Anyone who has spent years running search campaigns and lands on an app campaign gets a small shock: half the controls they know are gone. There are no keywords, no ad groups as you know them, you don't pick placements and you don't even write a complete ad. This lesson is the mental map so you don't get lost.
Difference 1 · There are no keywords (and it isn't a bug)
In an app campaign you don't choose queries. You give the system assets (text, images, video) and a goal, and it decides who sees them and where. Targeting exists, but you do it with other tools: the conversion goal itself, the country, the language and — above all — the creatives (Advanced level lesson: creatives ARE the targeting).
Difference 2 · Your ad assembles itself, across five shop windows
Google combines your assets and serves them on Search, Google Play, YouTube, Display and Discover, with different formats in each. There is no final version of the ad you can preview like in Search: there are many, and they change. You control the raw material and its quality, not the assembly.
Difference 3 · The landing page is the store listing
This is the biggest change and the most ignored. On the web, the click lands on YOUR landing page and you control everything. In apps, the click lands on your Google Play or App Store listing: screenshots, icon, title, description, ratings. If that listing converts badly, you pay more for every install, whatever you do inside Google Ads. Optimising the listing (ASO) stops being "the other team's job" and becomes part of your cost per install.
Difference 4 · The conversion happens where Google isn't
A purchase on your website is seen by your conversion tag. A purchase inside your app happens on a device, with no browser and no cookies: it has to be measured from inside the app with an SDK (Firebase or an attribution provider) and sent back to Google Ads. Without that circuit in place, your campaign only knows how to count installs — and optimising for installs is optimising for the cheapest, least valuable metric in the funnel.
Difference 5 · Android and iOS are two worlds, not two checkboxes
It isn't just two stores: iOS measures differently. Since Apple's tracking consent prompt, much of iOS attribution goes through its own aggregated, delayed system (SKAdNetwork), with less detail and less per-user data. Practical consequence: they are run as separate campaigns, with separate expectations and with numbers that will never match 100% between systems.
Difference 6 · Fewer controls, more patience
With so few levers, the temptation to fiddle is enormous… and it is exactly the worst thing you can do: every large change of goal or budget resets the system's learning. The discipline you learned in the Smart Bidding module — small changes, waiting for full cycles, judging in aggregate — is not a recommendation here: it is the only way to operate.
💡 Ninja trick: if you come from Search, translate it like this: the store listing is your landing page, the creatives are your keywords and the in-app event you choose is your value conversion. With those three equivalences, everything you already know still works.
⚠️ Pitfall: the app campaign looks easy to launch (four fields and off you go) and that is why it gets launched badly. The hard part isn't the campaign: it is before (measurement and listing) and after (creatives and patience). The next lesson is that checklist.
What you should remember
- No keywords and no placements: you provide assets and a goal.
- Your ad assembles itself across five Google surfaces.
- The store listing is your landing page: its conversion rate moves your cost per install.
- Whatever happens inside the app must be measured with an SDK and sent back to Google.
- Android and iOS: separate campaigns and different measurement rules.