Facebook Video Views

What counts as a view on a Facebook video?

Less watching than most people assume. Facebook's public video counter has long tripped after a watch of about three seconds, which is roughly the time it takes to scroll past something, notice it and change your mind.

That is a deliberate choice rather than an accident. A feed autoplays video as it passes, so a bar set at a few seconds captures the moment somebody stopped, and a bar set at thirty seconds would capture almost nobody.

The consequence is that a view on Facebook is a much lighter event than the word suggests. It records attention starting, not attention held, and those are two different things with very different worth.

Understanding that one detail is most of what you need before ordering. Everything else on this page follows from the gap between what the counter says and what actually happened.

Why does Facebook publish more than one view number?

Because one number cannot answer both questions people ask about a video. How many people started it and how many people watched it are separate facts, and Facebook measures both.

So there is a loose public figure and a set of stricter private ones. The public counter takes the low bar described above; alongside it Facebook reports things like one-minute views and ThruPlay, which require considerably more watching.

Average watch time and a retention curve sit with them. Those show how long the average person stayed and where in the video people left, which is a level of detail the public number cannot carry.

Meta renames and reorganises these regularly, so the labels move around between interface revisions. What does not move is the structure: one generous number on display, several demanding ones underneath, and the gap between them doing all the work described on this page.

Which number does the public see?

The loose one, under the video, where anybody scrolling past reads it. That is the figure a visitor forms an impression from, and it is the only one they have access to.

Nobody outside the page can see the watch time. A stranger cannot tell whether your fifty thousand views were fifty thousand people stopping for three seconds or fifty thousand people watching to the end, and Facebook gives them no way to find out.

That asymmetry is what makes this product coherent. The number that shapes what a viewer thinks is public and loose, which is exactly the number an order moves.

It is also why the honest sections of this page exist. A product that moves the visible number and not the private ones is useful for one purpose and useless for another, and the difference is worth being clear about before paying rather than after.

Which numbers actually matter?

The private ones, for anything that involves the algorithm or money. Facebook's video ranking is built around how long people watch rather than how many started, so watch time and completion are the figures with consequences.

The same is true of anything paid. A brand assessing a page, or the monetisation requirements Facebook publishes for video, looks at minutes watched rather than at the counter under the post.

The public number matters for exactly one thing, and it is not a small thing: what a human being thinks when they land on the video. Social proof is real and it operates before anybody watches anything, which is why the loose number is worth something despite being loose.

So the two sets of numbers serve two audiences. The counter talks to visitors and the watch metrics talk to Facebook, and an order here speaks fluently to the first and not at all to the second.

What is ThruPlay?

Facebook's name for a video watched to the end, or for around fifteen seconds of it when the video runs longer than that. It is the strictest of the commonly quoted figures and the one advertisers pay against.

The reason it exists is that the loose counter proved useless for buying decisions. An advertiser paying per view wanted to pay for watching rather than for scrolling, and ThruPlay is the metric built to answer that.

For an organic video it is still the most useful single number you have. A video with a high ThruPlay rate held people; a video with a high view count and a low ThruPlay rate stopped people for a moment and lost them.

Nothing in this order changes it. ThruPlay counts sustained watching, and that is not what a delivery produces, which is the plainest way to state the limit of this product, and the limit holds however large the order is.

Where do you find the other figures?

In the page's own tools, under the video's performance view. Open the video from the page rather than from the feed and the watch metrics are there alongside the public count.

The retention curve is the part worth learning to read. It shows the share of viewers still present at each second, and the shape of the drop tells you more about the video than any single figure does.

A cliff in the first seconds means the opening is not holding anybody. A gentle slope means people are staying and leaving gradually, which is what a video that works looks like.

Reading that curve before ordering is worth the five minutes. It tells you whether the video is one where more arrivals would help, or one where more arrivals would simply produce a taller cliff.

Does the view count help distribution?

Less than people hope, and this is the honest answer rather than the convenient one. Facebook decides what to show using how people behave with a video, and a counter is not behaviour.

What the counter does is influence the humans who see it, and that has a real second-order effect. A video with a substantial count gets stopped for by more of the people it reaches, and stopping is behaviour that Facebook does read.

So the honest chain is indirect: the number changes what viewers do, and what viewers do is what the system measures. Anybody claiming the count feeds ranking directly is describing a machine nobody outside Meta can see.

The practical version is that this helps a video that is already good and does very little for one that is not. That is not a disclaimer; it is how the mechanism works, and knowing it saves money on the wrong video.

How many views should a video have?

More than the page's follower count on a good video and a fraction of it on an ordinary one, which is a wide range because Facebook video genuinely behaves that way. A video can be shown far beyond the people who follow the page.

That makes the ratio less rigid here than on most products. Unlike a follower count or a like count, a view figure that exceeds the audience is completely ordinary and nobody reads it as strange.

What does read as strange is views without anything else. A video with a large count and no reactions, no comments and no shares describes an audience that arrived and felt nothing, which is a specific and unflattering claim.

Your own back catalogue is the guide. Open the last ten videos, note the views and the reactions on each, and the relationship between them on your page is what a new video is compared against.

How many should I buy?

Enough to change what a visitor thinks and not so much that the video stands alone in your catalogue. For most pages that means 500 or 1,000 rather than 100,000 — the round numbers people search for are not the ones that suit a small page. A page whose videos all sit in the hundreds and one that sits in the hundreds of thousands invites the comparison.

Sizing against reactions is the constraint that actually binds. The proportion of viewers who react is fairly stable for any given page, so a view count that races ahead of the reaction count is the pairing somebody notices.

Ordering a smaller number and a few reactions alongside it usually reads better than ordering a large number alone. That is not an upsell, it is the arithmetic of the ratio people actually look at.

And there is a ceiling worth respecting that is specific to this product. Every view you add divides the same total watch seconds, so a very large order does visible damage to the average — which the next section is about.

What happens to average watch time?

It falls, and this is the one thing about this product that nobody should discover after the fact. Average watch time is total seconds divided by total views, and an order grows the denominator without growing the numerator much.

So the arithmetic runs the wrong way by design. A video averaging twelve seconds across a thousand views will average considerably less across five thousand, and the drop is proportional to how large the order was.

That matters because average watch time is one of the figures Facebook's ranking actually reads. This product improves the number a visitor sees and worsens the number the system uses, at the same time, from the same order.

Which is why the sensible order is modest and the sensible target is a video that has already held people. A strong retention curve absorbs a small order without much change; a weak one gets weaker.

What do you need to buy views for a Facebook video?

The public link to the video. Open it, use the share menu, copy the link and paste it into the form — that is the whole requirement, and no password, login or page access is part of it.

The video has to be public while the delivery runs. A video limited to friends, to a specific audience or to a private group cannot receive a delivery, and changing the audience mid-order stops it where it is.

Copy the link from the video itself rather than from the page. A link to the page delivers nothing because there is no single video at that address, and this is the most common correction an order needs before it starts.

Everything else about the video is unaffected. Reactions, comments, shares and the description carry on as they were, and the order moves the view counter and nothing beside it.

Does this work on Reels and on live videos?

On Reels yes, and the counter behaves the same way, though Facebook distributes Reels differently from feed video. A Reel is shown to people who do not follow the page, so its own numbers tend to run on a different scale from the rest of the catalogue.

A live broadcast is a different case while it is running. Concurrent viewers during a broadcast is a separate figure from views on the recording afterwards, and the two are not interchangeable.

Once a live video has ended it behaves like any other video on the page, so buying Facebook live video views after the broadcast is the same order as any other. The recording carries a view counter that works the same way, and an order placed after the broadcast is an ordinary order.

A cross-posted video is worth a thought before ordering. If the same video exists on several pages, each copy carries its own counter, and a delivery moves only the one whose link you paste.

How fast should they arrive?

Front-loaded but not instant, because that is how a video actually behaves. Facebook shows a video most heavily in its first day or two, so a delivery that climbs quickly and then eases matches the curve the video would have had.

Spreading a delivery over too long looks stranger than compressing it. Video views arriving steadily for two weeks describes something no organic video does, since attention on a post decays rather than holding flat.

Instant Facebook video views have a real use when the video is about to be linked somewhere. If it is going into an email, an ad landing page or a message to a client this afternoon, a faster curve is a reasonable trade.

What no pacing changes is the watch-time arithmetic. The average falls the same amount whether the views arrive in an hour or a week, so pacing helps the appearance and not the underlying figure.

How much do Facebook video views cost?

Priced per thousand with the rate falling as the order grows, and the full ladder is in the picker above before anything is confirmed. Cheap Facebook video views are the norm rather than a warning sign here: views are among the cheapest things in the catalogue and the reason is the low bar described at the top.

A view asks for a few seconds of playback and nothing else. Nobody decides anything, nothing is published, nothing persists on any account, and that lightness is what keeps the unit price down and lets the tiers reach large quantities.

The cheapness is precisely why the sizing advice on this page pushes the other way. When a product is cheap per unit, the temptation is to order a lot of it, and this is the one product where ordering a lot actively damages a metric you care about.

The smallest tier is enough to see the whole shape. One order on one video, then the performance view the next day, shows the counter rise and the average fall in the same screen, which is the fastest way to understand this product.

Do the views stay?

Generally, and more reliably than engagement counts do. A view records something that already happened rather than an action an account is holding, so there is nothing for anybody to withdraw.

Our window runs 30 days from delivery. If the count falls below the level ordered inside that period, the shortfall is delivered again at no cost, and the order number is all that is needed to arrange it.

Facebook does reconcile video figures after the fact, and small movement in the days after a video goes up is ordinary. It happens to videos nobody ordered anything for, which is worth knowing before reading a dip as a failure.

Anybody promising a number that can never move is describing a system they do not run. The promise worth holding is the one that tops up a shortfall, and a month covers the period in which a correction would arrive.

Is it safe for the page?

On the security side the answer is clean: no credentials are involved at any point. The order needs a public video link, so the routes that actually cost people their pages — a shared password, a connected app, an admin role given to a stranger, a login on a page that resembles Facebook — stay shut throughout.

Nothing is added to the page's administration. No account is promoted, no app is installed, and nobody gains the ability to post, edit or change settings.

On the platform side, Meta's terms ask for authentic engagement and the company removes automated accounts in sweeps run on its own schedule. When a sweep lands, views attributable to those accounts are recalculated across the platform, which is ordinary maintenance rather than a fault in one order.

Where monetisation is involved the sensible advice is more cautious. Facebook pays on watch time, so a page in or near a monetisation programme has a real reason to keep its watch metrics clean and to keep any order small.

Can I see the delivery before paying for it?

Partly. The free Facebook comments tool sends a small delivery to a post of yours, which shows how quickly an order starts and how arrival behaves on a real page before anything is spent.

What that answers is the mechanical half — queue time, pacing, whether a delivery holds through the week — and that half behaves the same way whichever counter is being moved.

What it cannot show is the effect specific to this product, because a comment does not touch watch time and a view does. Only a small paid order on one of your own videos shows the counter and the average moving together.

Use an older video for that test rather than the one you care about. The information is identical and it leaves the video that matters free for a considered order.

How does this differ from ordering followers or comments?

A follower attaches to the page and a view attaches to one video, which is the first and largest difference. One is a standing relationship and the other is a single event on a single post.

If the goal is a page that looks established before anybody watches anything, Facebook followers is the product that speaks to that, and it sets the size of the audience your posts are eligible to reach.

If the goal is a video that looks discussed rather than merely watched, Facebook comments is the one, and it moves a counter that carries text rather than a number.

Views are for the video you want stopped for. That is a narrower job than the other two do, and it is the only one of the three where ordering more can make a metric worse.

What the counter carries, and what the video keeps

It carries the reason somebody stops instead of scrolling. A visitor deciding whether a video is worth three seconds has almost nothing to go on, and the number under it is the largest thing available.

That is a real job and it is worth paying for on the right video. Social proof works before anybody has watched anything, which is precisely the moment a video most needs help.

What stays with the video is everything the counter cannot supply: whether the first seconds hold anybody, whether the middle earns the end, whether anyone would send it to somebody else. Those are the things Facebook actually measures, and no order touches them. The counter opens the door; what happens after it is the video alone.

Which is why the useful order here is modest, placed on a video that already holds people, and sized with the watch-time arithmetic in mind. The Facebook services overview sets out how the other numbers on a page relate to this one.