Table of Contents
If your Google App Campaigns are finding installs but not enough paying users, start with the conversion signal, creative and campaign structure. A lower cost per install is only useful when the people you acquire also contribute to your business.
This guide covers campaign goals, bidding, creative tests and measurement checks for teams already running app acquisition. It draws on Google’s current guidance and linked Admiral Media case studies. Client results describe individual engagements; they are not a forecast for your app.
Technical guidance reviewed on 20 September 2026. Campaign options and eligibility can vary by account and app platform.
How Google App Campaigns Work in 2026
App campaigns use your creative assets, store listing, budget and chosen goal to assemble and distribute ads across Google’s eligible placements. Campaign types cover installs, engagement and pre-registration. Choose the type around the immediate outcome you need, then check the options available for your app in Google Ads.
App Connect is a separate setup for app measurement and deep links from supported web campaigns. It can help existing app users reach a relevant in-app destination. It does not turn every web campaign into an App campaign. For the wider acquisition plan, see our mobile app user acquisition guide.
Campaign Structure and Setup Best Practices
Group Markets Around Language and Economics
Separate markets when language, user value, acquisition cost or reporting needs justify independent budgets and targets. Avoid splitting a small programme into campaigns that cannot gather useful conversion evidence. Countries with compatible languages and business goals can share a campaign. Review country-level outcomes before deciding where to split.
That is a planning decision, not a rule that every country needs its own campaign. Google’s creative guidance includes selecting countries that share the relevant language.
Give Each Campaign a Clear Goal
Decide whether the immediate priority is installs, a valuable in-app action or conversion value. Use a separate campaign when a different goal needs its own optimization. For action bidding, avoid blending events with different economic values into one CPA target.
Choose an event that is meaningful and occurs often enough to support decisions. Do not automatically start with install volume if your account already has reliable downstream signals. Equally, a rare purchase event may not provide enough feedback for the planned budget. Review event volume and the available bidding options before committing.
Budget and Bid Settings
Google’s recommended bid-to-budget ratios are at least 50 times the bid for ACi target CPI and pre-registration campaigns, 10 times for ACi target CPA, and 15 times for ACe target CPA. These are campaign setup recommendations, not a promise of profitable delivery. Check whether the account shows budget or bid constraints.
Make material changes deliberately and record what changed. Frequent changes to bids, budgets and goals make results harder to interpret. Google’s setup guidance advises allowing 7 to 14 days for stabilization after significant changes. Also allow for conversion delay before judging recent performance.
Creative Asset Strategy
Use Distinct Assets That Work Together
Provide useful variation in message, visual treatment and format. Each headline, description, image and video should make sense in different combinations. Check the limits for your campaign subtype; App campaigns for installs support up to 20 images and 20 videos per ad group. Filling slots with near-duplicates is not a testing strategy.
Show the App and the Reason to Use It
Use the opening of a video to show a relevant benefit or product experience. Brief the aspect ratios, duration, captions and language needed for the placements. A different crop is a format variation; a different message or use case is a new testing idea. Track those distinctions so the next brief reflects what you learned.
Our KaufDA case describes creator-style testing on TikTok. It can suggest a hypothesis to test elsewhere, but it does not prove that UGC will outperform other formats in Google App Campaigns. Adapt the idea to the product and placement, then assess the result in that campaign.
Smart Bidding Optimization
Match the Bid Strategy to the Outcome
For installs, review the install-focused options available in your account, such as target CPI or Maximize conversions. Target CPA is for a selected action. For target ROAS, verify that values, measurement sources and campaign eligibility support value optimization. Thirty days of tracking alone is not a sufficient readiness test.
Use mature results from a comparable campaign when considering an initial return target. See Google’s initial target ROAS guidance. An aggressive target based on an incomplete recent period can make a campaign difficult to assess.
The Miles Mobility case describes Smart Bidding, measurement and Search campaign work. It is a separate campaign example, not a benchmark or promised outcome for an App campaign.
Check Conversion Data Before Changing Bids
Verify the app measurement integration, linked accounts, imported events and bidding goals. Check for duplicate events, missing values and reporting delay. Keep purchase revenue separate from estimated proxy values. For a subscription app, a trial start is not a paid subscription and its estimated value is not collected revenue.
Agree which event the campaign should optimize, how its value is calculated and when the cohort is mature enough to evaluate. Keep a record of changes so a tracking change is not mistaken for a business improvement. The app growth methodology guide explains the wider decision process.
Implement consent handling for the app and its measurement SDKs. Server-side delivery does not replace consent. Use the platform’s app consent guidance with your product and development teams.
Advanced Optimization Techniques
Connect Web Campaigns to the App Journey
For supported web campaigns, App Connect combines app conversion measurement and deep linking. Confirm campaign and platform compatibility, then test the destination and intended event with the app team. A link that opens the app is one check; a correctly recorded downstream action is another.
Match Store Messaging to the Ad
Where Google Play’s custom store listings fit your setup, use them to align the store message with the campaign’s audience or theme. Keep the app’s actual features and offer consistent. Measure the path from ad to install and subsequent action rather than assuming a tailored listing will improve conversion.
Review Asset Ratings in Context
Google’s asset reports compare assets using ratings such as Low, Good and Best. Wait for adequate traffic rather than replacing an asset on a fixed seven-day deadline. Relative ratings do not establish that an asset caused better business results. Add useful variations where slots remain, or replace weaker assets gradually when they are full. Review campaign outcomes alongside those ratings.
Case Study: NeuroNation’s Google Campaign Work
The NeuroNation case explains how Admiral organized communication ideas and tested them across audiences and markets. The published case reports higher ROAS and purchases alongside lower CPI. Read the case for the specific engagement context.
The useful lesson is to assess acquisition cost together with downstream outcomes. A cheaper install can still be a poor acquisition if it does not lead to a useful action. One client’s results do not establish what another app will achieve, and net cohort revenue is not interchangeable with lifetime value.
Common Mistakes to Avoid
Changing Several Things at Once
Changing the bid, creative, goal and tracking together obscures which change mattered. Keep a change log and review the campaign’s learning status and conversion delay. Correct a verified measurement or delivery failure promptly; avoid treating ordinary daily variation as proof that the strategy has failed.
Testing Variations Without a Question
Decide what each batch should teach you: whether a problem resonates, whether product footage explains the value, or whether a different format fits the placement. Use outcomes after the click where the data supports them. A higher click rate alone does not establish more paying users.
Ignoring Post-Install Quality
Track the steps that matter to the business, such as activation, purchase or paid subscription. Check whether the selected bidding event is a useful signal of that value. There is no universal number of events that makes a measurement setup complete.
Fragmenting the Budget Too Early
Before adding a campaign, explain why it needs a separate goal, market, budget or test. Then check whether the available spend and event volume can support that split. Consolidation is a decision based on the account, not a guarantee that fewer campaigns will perform better.
Google App Campaigns Checklist for 2026
- Measurement: verify the intended events, values, consent behavior, account links and reporting delay.
- Creative: review distinct messages and suitable formats, with enough observations to assess assets.
- Structure: document why campaigns are separated and whether each has a useful goal and viable budget.
- Bidding: check the strategy’s eligibility and signal quality before changing the optimization goal.
- Evaluation: agree the conversion window, cohort maturity and decision that would justify changing spend.
- App journey: inspect store destinations and deep links where relevant. Our Google App Campaigns setup guide provides another starting point.
Need Help Managing App Acquisition?
Share your app store link, target markets, current monthly media spend and the outcome you want to improve. Explain how you measure purchases or subscriptions, and whether you need campaign management, creative testing or measurement support. That context helps us assess where Admiral can contribute.
Explore our app marketing service or discuss your acquisition programme. Media spend, agency fees and creative production are separate scope decisions. If you are changing partners, use our agency handover checklist.
Frequently Asked Questions
What is the minimum budget for Google App Campaigns?
There is no single amount that makes every app campaign viable. Use Google’s bid-to-budget recommendations above, then assess your target event, market economics and available budget. A recommended ratio is not a guarantee of enough profitable customers.
How long does the learning phase last?
Use the campaign’s current status and conversion delay. Google’s setup guidance recommends allowing 7 to 14 days after significant changes for stabilization. Do not assume every edit restarts an identical fixed period.
Should I use target CPA or target ROAS?
Choose around the outcome you can measure reliably. Target CPA focuses on an action; target ROAS uses conversion value. Check value quality and account eligibility before switching. For install volume, review the available install-focused bidding options.
How many creative assets should I provide?
Use the limits for your campaign subtype and prioritize distinct, relevant messages and formats. Avoid an arbitrary minimum count or a library filled with near-duplicates. Review assets after they have enough traffic to be informative.
Can I choose search keywords in App campaigns?
Standard App campaigns do not provide manual keyword targeting. A separate Search campaign gives you keyword controls and may use supported app deep links. Treat that as a different campaign setup with its own measurement checks.
What should I check before increasing budget?
Confirm event accuracy, cohort maturity and the business value of acquired users. Review campaign constraints and creative evidence, then record why the increase is justified. Spend growth, installs and click rates are not substitutes for paying-user economics.
How are App campaigns different from Performance Max?
App campaigns are built for app promotion. Performance Max is a separate cross-channel campaign type and can also use app conversion measurement and deep links through App Connect. The right mix depends on the goal and supported setup, not a rule that Performance Max only measures web conversions.
