Google App Campaigns are Google's dedicated campaign type for driving mobile app installs, in-app actions, and pre-registrations across Search, Play, YouTube, Discover, and the Display Network, all from a single campaign. Setting up Google Ads for mobile apps in 2026 requires a fundamentally different approach than web-based advertising. You cannot pick keywords, manually set placements, or control ad creative combinations the way you would in a standard Search or Shopping campaign. Instead, you provide assets, choose a bidding goal, and let Google's machine learning decide where and how to serve your ads. This guide covers everything you need to run Google app campaigns in 2026: setup, tracking, creative, bidding, scaling, and the management decisions that separate profitable app advertisers from those burning through install budgets with nothing to show for it.
Google App Campaigns (formerly Universal App Campaigns) are the primary way to advertise a mobile app through Google Ads. Unlike other campaign types, App Campaigns automatically generate ads by combining your text, image, video, and HTML5 assets with information pulled directly from your app store listing. Google then distributes those ads across its entire inventory, including Search, Google Play, YouTube, Gmail, Discover, and millions of Display Network apps and sites.
The key distinction is automation. You do not select keywords. You do not choose placements. You do not write individual ads for each network. Google's systems handle all of that based on the conversion goal you select: installs, in-app actions, or pre-registrations.
This level of automation makes App Campaigns simultaneously easier to launch and harder to optimize. The levers you control are limited to assets, bids, budgets, and conversion event selection. That means every decision you do make carries disproportionate weight.
A common question from advertisers new to app promotion: why not just run a regular Search campaign targeting app-related queries? The short answer is that standard campaign types are built for web conversions, not app installs. They lack native integration with app stores, cannot track installs or in-app events without significant workaround, and do not have access to the same inventory that App Campaigns unlock, particularly Google Play placements and in-app ad units.
Google explicitly restricts app install ads to the App Campaign format. Trying to drive installs through standard Display or YouTube campaigns means losing access to app-specific bidding strategies, attribution, and reporting.
Android and iOS require separate App Campaigns. Google Play apps benefit from deeper integration since Google owns the ecosystem, meaning install attribution is more straightforward and you can leverage Google Play listing data directly.
iOS campaigns face ongoing attribution challenges due to Apple's App Tracking Transparency (ATT) framework. SKAdNetwork remains Apple's sanctioned attribution mechanism for iOS, and it introduces delays, limited conversion value data, and restricted optimization signals. We cover the specifics in the tracking section below, but the strategic implication is clear: Android campaigns typically ramp faster and produce more reliable data than iOS campaigns on Google Ads.
Google offers three App Campaign subtypes, and choosing the right one is the first structural decision you make.
App Installs drive new users to download your app. This is the most common starting point and the subtype where most budget is allocated. You optimize toward either a target cost-per-install (CPI) or a target cost-per-action (CPA) for a specific in-app event.
App Engagement re-engages existing users who have already installed your app but may have gone dormant. These campaigns deep-link users back into specific sections of your app and are critical for retention-focused strategies.
App Pre-Registration (Android only) lets you promote an app before it launches on Google Play. Users can pre-register, and Google notifies them automatically when the app becomes available.
Most advertisers should start with App Installs, graduate to running Installs and Engagement simultaneously, and consider Pre-Registration only for new app launches.
Asset quality is the single biggest lever you have in App Campaigns because Google's automation assembles your ads dynamically. Here is what you need:
Text: Up to five headlines (30 characters each) and five descriptions (90 characters each). Write distinct messages for each slot. Do not repeat the same value proposition five times. Cover different angles: feature highlights, social proof hooks, urgency, and benefit-driven statements.
Images: Upload up to 20 images. Prioritize 1200x628 (landscape), 1200x1200 (square), and 480x800 (portrait). Each format serves different placements. Missing a format means missing inventory.
Video: Up to 20 videos. This is where most advertisers underinvest. If you do not upload video, Google will auto-generate video from your app store listing assets. Auto-generated video almost always underperforms custom creative. Upload portrait (9:16), landscape (16:9), and square (1:1) videos.
HTML5: Optional but useful for interactive ad formats on the Display Network.
The more high-quality, diverse assets you provide, the more combinations Google can test, and the better your campaigns will perform. Thin asset libraries are one of the most common reasons app campaigns underdeliver.
Your bidding strategy directly determines how Google allocates your budget and which users it targets.
Target CPI (Cost Per Install): You set a target price per install. Google optimizes to hit that average. Best for early-stage campaigns focused purely on volume.
Target CPA (Cost Per Action): You set a target cost for a specific in-app event, like a registration, purchase, or subscription. This requires enough conversion data for Google's models to optimize effectively. Generally, you need at least 10 conversions per day before tCPA bidding becomes reliable.
Maximize Conversions: Google spends your full budget to get as many conversions as possible without a target. Useful for learning-phase data collection but can produce volatile CPIs.
For most advertisers, the path is: start with Target CPI or Maximize Conversions to accumulate install data, then shift to Target CPA once you have sufficient in-app event volume.
App Campaigns have a learning phase that typically lasts several days. During this time, performance fluctuates as Google's algorithm explores different audience segments, placements, and asset combinations.
Set your daily budget to at least 50 times your target CPI. If your target CPI is $2, budget at least $100 per day. Going below this threshold extends the learning phase and often leads to campaigns that never stabilize.
Do not make significant changes during the learning phase. Adjusting bids, budgets, or assets resets the learning process. Patience during the first week is critical, and it is also one of the hardest disciplines for advertisers to maintain. This is an area where having experienced management matters enormously. Services like groas, which combines AI agents monitoring campaigns 24/7 with a dedicated human account manager, can navigate the learning phase without the knee-jerk reactions that derail so many app campaigns early on.
Unlike Search or Display campaigns, App Campaigns do not allow you to select keywords, topics, or placements manually. Google determines targeting based on your app category, your assets, your conversion data, and its own user intent signals.
This means your ability to influence who sees your ads is indirect. The primary controls are your assets (which signal what your app is about), your conversion events (which tell Google what a valuable user looks like), and your geographic and language targeting.
You can apply audience signals to App Campaigns, including Custom Audiences built from search terms, URLs, or apps that your target users are likely to engage with. These signals help Google's automation orient toward the right user profiles.
Similar Audiences (now largely folded into Google's automated audience expansion) extend your reach to users who resemble your existing converters. The quality of these audiences depends directly on the quality of your conversion data, which brings us back to the importance of tracking setup.
App Engagement campaigns target users who have installed your app but stopped using it. You define audience lists based on in-app behavior: users who haven't opened the app in 14 days, users who started but didn't complete onboarding, users who made one purchase but never returned.
These campaigns use deep links to bring users back to specific screens within your app. Engagement campaigns are often the highest-ROI campaigns in a mature app advertising strategy because the user has already been acquired. The cost to re-engage is typically a fraction of the cost to acquire.
Firebase is Google's mobile app analytics and development platform, and it is effectively mandatory for running Google App Campaigns properly. Firebase provides the event data that Google Ads uses to optimize campaigns, attribute installs, and track in-app actions.
Without Firebase integration, you are flying blind. You can technically run App Campaigns without it, but you lose access to in-app event optimization, audience creation from app behavior, and accurate conversion reporting. Integrating Firebase with Google Ads is straightforward but must be done correctly: link your Firebase project to your Google Ads account, configure conversion events, and verify data is flowing before launching campaigns. For a broader look at auditing your entire Google Ads setup, including conversion tracking, see our complete account audit checklist.
Not all in-app events are equal, and choosing which event to optimize your campaign toward is a high-stakes decision.
Start by mapping your user funnel: install, onboarding completion, first session, key feature usage, subscription start, purchase. Then choose a conversion event that is far enough down the funnel to represent real value but frequent enough to give Google sufficient data for optimization. Optimizing for a rare event (like annual subscription) on a small budget means Google never gets enough signals to learn.
A common effective approach: optimize for a mid-funnel event (like registration or first meaningful action) while tracking deeper events for reporting. As volume grows, shift optimization to deeper, higher-value events.
iOS attribution remains constrained by Apple's privacy framework. SKAdNetwork (now in version 4.0 and beyond) provides aggregated, delayed attribution data. What this means practically for Google Ads:
Delayed reporting: Conversion data arrives 24 to 48 hours after the install, sometimes longer. Real-time optimization is impossible on iOS.
Limited conversion values: SKAdNetwork supports a restricted set of conversion value combinations, meaning you cannot pass granular revenue data the way you can on Android.
Reduced audience granularity: Remarketing and audience creation are more limited on iOS because user-level tracking requires ATT opt-in, which most users decline.
The strategic response is to invest proportionally more in Android campaigns where data is richer, while running iOS campaigns with broader optimization targets and accepting wider performance variance.
Measuring return on ad spend for app campaigns is inherently more complex than web campaigns. Install attribution can be affected by organic cannibalization (users who would have installed anyway), multi-touch journeys that span platforms, and the iOS attribution limitations described above. For scaling your budget effectively as performance data comes in, our Google Ads budget planning guide covers allocation strategies that apply to app campaigns as well.
Use incrementality testing when possible, compare ROAS across cohorts rather than relying solely on last-click attribution, and build reporting that accounts for delayed conversion windows. This kind of nuanced measurement is where most self-managed app campaigns fall apart, and where professional management provides genuine leverage.
Video is the highest-impact asset type in App Campaigns because it runs on YouTube, the single largest placement by volume. The first three seconds determine whether a user watches or skips.
Effective video hooks for app installs: show the core app experience immediately, lead with the problem your app solves, or open with a surprising visual that relates to your value proposition. Do not start with a logo animation or brand introduction. Nobody cares about your brand in the first three seconds. They care about what your app does for them. Our YouTube Ads strategy guide goes deep on creative that converts.
Portrait (9:16): Dominates mobile-first placements. This is your highest-priority format for reaching users on phones, which is where the vast majority of app installs happen.
Square (1:1): Strong performer across both mobile and tablet placements. Good middle ground.
Landscape (16:9): Still necessary for YouTube pre-roll and some Display placements, but secondary priority for app install campaigns.
Upload all three formats. If you can only produce one, make it portrait.
App Campaign creative fatigues faster than web campaign creative because the same users see your ads repeatedly across Google's network. Plan to refresh your asset library every four to six weeks.
Monitor asset performance reports within Google Ads. Google rates each asset as "Low," "Good," or "Best." Replace low-performing assets regularly and test new variations of your best performers. Creative management at this cadence is where many teams fall behind. It requires constant attention, which is precisely why groas assigns a dedicated human account manager to each account while AI agents handle the continuous monitoring and optimization of asset performance across campaigns.
Scaling a single App Campaign by increasing budget works until it doesn't. The general rule: if you increase daily budget by more than 20% at a time, you risk resetting the learning phase and destabilizing performance.
When your single campaign is performing well but you want more volume, add a second campaign targeting a different conversion event, a different geo, or a different creative angle. This preserves the stability of your first campaign while expanding reach.
For a deeper look at stage-by-stage scaling principles, our guide to scaling Google Ads from $5K to $100K/month lays out the methodology that applies across campaign types, including app campaigns.
Geographic expansion is one of the cleanest scaling levers for app campaigns. Start with your highest-value markets, establish benchmarks, then expand into adjacent markets.
Keep new geos in separate campaigns so their learning phase does not disrupt your established campaigns. CPI benchmarks vary dramatically by country, and mixing geos in a single campaign gives Google's algorithm conflicting signals.
Managing app campaigns at scale involves a specific set of challenges: creative refresh cycles, bid adjustments across multiple campaigns and geos, conversion event selection as your funnel matures, and constant monitoring of learning-phase stability. These are not tasks that benefit from once-a-week check-ins. They require continuous attention.
This is where groas provides a structural advantage over agencies, freelancers, or in-house teams. AI agents monitor campaign performance 24/7, catching learning-phase resets, creative fatigue, and bid inefficiencies in real time. Meanwhile, your dedicated human account manager handles the strategic decisions: when to shift optimization events, how to structure geo expansion, and which creative angles to test next. It is a combination that no traditional management approach can match for consistency or cost efficiency.
Running Google App Campaigns well requires continuous, detail-oriented management across creative, bidding, tracking, and scaling decisions. Most agencies assign a junior account manager who checks your campaign a few times a week. Freelancers offer even less coverage. In-house teams get stretched across too many priorities. And self-serve tools give you dashboards and recommendations but leave every execution step to you.
groas replaces all of those options. When you onboard, you get a dedicated human account manager who audits your existing app campaigns (or builds them from scratch), delivers a custom roadmap within 24 hours, and then implements the full plan. From there, groas AI agents manage your campaigns around the clock while your account manager oversees strategy, joins bi-weekly calls, and remains available via private Slack channel or email.
For app advertisers specifically, this means your creative refresh cycles happen on schedule, your bid strategies evolve as conversion data accumulates, your geo expansions are structured properly, and your learning phases are protected from premature changes. All of this for a fraction of what a single in-house hire or agency retainer would cost.
If you are spending on Google App Campaigns and not getting the install volume, cost efficiency, or in-app ROAS you need, the problem is almost certainly management, not the channel. groas fixes that.
Google App Campaigns are a dedicated campaign type within Google Ads designed to promote mobile apps across Search, Google Play, YouTube, Discover, Gmail, and the Display Network. You provide text, image, and video assets, set a bidding goal (installs or in-app actions), and Google's machine learning automatically generates and distributes ads across its inventory. They are the only campaign type Google supports for driving app installs at scale.
Set your daily budget to at least 50 times your target cost-per-install. If your target CPI is $2, your minimum daily budget should be $100. Going below this threshold extends the learning phase and often prevents campaigns from stabilizing. For serious scaling beyond $10K per month, you will need to run multiple campaigns across different geos and conversion events.
Yes. Firebase integration is effectively mandatory for running App Campaigns properly. Without it, you lose access to in-app event optimization, behavioral audience creation, and accurate conversion reporting. Link your Firebase project to your Google Ads account and verify data is flowing before launching any campaigns.
iOS attribution relies on Apple's SKAdNetwork, which provides aggregated and delayed conversion data. Install data typically arrives 24 to 48 hours after the event, conversion value granularity is limited, and remarketing audiences are restricted because most users decline App Tracking Transparency prompts. Android campaigns generally produce more reliable data and ramp faster on Google Ads.
Start with Target CPI or Maximize Conversions to accumulate install data during the early learning phase. Once you have at least 10 in-app conversions per day, shift to Target CPA bidding on a meaningful in-app event like registration or first purchase. This progression gives Google's algorithm enough signal to optimize effectively.
Refresh your asset library every four to six weeks. App Campaign creative fatigues faster than web campaign creative because users see ads repeatedly across Google's network. Monitor asset performance ratings in Google Ads and replace low-performing assets while testing new variations of your best performers.
Yes. groas is a full-service Google Ads management service that handles app campaigns end to end. AI agents monitor your campaigns 24/7, catching learning-phase resets, creative fatigue, and bid inefficiencies in real time, while your dedicated human account manager handles strategic decisions like conversion event selection, geo expansion, and creative direction. You get expert-level management at a fraction of the cost of an agency or in-house hire.
Absolutely. Android and iOS require separate App Campaigns because they use different attribution systems, produce different data volumes, and have different performance benchmarks. Mixing them would give Google's algorithm conflicting signals and make it impossible to evaluate performance accurately for each platform.
Underinvesting in assets, particularly video. If you do not upload custom video, Google auto-generates video from your app store listing, and auto-generated video almost always underperforms. The second most common mistake is making changes during the learning phase, which resets the optimization process. groas solves both problems: AI agents protect the learning phase from premature changes while your dedicated account manager ensures creative assets are refreshed on a proper schedule.
