App campaigns in Google Ads: installs, events and value bidding in 2026

App campaigns in Google Ads are among the most powerful yet frequently misunderstood campaign types the platform offers. While advertisers quickly learn the controls for Search or Performance Max (PMax), App campaigns often remain a black box for many marketers. In 2026, that is a significant missed opportunity: advertisers who configure App campaigns correctly, with the right conversion events and a well-considered bidding strategy, can grow their app in ways that organic reach or App Store Optimisation (ASO) simply cannot match. This article explains how App campaigns work, what the difference is between the three campaign goals (installs, in-app events and value bidding), when to use each, and how AdBrains' proprietary AI technology takes this campaign type to the next level.

What are App campaigns in Google Ads?

App campaigns are a fully automated campaign type within Google Ads, specifically designed to promote apps across the entire Google network. Google automatically distributes your ads across Search, Google Play, YouTube, Discover and the Display Network. You do not need to create separate ad groups or targeting settings per channel: Google combines the assets you provide (text, images, videos) into the best possible ad for each channel and each user.

The campaign type is available for both Android apps (via Google Play) and iOS apps (via the App Store). The basic setup requires four elements: text lines, images, video assets and a starting bid or Target CPA. Once the campaign has collected sufficient conversion data, Smart Bidding starts working and automatically optimises the bidding process per auction.

What makes App campaigns unique compared to other campaign types is that they are entirely dependent on high-quality conversion signals. The better and richer the data Google receives about which users perform valuable actions after installing, the better Smart Bidding can optimise. This makes the choice of your conversion events and bidding strategy absolutely critical.

The three campaign goals: installs, events and value bidding

Manual vs. AI-powered App campaign management

What changes when you use AI for App campaigns?

Manual management

• Bid strategy adjusted manually on a weekly basis
• Conversion events evaluated only after weeks
• Creative assets uploaded and tested manually
• No automatic signal enrichment for Smart Bidding
• High-value segments discovered late or never
• Target CPA/ROAS set based on gut feeling

AdBrains AI managementAdBrains AI

• Daily tCPA/tROAS adjustments based on conversion volume
• In-app events continuously monitored and validated
• RSA improvement system flags weak assets immediately
• Server-side signal enrichment for richer bid signals
• Automatic detection of high-value user segments
• Multi-agent verification prevents faulty bid changes

Google distinguishes three main goals for App campaigns. Each goal drives Smart Bidding in a different way and is suited to a different phase of your app lifecycle or business model.

1. App installs (ACi)

The most widely used variant is focused on maximising the number of app installs at a pre-set Target CPA. This is the right choice for apps in the growth phase that primarily want to attract more users. Smart Bidding then optimises for the probability that a user will actually install the app after seeing the ad. Think of a new mobile app in the education space: in the launch phase, the goal is to generate as many relevant installations as possible to quickly build up an active user base.

2. In-app events (ACe)

Once an app has sufficient installs, the priority shifts: not every install generates value. A user who installs the app but never makes a purchase or registers contributes nothing to revenue. App campaigns for in-app events (ACe) steer Smart Bidding towards achieving specific actions within the app, such as a purchase, completing a registration, adding an item to a basket, or completing a specific level in a game.

For an online learning platform like ToetsJeKennis.nl, which offers a mobile app where students can purchase exam bundles and take practice exams, the relevant in-app events are clear: starting a trial period, completing a first purchase, or activating a subscription. By steering Smart Bidding towards these events, the campaign attracts not just more installs, but specifically users who also actually convert.

3. Value bidding (ACe with In-App Purchase Values)

The most advanced variant combines in-app events with actual transaction value. Rather than optimising for the number of events, Smart Bidding targets maximising total in-app revenue at a set Target ROAS. This requires that you pass in-app purchases with dynamic values to Google Ads via the Firebase SDK or a Mobile Measurement Partner (MMP) such as AppsFlyer or Adjust.

Value bidding is particularly powerful for e-commerce apps with a varied product range, where one purchase is significantly more valuable than another. A user who purchases a premium annual licence is far more valuable than someone who buys a single low-cost practice test. With value bidding, Smart Bidding learns which user profiles make the highest-value purchases and actively targets those profiles.

How App campaign assets work

Because App campaigns are fully automated, the quality of your provided assets determines success. A strong asset mix consists of:

• At least four short text lines (each max. 25 characters) and two longer descriptions (each max. 50 characters)
• Landscape and portrait images in multiple aspect ratios (1:1, 1.91:1, 9:16)
• Videos in multiple formats: landscape (16:9), portrait (9:16) and square (1:1)
• HTML5 assets for interactive ads on the Display Network (optional but valuable)
• A strong app store listing with compelling screenshots and a clear description

The Ad Strength of an App campaign indicates the quality and diversity of your asset set. Just as with Responsive Search Ads (RSAs), the richer and more diverse the asset mix, the more combinations Google can test, and the better the ultimate ad performance. A Poor Ad Strength is a direct signal that assets are missing or that there is insufficient variety.

Setting up conversion events correctly: the foundation of everything

App campaign conversion funnel

Typical flow per phase — e-commerce app benchmark (per month)

100% Impressions

Total reach via Search, Display, YouTube and Play

6% Clicks / install attempts

Average CTR App campaigns: 5-7%

3% Successful installs

Install rate after click: approx. 50% of clickers install

0.24% In-app events (purchase / registration)

Event rate on total reach: 0.2-0.3%

Value bidding targets high-value events, not just volume

The choice and configuration of your conversion events is the foundation on which the entire App campaign rests. An error in event configuration has a direct impact on the quality of Smart Bidding signals and therefore on the return on the campaign. There are three common mistakes to absolutely avoid:

1. Switching to in-app events too early: Only move from ACi to ACe when your campaign consistently generates at least 100 installs per week. Without sufficient volume, Smart Bidding cannot learn reliable patterns.
2. Events with too little volume as primary goal: If you set a very rare action as your only bid goal, Smart Bidding has too little data to optimise on. Choose events that occur at least 10-20 times per week.
3. Double counting through multiple SDK integrations: If you use both Firebase and an MMP without proper deduplication, Google counts conversions twice, leading to overly aggressive Smart Bidding decisions.

Beyond correct event configuration, the quality of the conversion signal itself plays a major role. App campaigns benefit enormously from enriched conversion signals: the more first-party data Google receives about who performed a valuable action, the better Smart Bidding can predict who will perform the same action in the future.

App campaigns vs. other campaign types: when to choose what?

For E-4motion.com, the webshop for new electric folding bikes, an App campaign can be an interesting addition to existing Search and PMax campaigns, particularly for generating test ride requests via a mobile app. The high purchase price of an electric folding bike makes value bidding less relevant than for an e-commerce platform with many small repeat purchases. An ACe campaign focused on test ride requests (as a lead gen event) is then the most logical choice.

How AdBrains AI takes App campaigns to the next level

App campaigns are theoretically already largely automated by Google, but that does not mean there is no room for further optimisation. Quite the opposite: precisely because App campaigns are so dependent on the quality of conversion signals, asset sets and bidding strategies, there is enormous potential to gain with the right AI layer on top of Google's system.

AdBrains has developed several specific modules that work seamlessly together on App campaigns. First, the server-side signal enrichment system ensures significantly richer bid signals. Via AdBrains' own sGTM infrastructure, conversion events are enriched with first-party customer data before being passed to Google Ads. This means Smart Bidding not only knows that a conversion occurred, but also receives contextual information about the user who performed the action. Advertisers implementing server-side tracking see on average more than 50% more tracked conversions, giving Smart Bidding a structural advantage when learning high-value user patterns.

Second, AdBrains' multi-agent verification system reviews every bid change before it is implemented. Four independent AI agents evaluate every proposed adjustment to Target CPA or Target ROAS, based on conversion volume, trend data and margin targets per client. This prevents a tCPA from being lowered too early when Smart Bidding is still in its learning phase, or a bidding strategy switch being made on the basis of insufficient statistical evidence. Particularly for App campaigns, where the learning phase is more sensitive than for other campaign types, this is a crucial advantage.

Third, AdBrains' RSA improvement system also works on the asset sets of App campaigns. Automatic analysis of Ad Strength scores detects assets with a Poor status and generates concrete improvement suggestions for text lines and image formats, keeping asset diversity structurally at the required level. Finally, automated tCPA/tROAS optimisation monitors conversion volume per campaign daily and adjusts the bidding strategy based on actual performance and the client's margin goals. For App campaigns working with value bidding, this is particularly valuable: when the mix of in-app purchases changes due to seasonal influences, the tROAS adjusts automatically so the budget always flows to the most profitable user segments.

Common mistakes in App campaigns

App campaigns: benchmark KPIs in 2026

Average performance indicators for well-managed App campaigns

5-7%

Average CTR App campaigns

+34%

Higher in-app revenue with value bidding vs. installs only

50%+

More tracked conversions with server-side tracking

3-5x

ROAS improvement when optimising on high-value events

Beyond the event configuration errors discussed above, there are a number of patterns AdBrains regularly encounters in accounts that underperform:

• Increasing budget too quickly: A rapid budget increase of more than 20% per week returns the campaign to the learning phase, temporarily causing higher CPAs. Gradual budget increases give Smart Bidding time to re-acclimatise.
• No separate campaigns per goal: Mixing installs and in-app events in the same campaign gives conflicting signals to Smart Bidding. Keep goals strictly separated per campaign.
• iOS campaigns without SKAdNetwork configuration: Without correct SKAdNetwork integration, Smart Bidding misses a large portion of iOS conversion signals due to Apple's privacy restrictions, leading to underbidding in a crucial segment.
• Assets not refreshed for years: Google recommends adding new assets at least every quarter to maintain freshness and reactivate the learning phase on improved creatives.
• No audience segmentation for remarketing: App campaigns can be strengthened with RLSA audiences of existing users who have already completed certain events, but this is rarely utilised.

Frequently Asked Questions about App campaigns

What is the minimum budget for an App campaign?

Google recommends a daily budget of at least 50 times the set Target CPA for App install campaigns. If you set a Target CPA of two euros, you need at least 100 euros daily budget to generate sufficient conversion signals for Smart Bidding. For in-app event campaigns, the same principle applies: the budget must be large enough to generate multiple events per day. Going below this minimum keeps the campaign structurally in the learning phase, making results unreliable.

When should I switch from ACi to ACe?

The rule of thumb is: only switch to an in-app event campaign when your App install campaign consistently generates more than 100 installs per week. Only at that point does Smart Bidding have sufficient signal volume to learn from the relationship between user profiles and in-app behaviour. Switching too early results in a prolonged, costly learning phase where CPA temporarily rises sharply before the system stabilises.

Do App campaigns work for iOS apps after Apple's ATT changes?

Yes, but the setup requires more attention. Due to Apple's App Tracking Transparency (ATT) framework and the limited attribution windows of SKAdNetwork, Google receives less granular iOS conversion signals than for Android. This means iOS campaigns typically take longer to exit the learning phase. Correct SKAdNetwork configuration, use of Apple's Privacy Nutrition Labels, and setting a sufficiently high Target CPA are essential. AdBrains monitors iOS campaigns with extra intensity via the multi-agent verification system to prevent premature bid changes based on insufficient iOS data.

What is the difference between value bidding and a standard Target ROAS strategy?

Value bidding in App campaigns is functionally similar to Target ROAS in Search or PMax campaigns: you give Google a revenue target and Smart Bidding steers the bidding process to maximise in-app revenue. The crucial difference lies in the data source: with App campaigns using value bidding, purchase values must be passed via Firebase or an MMP as an event parameter (value), so Smart Bidding learns that an 80-euro purchase is structurally more valuable than a 10-euro purchase. Without these dynamic value signals, Smart Bidding optimises for the number of events, not revenue. Value bidding is therefore not simply a different setting, but a fundamentally different approach to campaign optimisation that only works when the measurement infrastructure is in order.