YouTube Shorts in App campaigns: what changes

Google's App campaigns have run on automated placement for years, spreading one set of uploaded assets across Search, Display, YouTube, Discover, with no buyer choosing where any single impression lands. Shorts inventory sits in that same pool. So most buyers are running Shorts campaigns without ever having decided to.

Shorts inventory is already in the mix

A studio running a standard App campaign for installs is almost certainly serving into YouTube Shorts placements already. Nobody on the team need have meant to buy them. Google's machine learning system allocates spend across its inventory on predicted performance, and Shorts, being a high-volume vertical video surface, takes a meaningful share of that allocation whenever a campaign has vertical assets that qualify.

That default matters, because on almost any other platform a buyer decides deliberately whether to run a Shorts-specific line item, whereas here the allocation happens on its own and the only influence left to the buyer runs through the assets supplied and the campaign's overall optimisation signal.

Why vertical assets change the auction

A campaign built entirely on horizontal or square video is not shut out of Shorts placement. It simply competes there with visibly weaker material than a campaign carrying purpose-built vertical creative. Google's ad serving crops and reformats what it has when nothing better exists, and a cropped horizontal video looks exactly like what it is inside a feed built around the full vertical frame.

Supply genuine 9:16 assets. Shot for the vertical frame, or edited for it, rather than adapted after the fact. That gives the algorithm a stronger candidate for Shorts placement and tends to lift the volume of Shorts impressions a campaign wins against an otherwise identical campaign running only horizontal creative. The asset is the lever here, not a manual placement choice.

What a buyer can and cannot control

A buyer cannot target Shorts specifically inside a standard App campaign. Nor can a buyer see a clean placement-level cost or conversion breakdown, the way a Meta or TikTok dashboard would provide it, or drop Shorts from an App campaign without dropping YouTube placements more broadly. Google's automated bidding treats the whole inventory pool as one optimisation problem, and the placement-level opacity that comes with App campaigns generally applies here too.

What a buyer can control is asset supply and campaign structure. Feed the system a deep pool of vertical video variants and keep the horizontal and square assets that other placements still need. Campaign-level conversion goals, install versus a deeper in-app event as the optimisation target, also shape which assets and which placements the system favours, since it learns toward whichever creative produces the chosen outcome most efficiently.

Asset group hygiene as the real lever

Treat asset groups the way a buyer treats a Meta ad set. Retire assets that are not getting served. Refresh vertical creative on a cycle instead of leaving the same three clips running for months, and read the asset-level performance reports for signals of which formats the system actually favours, even without full placement transparency.

A studio auditing its App campaigns for the first time with this in mind often finds a vertical asset gap it did not know it had, a campaign running mostly horizontal creative that is quietly losing out on a placement it never chose to opt out of. Closing that gap is one of the few concrete, low-cost moves left inside a campaign type that otherwise hands most of the decision-making to the algorithm.

Once a team accepts that Shorts is part of the pool, the same logic reaches creative testing more broadly. A concept that tests well as a quick vertical cut and poorly as a longer horizontal edit is not necessarily a weak concept. It may simply suit the placements the algorithm already routes it toward. Read asset-level results with that distinction in mind rather than judging every asset against one blended average, and a buyer gets a clearer sense of which creative to keep feeding the system and which to retire, even without the placement-level detail a more transparent platform would provide. Which argues, finally, for producing vertical assets as a first-class deliverable rather than an afterthought derived from a horizontal master, since a purpose-shot vertical asset consistently outperforms a cropped one in the placements where the frame itself is the format.

Related archive reading

These articles provide related context and remain subject to their stated review status.

• What Meta Advantage+ App Campaigns Mean for Game UA
• Tripledot's AppLovin Apps Portfolio: A Buyer's Guide