YouTube Shorts Ads: What Works and What to Skip

You can't buy Shorts as a clean, standalone placement the way you buy TikTok. There's no "Shorts campaign" button that does only what you think it does. You buy through Demand Gen or Video campaigns, tell Google your objective, and your vertical creative gets slotted into the Shorts feed alongside in-stream pre-roll and in-feed discovery. Where it lands depends as much on your asset's aspect ratio and your bid type as on any setting you picked.

That's the first thing to get right, because most founders pour budget into a campaign assuming it's running on Shorts and never check where the impressions actually came from. This guide covers the buying mechanics that decide that, and then the creative patterns that make a vertical Short out-earn the same dollars spent on a 15-second skippable pre-roll.

How you actually buy Shorts inventory

There is no isolated Shorts ad product. Shorts placements are served through two campaign types, and your creative aspect ratio is the strongest signal for whether you end up there.

Demand Gen campaigns are the closest thing to a "social-style" buy on Google. They run across Shorts, the main feed, Discover, and Gmail, optimized for conversions or clicks. Upload a 9:16 asset and Shorts becomes a primary surface for it.
Video campaigns (Video reach / Video views) can also serve Shorts, but they're built around in-stream pre-roll first. A vertical asset is eligible for Shorts; a horizontal asset effectively isn't.

The practical rule: aspect ratio is the placement lever. Upload 9:16 and you opt into Shorts and other vertical surfaces. Upload only 16:9 and Google has no vertical asset to serve there, so your spend drifts to pre-roll regardless of intent.

You cannot perfectly fence spend to Shorts-only, and forcing it usually raises cost-per-result. The better move is to feed the system vertical creative built for the Shorts context, let it allocate, then read the placement report to see what it did.

Bidding and objective, briefly

For a founder selling something, run Demand Gen on a conversion objective (Maximize conversions, or target CPA once you have ~15-30 conversions for the algorithm to learn from). Use Video reach only for genuine top-of-funnel awareness. The common waste is running a "views" objective, celebrating cheap views, and wondering why nothing sold — views are cheap because the system optimizes for the cheapest watchers, not buyers.

Shorts versus pre-roll on the same budget

Both can serve from the same campaign, but they reward opposite creative instincts, and that's why one usually wins.

Skippable in-stream pre-roll plays before a video the viewer chose to watch. Their mindset is "let me past this." Your enemy is the Skip button at five seconds, and everyone in the audience knows you're the interruption.

Shorts plays inside a feed the viewer is swiping for entertainment. There's no Skip button — there's a thumb flick, faster and more ruthless. But the difference is intent: in pre-roll the viewer wants you gone so the real video starts; in Shorts your ad is the content, and a good one is indistinguishable from the organic Short above it.

On equal spend, vertical-native Shorts creative tends to win for direct response for three reasons:

No skip-at-five cliff. Pre-roll trains you to cram the pitch before the Skip button arms. Shorts lets a hook breathe a beat longer because the exit is a swipe, not a countdown.
Full screen, sound-forward. Shorts is full-bleed vertical with audio viewers often leave on. Pre-roll shares the screen with the page and the "Video will play after ad" frame.
Native creative is cheaper to serve. Ads that match feed behavior get watched longer, and watch-through is part of what keeps your effective CPM down.

The honest caveat: pre-roll still wins with one polished, broadly-appealing brand film and an awareness goal. For founders shipping many direct-response variants on a tight budget, Shorts is the higher-leverage surface.

The creative patterns that outperform pre-roll

A pre-roll ad and a Shorts ad should not look the same, even for the same product. Re-uploading your horizontal pre-roll as a Short is the single most common mistake, and it underperforms because it carries every "this is a commercial" signal the Shorts feed is trained to flick away.

What outperforms:

Vertical-first framing, not a cropped horizontal. A center-cropped 16:9 leaves dead space and amputates the parts of the frame the eye lands on. Compose for 9:16 from the start.
A hook that doesn't announce itself as an ad. No logo sting, no slow brand reveal. Open mid-action with a spoken or captioned claim. The first second decides the swipe.
Burned-in captions. Even though Shorts is more sound-on than most feeds, captions hold viewers who scroll muted and reinforce the message for everyone else. Burn them in rather than relying on auto-captions, which render inconsistently.
One idea, demonstrated. A Short that shows the product doing one specific thing beats a montage of benefits. Specificity is what reads as real.
A single, spoken CTA near the end. "Link below" or "search [brand]" — one instruction, once.

A reusable 30-second Shorts ad skeleton

Write to this structure, then cut anything that doesn't earn its place. Shorts can run longer than 30 seconds, but for direct response this is the band where completion stays high.

0-2s — Hook with motion. No static frame, no logo. A line a real person would say plus visible movement. "I rewrote this ad four times. The version that worked took two minutes to make."
2-6s — Stakes, in their words. Name the problem the way the viewer would phrase it. "Hiring an editor for every variant was eating my whole test budget."
6-20s — Demonstration. Show the mechanism. Screen recording, before/after, a hand using the thing. This is the proof and it must be concrete.
20-27s — Payoff. The specific result. "Five variants live by lunch, and I could finally tell which hook actually sold."
27-30s — One CTA. "Link's in the description." Stop there. A second CTA splits intent and converts worse.

A worked example for a hypothetical meal-prep service:

0-2s: (hands opening a fridge full of labeled containers) "I haven't decided what's for dinner in three weeks."
2-6s: "I was spending an hour every night staring into the fridge, then ordering takeout anyway."
6-20s: (unboxing, then a 10-second cook) "These show up Sunday, prepped. Each one's done in the pan in under ten minutes."
20-27s: "Cut my food spend by a third and got my evenings back."
27-30s: "First box is in the description."

It never says "the best meal kit" or names the brand in the first second. It states one result and shows how it happens.

Reading the placement report so you know what you bought

Because Shorts and pre-roll share campaigns, the report is the only honest record of where your money went. Check it within the first few days, not at the end of the month.

Pull the "Where ads showed" / placement breakdown. Confirm a real share of impressions came from Shorts, not just in-stream. If almost everything is in-stream, your vertical asset is missing or weak.
Compare cost-per-conversion by surface. If Shorts converts cheaper, that's your signal to lean harder into vertical variants.
Watch the view-rate trap. A high view rate with no conversions usually means the objective is optimizing for cheap watchers. Re-check that you're on a conversion goal, not a views goal.
Track creative fatigue. Shorts creative burns out fast in a high-velocity feed. When cost-per-result on your winner starts climbing week over week, it's fatigue, not a broken account — refresh the creative.

Volume is the unlock, and the bottleneck

The reliable Shorts strategy is the same as the reliable strategy on any feed: ship many variants of one core message, change one variable per version, and let spend find the winner. You can't predict which hook lands, so betting an entire budget on one polished Short is a sample size of one.

For a single offer, a sane test set is:

4-6 hooks — same demonstration, different opening lines.
2-3 framings — problem-first, result-first, comparison.
One variable changed per variant so a winner tells you something reusable. Swap the hook and the music and the CTA at once and a win teaches you nothing.

That's a dozen-plus creatives for one product. Producing them by hand or by agency is slow and expensive enough that most founders ship one Short, watch it underperform, and conclude Shorts doesn't work for them. The surface didn't fail. The test was too small to read.

Honest trade-offs

A few things worth being straight about before you commit budget.

You give up precise placement control. Google decides allocation across vertical surfaces; you influence it with creative and objective, you don't dictate it. If you need surgical channel separation, Shorts inside Demand Gen will frustrate you.

Native creative fights brand guidelines. The Shorts that perform look less polished than your brand book wants. You'll trade some control for lower CPM, or pay the CPM.

Fatigue is faster than on search or display. Plan a rolling creative refresh, not a quarterly one — which loops back to volume: the strategy that works is exactly the one that's painful if every video is a manual edit.

FAQ

Can I run ads only on YouTube Shorts?

Not cleanly. There's no Shorts-only placement. You serve Shorts through Demand Gen or Video campaigns by uploading 9:16 vertical creative, then confirm in the placement report how much spend actually landed on Shorts. Trying to force Shorts-exclusive delivery usually raises your cost-per-result.

Are YouTube Shorts ads better than skippable pre-roll?

For direct response on a small budget, vertical-native Shorts creative usually wins on the same spend, because there's no skip-at-five cliff and the ad reads as feed content rather than an interruption. Pre-roll still wins for broad awareness with one polished brand film. The mistake is using the same asset for both.

How long should a YouTube Shorts ad be?

For conversion-focused ads, 15 to 30 seconds is the practical band: long enough to hook, demonstrate, and ask for the click, short enough to keep completion high. If yours runs longer, cut the slow middle, not the hook or the CTA.

Shipping a dozen vertical, captioned variants is the part that stops most founders from testing Shorts properly. Aitachyon takes a website URL and returns a finished, captioned video ad in about two minutes: three script variants from one brand scrape, exported in 9:16, 16:9, or 1:1 for Shorts, Reels, TikTok, Meta, and LinkedIn. Plans start at $29/mo, with Pro at $79 and Agency at $299, all under a 14-day money-back guarantee — roughly the cost of a single freelance edit for a month of variants to test against.