How to run Meta app ads in multiple countries

Running Meta app ads across multiple countries is one of the fastest ways to scale a mobile app, but it is also one of the easiest ways to waste budget if your structure is wrong.

At RocketShip HQ, we have managed multi-geo campaigns across 50+ countries for B2C apps spanning gaming, health, and finance verticals. The difference between a well-structured multi-country campaign and a poorly structured one can be 2-3x in CPI efficiency.

In this guide, you will learn exactly how to set up your geo-tiering, campaign structure, creative localization, and budget allocation so every dollar works harder across every market.

Prerequisites: You should have a working Meta Ads Manager account with app install campaign experience, a functioning mobile measurement partner (MMP) like AppsFlyer or Adjust, at least one proven creative concept that has worked in your home market, and a minimum monthly budget of $10K-$15K to make multi-geo testing statistically viable.

Step 1: Build Your Geo-Tier Strategy Based on LTV and CPI Data

Before you touch Ads Manager, you need to segment your target countries into tiers based on the relationship between cost per install and lifetime value. This is the single most important structural decision in multi-geo campaigns because it determines how you allocate budget, set bids, and measure success. As of Q1 2026, median Meta CPIs by app category, with wide variance across geos—making tier-based segmentation essential.

Most apps we work with at RocketShip HQ use a 3-tier system, though some scale to 4 or 5 tiers.

Define Tier 1: High LTV, High CPI markets

Tier 1 typically includes the US, UK, Canada, Australia, and parts of Western Europe (Germany, France, Netherlands). CPIs here range from $2–$8 for casual games and $15–$50+ for subscription apps. These markets justify higher spend because LTV supports it. Expect 60-70% of your total budget to concentrate here.

Define Tier 2: Moderate LTV, Moderate CPI markets

Tier 2 often includes Southern and Eastern Europe (Spain, Italy, Poland), parts of Latin America (Mexico, Brazil, Argentina), and select Asian markets (Japan, South Korea, Taiwan). CPIs are typically 40-60% lower than Tier 1, but LTV follows a similar ratio. These markets offer volume scaling once Tier 1 is optimized.

Define Tier 3: Low LTV, Low CPI markets

Tier 3 covers Southeast Asia (Indonesia, Philippines, Vietnam, Thailand), India, and parts of Africa. CPIs can be as low as $0.10–$0.50, which looks attractive until you realize LTV might be $0.30–$1.00. Only run Tier 3 if your monetization model (ad-based, for example) supports it or you need volume for algorithm training.

Use your MMP data to validate tiers quarterly

Pull cohorted LTV data at Day 7, Day 30, and Day 90 by country from your MMP. Calculate LTV-to-CPI ratios for each market. Any country with a ratio above 1.5x at Day 30 belongs in a higher tier than its CPI alone would suggest. Because ensure your MMP attribution is properly configured, with gaming at the lower end and fintech higher, ensure your MMP attribution is properly configured before making tier decisions. Revisit tiers every quarter because currency fluctuations and competitive dynamics shift market economics.

Do not rely on industry benchmarks alone to build your tiers. We have seen apps where Brazil outperforms Germany on LTV-to-CPI ratio because their product resonated uniquely with that audience. Your own data is the source of truth.

Step 2: Structure Campaigns by Tier, Not by Individual Country

The biggest structural mistake in multi-geo advertising is creating a separate campaign for every single country. This fragments your budget and starves Meta's algorithm of the data it needs to optimize. Instead, group countries by tier into campaigns and use ad set level controls for more granular targeting.

This approach gives the algorithm enough conversion signal per campaign while still letting you manage budgets by market economics.

Create one campaign per tier

Run a Tier 1 campaign, a Tier 2 campaign, and a Tier 3 campaign. Set campaign-level budgets proportional to each tier’s expected return. A typical split is 60% Tier 1, 25% Tier 2, 15% Tier 3. Use Advantage Campaign Budget (CBO) within each tier so Meta can shift spend toward the best-performing ad sets within that tier. CBO campaigns typically deliver 10-20% lower CPAs compared to ABO setups after exiting learning phase, making them ideal for multi-geo structures.

Separate ad sets by language group within each tier

Within your Tier 1 campaign, you might have an English-speaking ad set (US, UK, CA, AU) and a German-speaking ad set (DE, AT, CH). This ensures your creative language matches the audience without splitting budget too thin. Language alignment is more important than individual country separation for Meta's delivery system.

Consider isolating high-spend countries into their own ad sets

If one country (usually the US) represents more than 40% of your total spend, give it a dedicated ad set or even a dedicated campaign. This prevents the algorithm from over-allocating to the cheapest installs in a multi-country ad set while starving your most valuable market. At RocketShip HQ, we almost always run the US as a standalone.

When grouping countries, be cautious about the perils of asset stuffing. Throwing every country and every creative into one ad set prevents the algorithm from identifying appropriate audience segments. Thematic separation by language or region is essential.

Step 3: Localize Creatives Beyond Just Translation

Translation is table stakes. True localization means adapting the creative concept, visual cues, and emotional hooks to resonate with each market. Since creative quality drives performance variance, investing in culturally-adapted creatives rather than just translated text becomes critical for multi-geo success.

For example, shifting copy based on psychological profiling improved IPM from 0.97 to 2.4 for one solitaire game. The same principle applies across geos: what motivates a user in Japan is fundamentally different from what motivates a user in Brazil.

Adapt emotional hooks by culture

As highlighted by Tactile Games' approach with Lily's Garden, exploring emotions like sadness and anxiety when competitors relied on 'funny or cute' drove massive creative differentiation. Apply this thinking per geo: competitive creative landscapes differ by country, so audit what local competitors are running and find the emotional white space.

Localize visual elements, not just text

Adjust character appearances, color palettes, and cultural references. In Japan, cute or detailed UI screenshots perform well. In Brazil, bold colors and social proof drive engagement. In Germany, functional and precise messaging outperforms emotional appeals. These differences can account for 30-50% variations in IPM.

Use native speakers for voiceover and copy

Machine translation is obvious to native speakers and erodes trust. Invest in native copywriters or at minimum use native review for all ad copy. For video ads with voiceover, always use native speakers. The cost difference is marginal ($50–$200 per language) but the performance difference can be 20-40% in CTR.

Match creatives to Custom Product Pages per geo

Pair localized creatives with Custom Product Pages on the App Store that mirror the same messaging and visual theme. This continuity from ad to store listing improves conversion rate by 10-25% compared to sending all traffic to your default store listing.

Start with your top 2-3 performing concepts from your home market and localize those first. Do not create entirely new concepts for new geos until you have validated that your winners do not transfer. In our experience, about 40% of top-performing creatives in the US also perform well in other Tier 1 markets with proper localization.

Step 4: Configure Language Targeting and Audience Settings Correctly

Meta's language targeting and audience settings behave differently than most advertisers expect, especially in multilingual countries. Getting these settings wrong can either restrict your reach dramatically or serve ads in the wrong language to the wrong people.

We recommend broad targeting over interest targeting for most multi-geo app campaigns because Meta's algorithm is better at finding app installers than manual interest stacking, particularly at scale.

Use language targeting only when your creative is in a non-default language

If you are targeting France with French-language creatives, you generally do not need to add a French language filter because the geo itself handles it. But if you are targeting the US with Spanish-language creatives to reach Hispanic audiences, you must add Spanish as a language target.

The rule: use language targeting when the creative language differs from the country's primary language.

Handle multilingual markets with dedicated ad sets

Countries like Canada (English/French), Belgium (Dutch/French), Switzerland (German/French/Italian), and India (Hindi/English plus regional languages) need separate ad sets per language with corresponding creatives. Do not run English and French creatives in the same Canadian ad set. Split them so the algorithm matches the right creative to the right user.

Set exclusions to prevent geo overlap

If you run English-speaking Tier 1 and a separate US-only campaign, exclude the US from your English-speaking Tier 1 ad set. Overlapping audiences across campaigns causes you to bid against yourself, inflating CPIs by 10-20%. Audit for overlap monthly using the Audience Overlap tool in Ads Manager.

In our experience, leaving language targeting blank (while using geo targeting and language-appropriate creatives) often delivers better results than restricting by language. Meta's algorithm is surprisingly good at matching the right creative to the right user based on behavioral signals. Test both approaches.

Step 5: Set Budgets and Bids Accounting for CPI Variance Across Geos

CPI differences between countries can span 50x or more (from $0.10 in India to $5+ in the US for the same app). Your budget allocation and bidding strategy must account for this variance or you will either overspend in cheap markets or starve expensive ones.

The goal is to maximize total LTV-positive installs across all geos, not to maximize raw install volume.

Allocate budgets proportional to LTV opportunity, not CPI

A common mistake is giving equal budgets to each tier. Instead, calculate the total addressable LTV per tier by multiplying expected installs by expected LTV, then allocate budget to maximize total return. Tier 1 usually gets the lion's share because even though CPIs are higher, the LTV-to-CPI ratio is often the best.

Use cost caps or bid caps for Tier 1 markets

In expensive markets like the US, set cost caps at your target CPI to prevent Meta from overspending on low-quality installs. Start your cost cap at 10-15% above your target CPI to give the algorithm room to learn, then tighten over 2-3 weeks.

For Tier 2 and 3, lowest cost bidding often works fine because the CPI ceiling is naturally lower.

Account for currency and reporting

Set your ad account currency to USD (or your primary reporting currency) to simplify cross-geo comparison. If you have local entity ad accounts in other currencies, use your MMP as the source of truth for cross-geo CPI and ROAS comparisons. Because SDK attribution gaps on iOS when attribution is fragmented across geos, implement Conversions API in every market to ensure accurate cross-geo performance measurement. Currency fluctuations can swing CPIs by 5-15% quarter over quarter in emerging markets.

Use RocketShip HQ's Weighted Anomaly Scoring approach to monitor multi-geo performance. Weight metric changes by business impact: abs(% change) x sqrt(spend). A 15% ROAS drop on your $5K/day US campaign is far more critical than a 40% ROAS drop on your $200/day Philippines campaign. This eliminates over 70% of false alarms when monitoring dozens of geos simultaneously.

Step 6: Test and Iterate Creatives Across Geos Systematically

Multi-geo creative testing multiplies complexity fast. If you have 5 concepts across 4 language groups, that is 20 creative variants before you even account for format differences. You need a disciplined testing framework or you will burn budget on inconclusive tests.

Be especially mindful of the hidden testing costs of scaling creative output: more creative variants require proportionally larger test budgets to reach statistical significance.

Test concepts in your highest-spend market first

Run all concept tests in the US (or your highest-spend market) first because you will reach statistical significance fastest. A concept that wins in the US has about a 40-50% chance of winning in other Tier 1 markets after localization. Only localize proven winners to avoid wasting localization costs on concepts that do not work.

Control the number of creatives per ad set

Do not overload ad sets with too many creatives. We recommend keeping 3-6 creatives per ad set so each gets enough impressions to generate meaningful performance data. In smaller geos, this might mean running even fewer variants.

Run geo-specific tests for Tier 2 and Tier 3 separately

Some concepts perform completely differently in non-English markets. Once you have localized your top 3 US winners, run them as a controlled test in each tier for 7-14 days. If a concept that won in the US underperforms in a specific geo by more than 30% on CPI, develop a geo-specific alternative.

Track creative performance by geo in a shared creative scorecard. We use a simple matrix: concept vs. geo with CPI, IPM, and Day 7 ROAS in each cell. This makes it immediately obvious which concepts are universal winners and which are geo-specific.

Step 7: Scale Winners and Cut Losers Using a Weekly Review Cadence

Multi-geo campaigns require more frequent monitoring than single-market campaigns because you have more variables in play. Establish a weekly review cadence with clear rules for scaling, pausing, and iterating. Without this discipline, underperforming geos can quietly drain budget for weeks.

Review Tier 1 markets every 3-4 days

Because Tier 1 represents the majority of spend, review these markets twice per week. Look at CPI, CPE (cost per event for downstream actions), and ROAS. Any ad set that has spent 3x your target CPI without hitting your CPA goal should be paused or have its creatives refreshed.

Review Tier 2 and 3 markets weekly

These markets have lower spend, so they need more time to accumulate data. Review weekly and make decisions based on 7-day rolling averages, not daily fluctuations. For Tier 3, also monitor install quality metrics (Day 1 retention, session length) because cheap installs often have poor engagement.

Reallocate budget across tiers monthly

Every month, compare actual LTV-to-CPI ratios across tiers against your plan. If Tier 2 is outperforming Tier 1 on efficiency, shift 10-15% of budget from Tier 1 to Tier 2. Avoid dramatic shifts (more than 20% in a single month) because they disrupt Meta's learning phase.

Set up automated rules in Meta Ads Manager to pause ad sets that exceed 2x your target CPI after spending a minimum threshold. This acts as a safety net between your manual review cycles and prevents runaway spend in underperforming geos overnight.

Common Mistakes to Avoid

1. Running one campaign per country: This fragments your budget across too many campaigns, starving each one of the 50+ conversions per week Meta needs for proper optimization. Group countries by tier and language instead.
2. Optimizing for the cheapest CPI across all geos: A $0.15 CPI in India looks great until you realize the LTV is $0.20. Always optimize for LTV-to-CPI ratio or ROAS, never raw CPI. We have seen advertisers waste $50K+ chasing volume in low-LTV markets. Testing new creative concepts monthly than those running just 3-5 creatives, so combining geo-specific creative iteration with LTV-focused optimization delivers the best results.
3. Using machine-translated creatives without native review: Users can spot poor translation instantly, and it destroys credibility. This is especially damaging for subscription apps where trust is essential to conversion. Budget $100–$300 per language for native copy review.
4. Ignoring audience overlap between campaigns: If your US campaign and your 'English-speaking Tier 1' campaign both target the US, you are bidding against yourself. Always check for geo exclusions and use the Audience Overlap tool. This single mistake can inflate CPIs by 15-25%.
5. Scaling to new geos before optimizing core markets: Adding new countries feels like progress, but spreading budget thin across 20 countries before you have profitable, stable performance in your top 3-5 markets just dilutes your data and slows learning. Nail Tier 1 first, then expand.

Running Meta app ads across multiple countries is a powerful growth lever, but only if your structure matches the complexity. Start by building clear geo tiers based on your own LTV and CPI data. Structure campaigns by tier with language-based ad sets.

Localize creatives beyond translation by tapping into cultural and emotional nuances. Set budgets proportional to LTV opportunity, not install volume. And maintain a disciplined weekly review cadence with weighted anomaly scoring to focus on what actually moves the needle.

At RocketShip HQ, we have seen well-structured multi-geo campaigns deliver 30-50% more LTV-positive installs at the same budget compared to poorly structured ones. The framework above is exactly how we approach it for the apps we manage.

Start with 3-5 countries in your top tier, prove the structure works, then expand tier by tier.

Looking to scale your mobile app growth with performance creative that delivers results? Talk to RocketShip HQ to learn how our frameworks can work for your app.

Not ready yet? Get strategies and tips from the leading edge of mobile growth in a generative AI world: subscribe to our newsletter.

Related Reading

• Meta Ads for mobile apps: the complete playbook (comprehensive guide)
• Does Broad Targeting Outperform Interest Targeting on Meta?
• What Are Custom Product Pages and How Do They Improve Meta Ad Performance?
• How Many Creatives Should You Run Per Meta Ad Set?
• Meta’s Ad Auction Actually Work