ThruPlay: what it is and when to use it as your objective

A ThruPlay is counted when someone watches your video to the end if it runs under 15 seconds, or for at least 15 seconds if it runs longer. It is Meta's answer to the old three-second view, which counted almost everyone who scrolled past. ThruPlay is a stricter bar, and it is a reasonable measure of whether your message was actually delivered. It is not a measure of whether anyone wanted what you were selling.

Formula

Cost per ThruPlay = ad spend ÷ ThruPlays.

Say you spend US$ 300 and record 6,000 ThruPlays. That is US$ 0.05 per play. Take the count from the platform; do not try to reconstruct it from impressions or watch-time percentages.

Two things the formula hides:

• ThruPlays are plays, not people. The same viewer can produce several, so this number never behaves like reach.
• The bar depends on video length. A 12-second ad has to be finished to count; a 60-second ad only needs its first 15 seconds. Comparing their cost per ThruPlay directly compares two different achievements.

What a good number looks like

Judge it in two steps, and only ever against videos of similar length and role.

A cheap ThruPlay by itself is not a result. Delivery costs fall in low-intent placements and among people who watch a lot of video, so the cheapest plays often come from the audience least likely to convert. If cost per play drops while CPL rises, the video objective is buying you watch time and nothing else.

A video objective earns its place in narrow cases: launching a new message you need explained before anyone will convert, building a warm audience you will retarget later, or testing which of several hooks holds attention. Outside those, optimize for the action you sell and let ThruPlay be a diagnostic column.

What makes the number move

• Video length, since the qualifying bar changes with it.
• The first two seconds. Most drop-off happens before anyone reaches 15.
• Sound-off legibility. Captions and on-screen text change completion more than production quality does.
• Placement. Reels and Stories behave nothing like in-feed video, and a mixed report averages them into a number that describes neither.
• The optimization event itself, which decides who gets shown the ad in the first place.

Decide what the video is for, then pick the column

Write down the job of the video before you launch it. If the job is comprehension, cost per ThruPlay is a fair scoreboard. If the job is revenue, judge it on CPA and treat plays as context. The paid media reporting page shows how to keep both readings side by side.