App Event Optimization (AEO) is the campaign optimization type in Meta Ads that tells the algorithm to find users likely to complete a specific in-app event, not just install your app.
It sits between install optimization and value optimization in Meta's hierarchy, and understanding how to deploy it correctly separates efficient campaigns from wasteful ones.
Page Contents
What exactly is AEO (app event optimization) in Meta Ads?
AEO is a Meta Ads optimization setting that targets users most likely to complete a specific post-install event, such as a purchase, subscription start, or registration.
Meta's developer documentation confirms AEO uses machine learning trained on your app event data to predict which users will convert beyond the install.
Key insight: AEO optimizes for downstream actions, not just installs, making every dollar work harder toward revenue.
- Targets post-install events, not just installs
- Requires SDK event data to train the model
- Higher CPI but better downstream conversion
- Sits between MAI and VO in Meta's hierarchy
- Available for both iOS and Android campaigns
| Optimization Type | What Meta Optimizes For | Best For |
|---|---|---|
| MAI (Install) | Cheapest install | Top-of-funnel scale |
| AEO (App Event) | Specific in-app event | Mid-funnel quality |
| VO (Value) | Highest predicted LTV | Revenue maximization |
When you select AEO, you choose a target event from your app's SDK event stream. Meta then bids in its auction to reach users whose behavioral signals match past converters for that event.
This is fundamentally different from MAI (Mobile App Install) optimization, which only cares about getting the cheapest install. AEO accepts a higher CPI in exchange for better post-install quality.
According to AppsFlyer's 2024 Cost of App Install report, AEO campaigns on iOS typically run 30-50% higher CPIs than install-optimized campaigns. The tradeoff pays for itself because post-install event rates climb substantially when the algorithm targets downstream behavior.
How does AEO differ from value optimization (VO)?
AEO optimizes for a binary outcome: did the user complete the event or not? VO goes further, predicting how much revenue each user will generate. Meta's optimization documentation specifies that VO requires purchase value data passed through the SDK.
VO demands significantly more conversion volume to exit learning. Per Meta's learning phase guidelines, if your app generates fewer than 50-100 purchase events per week per ad set, AEO typically outperforms VO because the algorithm has enough signal to learn.
How does Meta's algorithm calculate and deliver AEO campaigns?
Meta's algorithm uses your historical app event data, combined with its user-level behavioral graph, to predict which users in your target audience will complete the specified event.
The system needs a minimum of approximately 128 events per week per ad set to exit the learning phase, per Meta's learning phase documentation.
Key insight: AEO needs 128+ weekly events per ad set to exit the learning phase and stabilize performance.
- Learning phase requires ~128 events/week per ad set
- Behavioral signals from Meta's user graph power predictions
- Volatile CPAs during learning, stabilizes after
- Consolidation accelerates learning phase exit
- Each ad set edit resets the learning counter
When you launch an AEO campaign, Meta enters what it calls the "learning phase." Performance during this window is volatile, and CPAs can swing significantly above steady state levels.
The model considers hundreds of signals: engagement patterns across Facebook and Instagram, device type, cross-app activity, and time-of-day behavior. These signals collectively predict whether a given user resembles past converters.
Once the ad set accumulates enough conversions, it exits learning and enters a more stable optimization state. Post-ATT best practices covered on the Mobile User Acquisition Show recommend consolidating AEO campaigns to concentrate event volume and exit learning faster.
What happens during the AEO learning phase?
Expect CPAs to swing wildly for the first 50-75 conversions. Resist the urge to pause or edit the ad set during this window, because each edit resets the learning phase per Meta's documentation.
Budget the learning phase as a sunk cost. If your target CPA is $20, allocate roughly $2,500-$3,000 per ad set for the learning investment, calculated as 128 events at an inflated $20-$25 each during the exploration window.
Which app events should you optimize for with AEO?
Optimize for the event closest to revenue that still generates enough volume to feed the algorithm.
For subscription apps, that's typically "start trial" or "subscribe." For e-commerce, it's "purchase." According to RevenueCat's 2024 State of Subscription Apps report, the median trial-to-paid conversion rate across paywalled apps sits at approximately 53%.
Key insight: Pick the event closest to revenue that still gives you 128+ weekly conversions per ad set.
- Subscription apps: optimize for trial start or subscribe
- E-commerce apps: optimize for purchase or add-to-cart
- Gaming apps: optimize for in-app purchase events
- Avoid optimizing for low-intent events like app open
- Volume threshold trumps funnel proximity
| App Category | Recommended AEO Event | Fallback Event If Volume Is Low |
|---|---|---|
| Subscription / SaaS | Start Trial | Registration |
| E-commerce | Purchase | Add to Cart |
| Gaming (IAP) | In-App Purchase | Level Complete |
| Fintech | Account Funded | Registration |
| Health & Fitness | Subscribe | Complete Onboarding |
The classic mistake is optimizing for an event too far down the funnel. If your app only generates 15 purchases per week, Meta cannot learn effectively. Moving up to "add to cart" or "registration" and layering a retargeting funnel often produces better results.
Conversely, optimizing for events too high in the funnel (like "app open" or "tutorial complete") gives the algorithm a weak quality signal. Cheap events flood in, but downstream LTV suffers.
For subscription apps running Meta Ads, the sweet spot is usually "initiate checkout" or "start trial" because these events carry strong purchase intent while generating sufficient volume.
How much does AEO cost compared to install optimization on Meta?
AEO campaigns typically cost more per install than MAI campaigns, but the cost per paying user drops substantially. According to Liftoff's 2024 Mobile Ad Creative Index, the average iOS CPI on Meta in North America runs $3.80 for MAI versus $6.20 for AEO campaigns.
Need help scaling your mobile app growth? Talk to RocketShip HQ about how we apply these strategies for apps spending $50K+/month on UA.
Key insight: AEO costs more per install but less per paying user, making ROAS the metric that matters.
- AEO CPIs run meaningfully above MAI campaigns
- Cost per revenue event typically drops significantly
- ROAS and CPA matter more than CPI
- iOS AEO is pricier than Android AEO
- Learning phase inflates early costs substantially
| Metric | MAI Campaign | AEO Campaign | Source |
|---|---|---|---|
| Avg iOS CPI (North America) | $3.80 | $6.20 | Liftoff 2024 Creative Index |
| Typical Trial Start Rate | 2-4% | 10-14% | Industry observation |
| Effective Cost Per Trial | $95-$190 | $44-$62 | Derived from above |
The CPI increase scares many early-stage marketers into sticking with install optimization. That's a trap. Cost per revenue event, not cost per install, determines profitability.
Consider a subscription app paying $3.00 CPI on MAI with a 3% trial start rate, effectively spending $100 per trial. The same app on AEO might pay $5.50 CPI with a 12% trial start rate, yielding a cost per trial of $45.83.
That trial is 54% cheaper on the metric that actually drives LTV.
This is precisely why choosing the right bidding strategy requires thinking about downstream economics, not surface-level CPI.
How do you structure AEO campaigns in Meta Ads for best results?
Consolidate aggressively. Run 1-3 ad sets maximum per AEO campaign to concentrate conversion data. Based on testing Advantage+ app campaigns versus manual setups, consolidated structures consistently outperform fragmented ones on CPA because the algorithm receives a stronger learning signal.
Key insight: Fewer ad sets with larger budgets beat many ad sets with small budgets for AEO every time.
- Run 1-3 ad sets maximum per campaign
- Use broad targeting, not interest stacking
- Budget at least 10x target CPA per day
- Start with 4-6 creatives per ad set
- Don't edit ad sets during learning phase
The ideal AEO structure in 2026 looks radically simpler than what most marketers expect. Broad targeting now outperforms interest-based targeting on Meta for most app categories, so audience fragmentation is unnecessary.
Start with one campaign, one ad set, broad targeting (age and geo only), and 4-6 creatives. Set your daily budget to at least 10x your target CPA. A $30 CPA goal means budgeting at least $300/day per ad set.
As Eric Seufert discusses on MobileDevMemo, concentrating spend on fewer structures accelerates algorithmic learning. The same principle applies within Meta: concentrate into fewer ad sets rather than fragmenting across many.
Should you use Advantage+ app campaigns or manual structure for AEO?
Advantage+ app campaigns automate much of the structuring decision. For apps with sufficient event volume (128+ events/week), Advantage+ often matches or beats manual AEO setups because Meta can allocate budget dynamically across audiences.
Manual AEO campaigns give you more control over creative testing and budget pacing. Our comparison of the two approaches covers the tradeoffs in detail. The short answer: test both and let CPA data decide.
How does AEO work with SKAdNetwork and iOS privacy changes?
Post-ATT, AEO on iOS relies on SKAdNetwork conversion values to receive aggregated event signals. According to AppsFlyer's 2024 SKAN performance benchmarks, apps using optimized conversion schemas see approximately 25% better modeled CPA accuracy versus those using default mappings.
Key insight: Your SKAN conversion value schema directly determines how well AEO can optimize on iOS.
- SKAN conversion values feed iOS AEO models
- Map your AEO event into the first SKAN window
- Optimized schemas improve modeled CPA accuracy
- Android AEO avoids most iOS privacy complications
- SKAN 4.0 coarse values limit fine-grained optimization
SKAN 4.0 introduced three conversion windows and coarse values, making your conversion value mapping critical. A poorly configured schema wastes the limited signal SKAN provides.
Map your primary AEO target event into the first conversion window (0-2 days). Secondary monetization signals belong in later windows. Tools like AppsFlyer's SKAN solution and Adjust's conversion value hub help automate schema optimization.
Android AEO avoids most of these complications because Google's Privacy Sandbox is less restrictive than ATT. Expect Android AEO to deliver cleaner data signals and lower CPAs, which is consistent with the platform pricing differences covered in Liftoff's 2024 Creative Index.
When should you switch from install optimization to AEO?
Switch to AEO once your app consistently generates 128+ target events per week across all ad sets combined, per Meta's learning phase requirements. Launching AEO before reaching this volume threshold starves the algorithm of signal.
Key insight: Don't switch to AEO until your event volume can sustain Meta's learning phase requirements.
- Start on MAI to build event volume first
- Switch when you hit 128+ weekly target events
- Run MAI and AEO side-by-side for 2-3 weeks
- Compare cost per event, not cost per install
- Shift budget toward the winner gradually
Early-stage apps with minimal installs should start on MAI to build a user base and generate enough downstream events for AEO to learn from. Once SDK event data reaches the volume threshold, begin testing AEO alongside your existing MAI campaigns.
Run both simultaneously for at least 2-3 weeks. Compare cost per target event (not CPI) between the two. According to budget guidelines for app install testing, allocate enough budget to each approach so both can exit the learning phase.
Gradually shift budget toward whichever optimization delivers better cost per revenue event. Most apps find AEO wins on this metric once the algorithm has enough data, but the transition timeline varies by category and event volume.
How do creatives impact AEO campaign performance?
Creative quality is the single largest lever for AEO performance because it determines who engages with your ad, which in turn shapes the audience the algorithm learns from.
Per Liftoff's 2024 Creative Index, video creatives generate 22% lower CPA on average than static images for app event optimization campaigns.
Key insight: Creatives pre-qualify users for AEO, so performance hinges on creative messaging as much as targeting.
- Video creatives outperform static for AEO
- Pre-qualify users through clear value messaging
- Test 4-6 creatives per ad set across formats
- Rotate new creatives every 7-14 days
- Creative fatigue hits AEO audiences faster
RocketShip HQ's approach to AEO creative centers on pre-qualifying users within the ad itself. If you're optimizing for trial starts, your creative should make the value proposition of the trial crystal clear. Vague brand awareness ads attract low-intent users that inflate CPA.
Test 4-6 creatives per ad set, spanning formats: 9:16 vertical video for Reels placements, 1:1 square for Feed, and static for Audience Network. Each format reaches different user pools at different CPMs. Structuring creative across placements unlocks incremental reach that a single format misses.
Rotate in fresh creatives every 7-14 days. Creative fatigue accelerates in AEO campaigns because the algorithm targets a narrower, higher-value audience that sees your ads more frequently.
Can you use AEO with custom product pages on iOS?
Yes, pairing AEO with custom product pages (CPPs) creates a powerful conversion funnel by aligning the App Store landing page with your ad messaging. According to Apple's Search Ads documentation, CPPs can drive measurable improvements in tap-to-install conversion rates versus default store listings.
Key insight: Custom product pages extend AEO's intent signal from ad click through App Store conversion.
- CPPs match store page to ad messaging
- Subscription apps: emphasize trial on the CPP
- Consistent user journey boosts install conversion
- Pair with Apple Search Ads for reinforcement
- Up to 35 CPPs available per app on iOS
When a user clicks your AEO ad and lands on a generic App Store page, there's a message mismatch. The ad promised a specific value; the store page shows something generic. CPPs solve this by matching the store experience to the ad creative.
For subscription apps, create CPPs that emphasize the trial offer shown in your ad. E-commerce apps should mirror the specific product category or promotion. This alignment keeps the user journey consistent from impression through install.
Coordinating Apple Search Ads with Meta Ads amplifies this further. Users exposed to your Meta AEO campaign who later search organically can be captured by Search Ads pointing to the same CPP, reinforcing the message.
What mistakes kill AEO campaign performance on Meta?
The most common AEO killer is insufficient event volume combined with excessive campaign fragmentation. Per Meta's best practices, ad sets that never exit learning phase waste budget and deliver CPAs 30%+ higher than optimized ad sets.
Key insight: Fragmentation and premature edits are the two fastest ways to destroy AEO performance.
- Fragmented ad sets starve the algorithm of data
- Editing during learning phase resets progress
- Wrong event choice undermines optimization quality
- Budget changes over 20% trigger re-learning
- Lack of creative rotation accelerates fatigue
Splitting budget across too many ad sets is the number one structural mistake. Five ad sets at $50/day each will underperform one ad set at $250/day because none of the five can accumulate enough events to exit learning.
Editing ad sets during the learning phase is the second killer. Changing budget by more than 20%, swapping creatives, or adjusting targeting resets Meta's learning counter. Patience is essential during the first 50-75 conversions.
Optimizing for the wrong event ranks third. As detailed in the event selection section above, choosing an event with too little volume or too little purchase intent undermines the algorithm. A comprehensive guide to Meta app ads covers these structural pitfalls in more depth.
AEO sits at the intersection of algorithmic power and practitioner discipline. Get the event choice, volume threshold, and campaign structure right, and it consistently delivers lower cost per revenue event than install optimization.
Start with one consolidated ad set, feed it enough budget to exit learning, and let the algorithm do what it does best.
Frequently Asked Questions
Does AEO work for Android apps or only iOS?
AEO works on both platforms, but Android AEO typically delivers 15-30% lower CPAs than iOS AEO, according to Liftoff's 2024 Creative Index, because Android doesn't face the same ATT signal loss. Android's richer data environment helps Meta's algorithm learn faster.
Can you run AEO and MAI in the same Meta account without auction overlap?
Yes, but expect some audience overlap. Per Meta's audience documentation, separate optimization types bid into the same auction. Using audience exclusions or running them in a single Advantage+ campaign with auto-allocation minimizes waste from self-competition.
What SDK events does Meta support for AEO?
Meta supports all standard app events defined in the Facebook SDK event reference, including purchase, subscribe, start trial, add to cart, complete registration, and achievement unlocked. Custom events work too, but standard events receive better model support because Meta has billions of training signals for them.
How long should you wait before judging AEO campaign results?
Wait at least 7 days after exiting the learning phase before making performance judgments. Per Meta's guidelines, the learning phase itself takes approximately 50 conversions. Combined with post-learning stabilization, budget at least 2-3 weeks total evaluation time.
Is AEO or VO better for high-LTV subscription apps?
If you generate 100+ purchase events per week per ad set, VO typically outperforms AEO for high-LTV subscription apps. Below that threshold, AEO delivers more stable results. RevenueCat's 2024 report shows that top-quartile subscription apps generate enough revenue events to sustain VO's data requirements.
Can AEO campaigns target lookalike audiences or only broad audiences?
AEO supports both, but broad targeting consistently outperforms lookalikes and interest audiences in 2026 because Meta's algorithm already identifies high-propensity users. Lookalikes add a constraint that often limits reach without improving quality, especially at daily budgets above $200.
What's the minimum budget needed to test AEO effectively?
Budget at minimum 128 events multiplied by your expected CPA. Per testing budget frameworks, a $25 CPA target means allocating roughly $3,200 to exit the learning phase on a single ad set. Underfunding guarantees the algorithm never learns properly.
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)
- Advantage+ app campaigns vs manual campaigns for Meta app installs (2026)
- How Do Apple Search Ads and Meta Ads Work Together?
- Broad targeting vs interest-based targeting for Meta app campaigns (2026)
- Does Broad Targeting Outperform Interest Targeting on Meta?
