YouTube Shorts Ads: Vertical Video That Works

Almost every underperforming YouTube Shorts campaign we look at has the same problem, and it is not the targeting or the bid. It is a 16:9 video that was cut for a horizontal player, dropped into a vertical feed, and left sitting in the middle of the screen with black bars above and below it.

A viewer swiping through Shorts is moving fast. Letterboxed creative announces itself as an ad before the first frame has registered, and it gets swiped past. The account owner then concludes that Shorts does not work for their business, when what did not work was the aspect ratio.

Shorts is the fastest-growing surface on YouTube and, for now, some of the cheapest attention on the platform. Here is how the inventory is actually bought, what vertical creative has to do differently, and why anyone already producing Instagram Reels is far closer to running it than they think.

What a YouTube Shorts ad actually is

Shorts ads are vertical, full-screen video ads that appear between Shorts as the viewer swipes through the feed. They play with sound on by default, they fill the whole phone screen, and the viewer can leave at any moment by swiping up.

That last point is the mechanical difference that matters. A skippable in-stream ad gives you five protected seconds before the skip button becomes active. In Shorts you get none. There is no timer, no gate, nothing holding the viewer in place. The swipe is the skip, and it is available from frame one.

Everything about how you write and shoot for this format follows from that single fact. The first second is not a warm-up. It is the entire audition.

• Placement: between Shorts in the vertical swipe feed
• Orientation: vertical, full-screen, 9:16
• Sound: on by default, unlike most social feeds
• Skip behaviour: instant, by swipe, with no protected window

How Shorts ads are bought and served

This is where most explanations go stale within six months, because Google renames YouTube campaign types regularly. So here is the part that stays true regardless of what the interface calls things this quarter.

Shorts is a placement, not a campaign type. You do not usually buy a standalone Shorts campaign. You run a video or demand-generation campaign, supply vertical assets, and Shorts becomes one of the surfaces your ad is eligible to serve on alongside in-stream and in-feed. Several campaign types can reach Shorts inventory, including video campaigns built for reach or action, demand-generation campaigns, app campaigns, and Performance Max.

The practical consequence is the one people miss: the most reliable lever you have over Shorts delivery is your asset library. A campaign that only contains a 16:9 master will either serve poorly in Shorts or get auto-cropped into something you did not art-direct. A campaign that contains a purpose-built 9:16 asset gets served properly. If you want Shorts volume, upload vertical.

Before you scale anything, check your placement or 'where ads showed' reporting to see how much of your spend actually landed in Shorts versus in-stream. Reporting granularity varies by campaign type, and it is worth confirming rather than assuming. Judging a campaign on blended numbers when 80 per cent of the spend went to in-stream tells you nothing about whether Shorts worked.

What vertical creative actually requires

Shooting for Shorts is not a cropping exercise. A handful of things change, and they change together.

The hook moves to the first second. On in-stream you have five seconds to earn the sixth. Here you have roughly one. Whatever the strongest visual or claim is, it goes at the top with no logo sting, no establishing shot, and no slow zoom into a product.

The frame gets tighter. Vertical is a narrow, tall window on a screen held at arm's length. Faces should be large, products should fill the frame, and wide group shots read as mush. If the shot needs the viewer to squint, it does not belong.

Sound is on, but the ad still has to work muted. Shorts plays with audio by default, so you can lead with a spoken line rather than a text card. But a meaningful share of viewers are in a context where they will mute or look away, so burn captions in. Captions also buy you a second hook independent of the audio one.

Respect the safe zones. The Shorts interface puts controls, account handles, descriptions and calls to action along the bottom of the screen and down the right edge. Any text you place there gets covered. Keep essential copy inside the middle band of the frame and away from the lower fifth.

Cut faster than feels right. Early cuts every one to two seconds hold a swiping viewer far better than a single continuous shot, even if the continuous shot is more beautiful.

YouTube vertical video ads: the specs worth getting right

• Aspect ratio: 9:16 vertical, full-frame. 4:5 will serve on some surfaces but does not fill a Shorts screen
• Resolution: 1080 x 1920 is the sensible target. Higher is fine, lower starts to look cheap on a modern phone
• Length: not a platform limit, but in our experience most Shorts ad creative that works sits in the 10 to 30 second range, and the shorter end holds up better on cold traffic
• Captions: burned in, high contrast, positioned in the middle band of the frame
• Safe zone: keep essential text and logos clear of the bottom fifth and the right edge, where the interface sits
• Audio: designed for sound-on, but comprehensible with sound off
• Source file: upload a clean master to YouTube rather than a re-downloaded copy from another platform

Why Shorts inventory is currently cheap

There is a genuine arbitrage here, and it comes from a supply and demand gap rather than from anything clever.

Shorts consumption has grown very fast. Advertiser supply of vertical creative has not kept pace, because most brands still commission a single horizontal hero film and cut everything else out of it. Fewer advertisers competing for a rapidly expanding pool of impressions means the clearing price stays soft. In most markets, Shorts CPMs sit noticeably below in-stream.

That gap will close. It always does. The advertisers who benefit are the ones who build a vertical production habit now, while cheap inventory is still absorbing mediocre creative, rather than the ones who arrive after the price has normalised and mediocre creative no longer clears.

If you already run Instagram Reels, you are most of the way there

This is the most under-exploited fact about Shorts advertising. If you are already producing Reels creative for Meta, you own a library of 9:16, sound-on, fast-hook video that was built under exactly the same constraints Shorts imposes. The adaptation work is small.

What to change before you run them:

• Strip platform-specific language. 'Link in bio' means nothing on YouTube, and it reads as a recycled asset
• Never upload a watermarked re-download. A visible Reels or TikTok watermark marks the ad as second-hand and tends to depress performance. Go back to the master file
• Re-check the safe zones. The Shorts interface does not sit in the same place as the Reels interface, and text that was clear on Meta can land under a button here
• Swap the call to action to something YouTube can actually deliver, and make sure the clickable element in the ad matches what the voiceover is asking for
• Re-test the hook. Reels audiences are often warmer, having already encountered the brand elsewhere. Shorts prospecting traffic is frequently colder, so a hook that assumed recognition may need to state the proposition outright

Where Shorts fits, and where it does not

Shorts is strong for e-commerce, app installs, and top-of-funnel reach, and it is very good as a cheap retargeting layer against people who have already watched something longer.

It is weak as a standalone channel for a considered or complex sale on cold traffic. You cannot explain a high-ticket service, a B2B product, or anything requiring a comparison inside a swipe feed. If the decision needs an argument, the argument belongs in skippable in-stream, and Shorts becomes the thing that keeps you present afterwards at a fraction of the cost.

For most Indian accounts we see, the useful mental model is that Shorts buys attention cheaply and in-stream converts it. Asking Shorts to do both, on cold traffic, for a product that takes thirty seconds to explain, is how a perfectly good placement gets written off.

The mistakes that cost the most

• Letterboxing a 16:9 asset into a vertical feed. This single decision explains more Shorts underperformance than everything else combined
• Putting the hook at second five, out of in-stream habit, when the swipe is available at second zero
• Placing the offer or the price under the interface, in the bottom fifth of the frame
• Running Shorts and in-stream blended in one campaign and then judging Shorts on the combined numbers
• Uploading a watermarked download from Instagram or TikTok instead of the clean master
• Expecting a swipe feed to close a considered purchase on the first touch

What we would actually run

Start by auditing what you already own. If there is a Reels library, that is your first Shorts test and it costs a day of light editing rather than a shoot.

Build the campaign so you can read Shorts separately from in-stream, either through placement reporting or by keeping the assets and campaigns distinct enough to attribute. You cannot optimise a placement you cannot see.

Shoot the next round vertical-first rather than vertical-as-an-afterthought. That means the horizontal cut becomes the derivative, not the master. It is a small change in how a shoot is planned and it removes the letterbox problem permanently.

Then use Shorts for what the price makes it good at: cheap reach at the top, and cheap frequency against warm viewers at the bottom. Let in-stream carry the part of the message that needs more than a second to land.