What is cost per completed view, and how should buyers read it?
Part of Programmatic media planning
The short answer
Cost per completed view, or CPCV, is video media cost divided by the number of times the ad played to its end. A complete is counted when the player fires the VAST complete event, meaning 100 percent of the creative played at normal speed. CPCV falls as completion rate rises, so it depends heavily on placement, autoplay and skip settings.
Key takeaways
- CPCV is cost divided by complete views, and a complete means the player reported the whole creative played at normal speed.
- At a fixed CPM, CPCV equals the CPM divided by 1,000 times the completion rate, so a higher completion rate lowers CPCV with no change in price.
- Instream placements carry user intent to watch content; outstream placements play beside text and can stop when they leave the screen, so their completion rates are not comparable.
- A complete is not a viewable or audible view: completion, viewability and sound are separate measures.
- CPCV works as a secondary efficiency check inside one placement type, not as the single goal of a video plan.
What is cost per completed view?
Definition
- Cost per completed view (CPCV)
- The average cost of each video ad impression that played to the end, calculated as video cost divided by complete views.
Google's Display & Video 360 metrics reference (opens in a new tab) shows that even one platform has several CPCVs. "Video Media Cost eCPCV" is "The average cost of each complete video view, based on the Media Cost spent." "Total Video Media Cost eCPCV" uses total media cost, which adds data fees and partner costs to the raw media cost, and "Video Revenue eCPCV" uses revenue, calculated from the revenue model set on the line items. Two reports can show different CPCVs for the same flight simply because they divide different costs.
An eCPCV is computed after the fact from cost and completes, even when the media was bought by the thousand impressions. The plan should name which cost sits in the numerator, so that the CPCV the client sees matches the CPCV the buyer optimizes.
This page covers open web video bought programmatically. YouTube uses its own view definition and bidding, explained on the YouTube cost per view page; a YouTube view is not a VAST complete, and the two should not be added together.
How is a completed view counted?
Open web video ads are delivered through VAST, the IAB Tech Lab's ad serving template, and the player reports progress by calling tracking URLs. VAST 4.3 (opens in a new tab), released December 2022, defines the events in section 3.14.1. Its definition of complete is short: "The creative was played to the end at normal speed so that 100% of the creative was played."
| Event | VAST 4.3 definition | Report metric it feeds |
|---|---|---|
| start | "an individual creative within the ad was loaded and playback began" | Starts (Video) |
| firstQuartile | "played continuously for at least 25% of the total duration at normal speed" | First quartile views |
| midpoint | "played continuously for at least 50% of the total duration at normal speed" | Midpoint views |
| thirdQuartile | "played continuously for at least 75% of the duration at normal speed" | Third quartile views |
| complete | "played to the end at normal speed so that 100% of the creative was played" | Complete Views (Video) |
| skip | "the user activated a skip control to skip the creative" | Skips (Video) |
Display & Video 360 mirrors those events in its reports. "Complete Views (Video)" is "The number of times a video has been played to its end," and quartile metrics count plays to 25, 50 and 75 percent of the ad's length. Because the counts come from the player, a complete depends on the player firing the event correctly. A player that fails to call the complete URL undercounts; one that fires it early overcounts.
How does completion rate relate to CPCV?
Google defines "Completion Rate (Video)" as "The percentage of video impressions that played to completion." When the cost and the impressions come from the same report, CPCV and completion rate are tied by simple arithmetic: CPCV equals the CPM divided by 1,000 times the completion rate. The price per thousand impressions and the share that finish together set the cost of each complete.
| CPM paid | Completion rate | Completes per 1,000 impressions | CPCV |
|---|---|---|---|
| $20.00 | 20 percent | 200 | $0.100 |
| $20.00 | 40 percent | 400 | $0.050 |
| $20.00 | 80 percent | 800 | $0.025 |
| $28.00 | 80 percent | 800 | $0.035 |
The table shows why CPCV can mislead. A placement with a higher CPM and a high completion rate can produce a lower CPCV than a cheap placement where few ads finish, and a placement where ads finish because nobody can stop them can look the cheapest of all. CPCV rewards completion, so anything that inflates completion, such as forced duration or a player playing off screen, also lowers CPCV.
Reading a CPCV figure in four steps
- Name the cost. Confirm whether the figure uses media cost, total media cost or revenue.
- Check the denominator. Confirm completes and impressions come from the same report, the same dates and the same line items.
- Split by placement type. Break CPCV out by instream and outstream before comparing sites or deals.
- Pair it with viewability and starts. A low CPCV with low viewable completes or few starts is a reason to look closer, not to scale up.
Why do outstream and instream completion rates differ?
The difference starts in how the industry defines the placements. The IAB Tech Lab's AdCOM specification (opens in a new tab) says instream video "must be set to 'sound on' by default at player start, or have explicitly clear user intent to watch the video content," with the video content as "the focus of the user's visit." No Content/Standalone video, by contrast, is played "without streaming video content," in placements "like slideshows, native feeds, in-content or sticky/floating." The outstream video ads page covers those placement values in depth.
| Placement | How it plays | What pushes completion up or down |
|---|---|---|
| Instream (pre, mid or post roll) | Before, during or after video the person chose to watch | The person is waiting for content; skip controls, where offered, reduce completes |
| Accompanying content | Player between paragraphs that "starts playing only when it enters the viewport" | Playback can end when the reader scrolls on, unless the player floats |
| In-article and in-feed outstream | Google: ads "between the paragraphs of pages" or "in a scrollable stream of content" | The reader came for text, so the ad competes with reading and scrolling |
| Interstitial | Plays without content; AdCOM says it takes "the majority of the viewport and cannot be scrolled out of view" | Completion rises because the ad cannot be scrolled away, which does not prove interest |
AdCOM also defines how playback ends, through its playback cessation modes: "On Video Completion or when Terminated by User", "On Leaving Viewport or when Terminated by User", or continuing "as a Floating/Slider Unit until Video Completion." A player that stops on leaving the viewport loses the complete whenever the reader scrolls past; a floating player keeps playing in a corner. The two can carry the same label in a report and very different completion rates.
The IAB and MRC video impression guidelines (opens in a new tab) of June 2018 ask that auto-play and forced duration be disclosed. They define forced duration as the portion of a video ad "during which a user cannot skip the ad," and for material levels require reporting the share of impressions from auto-play, click to play and unknown, noting that "assuming something is not Auto-Play just because it wasn't signaled as such is a false assumption."
Should CPCV be a primary KPI?
CPCV fits as a secondary efficiency measure inside one placement type, where the conditions behind each complete are comparable. As the primary goal of a mixed plan, it can push budget toward whatever finishes most cheaply, which may be the placement least able to be stopped rather than the one most watched. A plan that buys a full message, such as a 30 second story, is better served by a goal that combines completion with viewability and sound, and by a business outcome measured separately.
Before CPCV goes into a plan or a deal
- The plan names the cost basis for CPCV: media cost, total media cost or revenue.
- CPCV targets are set separately for instream and outstream placements.
- Sellers have disclosed auto-play, sound and forced duration settings for each placement.
- Viewable completes or completes with sound on are reported next to CPCV where available.
- Creative length is fixed across the comparison, since a 15 second and a 30 second ad do not finish at the same rate.
- The YouTube portion of the plan is reported with its own view metric, not merged into CPCV.
How does PlacementGraph look at video and ad layout on a site?
What does a CPCV review look like for a regional advertiser?
Fictional example
A furniture maker's open web video review
An unnamed solid wood furniture maker with showrooms in Birmingham, Huntsville and Mobile runs a 30 second video on home design, regional lifestyle and local news sites in Alabama. The buy includes instream placements on two publishers with their own video programming and outstream in-article placements on 20 smaller sites.
The client asks why the outstream sites show a higher CPCV than the instream publishers. The buyer splits the report by placement type before answering, confirms that both figures use media cost, and asks each outstream seller whether its player stops on leaving the viewport or floats. Two sellers report floating players with auto-play and no sound by default.
The buyer sets separate CPCV references for instream and outstream, adds viewable completes to the report, and keeps the floating player sites on the list but labels their completes as auto-play in the client summary. No site is dropped for CPCV alone.
Frequently asked questions
Is CPCV the same as cost per view?
No. CPCV divides cost by completes, where the whole creative played. Cost per view depends on the platform's view definition, which can count a view before the end of the ad. Compare the two only after checking what each platform counts as a view.
Does a skippable ad lower completion rate?
It can. VAST 4.3 defines a skip event for when the person activates a skip control, and a skipped ad does not reach the complete event. AdCOM lets placements signal whether they impose skippability and the creative length above which skipping is allowed, so buyers can separate skippable from non skippable supply.
Do completes count when the sound is off?
Yes. The VAST complete event only says the creative played to the end at normal speed. VAST 4.3 offers macros for auto-play and muted states, and the IAB and MRC video guidelines ask that auto-play be disclosed, so sound has to be checked as a separate signal.
Should CPCV be compared across different creative lengths?
No. A complete on a 15 second ad and a complete on a 30 second ad describe different amounts of playback, since VAST counts a complete only when 100 percent of the creative played. Compare CPCV within one creative length, or report quartile completion for the longer ad next to it.
Sources
- Metrics in reports (opens in a new tab)Google, Display & Video 360 Help
- Video Ad Serving Template (VAST) 4.3 (opens in a new tab)IAB Tech Lab
- AdCOM v1.0 (opens in a new tab)IAB Tech Lab
- Position targeting (opens in a new tab)Google, Display & Video 360 Help
- Digital Video Impression Measurement Guidelines (June 2018) (opens in a new tab)IAB and Media Rating Council
- Viewable Ad Impression Measurement Guidelines v2.0 (opens in a new tab)Media Rating Council
Researched and written by PlacementGraph Research. Read how we research, source and update these pages. Corrections: diana@placementgraph.com.