precise about the job they do.

A video showing 31 views reads as something nobody watched, and people scroll past videos nobody watched. The count sits under the post before a single second has been played, which means it is already shaping the decision to press play. That is the specific job: social proof at the moment of the choice, on a video where the content is good and the post is young.

Delivery starts within minutes from a public video link, no password is involved, it is priced per thousand, and it carries a 30-day refill guarantee.

They work best alongside the work described above. The view count buys the click; the first three seconds keep the person; the next fifty-seven build the watch time that decides how far the video travels. Each part hands off to the next, and the video that has all three is the one that runs for days.

Questions people ask

Does Facebook count my own views of my video?

Yes, plays from your own account register like any other. On a video with a handful of views this is visible in the numbers, so it is worth being aware of when reading a fresh post.

Why does the view count change after a few hours?

Insights figures settle over roughly a day as the platform reconciles playback records and removes duplicate and automated activity. Reading a number in the first hour and again the next morning will often give you two different figures, and the later one is the real one.

Do Reels views and video views appear in the same total?

They are recorded on separate surfaces and it is worth reading them separately. Business Suite lets you filter by content type, and doing so is what makes a month-over-month comparison meaningful.

Does a share help more than a like?

Considerably. A share puts the video into a new feed, so it produces both a stronger ranking signal and a fresh set of viewers who have never seen your Page. It is the single most valuable action a viewer can take.

How many views make a video successful?

Compare against your own Page rather than any published benchmark. A video that reaches twice your follower count and holds an average watch time above half its length is performing well at any size.

The short version

Facebook counts a view at three seconds and counts a separate one at sixty, and the second number is the one that decides how far the video travels. The three-second count reports reach and opening frame; the one-minute count reports whether the video held the room. Reels are a different surface with different behaviour, sound-on rate is a measure of intent, and captions are the cheapest fix on the list. Add the right columns in Meta Business Suite, read the retention curve rather than the view total, and judge the month by minutes viewed. Everything else on this page is detail hanging off those two numbers.