App Reengagement vs. New Installs: How to Split Your Q3 Budget
Most app teams treat the reengagement vs. new-installs question like a coin flip. They either default to acquisition because "growth" sounds better in a board deck, or they pivot hard to reengagement after a bad CPI month and call it "efficiency." Neither is a strategy.
This post gives you a repeatable framework for making that call with your actual data—specifically in Q3, where seasonal patterns, platform auction dynamics, and user behavior create a different environment than the rest of the year.
Why Q3 Changes the Equation
Q3 (July–September) sits in a tricky spot on the mobile advertising calendar. Advertiser spend is lower in July and August because most big consumer brands are saving budget for Q4. That suppresses CPMs and CPIs on Meta, TikTok, and Google App Campaigns—which makes new user acquisition relatively cheaper than it is in October or November.
At the same time, summer churn is real. Apps in fitness, productivity, and B2B SaaS typically see a dip in DAU through July as users are traveling or distracted. That lapsed-user pool is large and warm—these are people who already downloaded your app, already went through onboarding, and already formed some intent. Reengaging them costs less than acquiring someone cold, but only if your reengagement targeting and creative are sharp.
So Q3 isn't a time to pick one channel and go all in. It's a time to be precise about which users fall into which bucket and price both correctly.
The Four Variables That Drive the Split
Before you land on any percentage, you need four numbers from your analytics stack. If you don't have clean data on all four, fix that before you spend another dollar on either channel.
1. Reactivation rate of lapsed users Of users who churned in the past 60–90 days, what percentage came back after a reengagement push (push notification, email, retargeting ad)? If this number is very low—say under 5%—your lapsed users may be genuinely gone, and dollars spent chasing them are mostly wasted.
2. Cost per reactivation vs. cost per new install Pull your last 90 days of paid data and calculate what you actually spent per reactivated user vs. per new install. In our engagements, reactivation typically runs 30–60% cheaper per event than new install—but that gap narrows if your lapsed audience has already been hammered with retargeting and is showing ad fatigue.
3. LTV delta between reactivated and new users A reactivated user who has already churned once may have a lower predicted LTV than a new user who's in a higher-intent acquisition cohort. Check your cohort data. If reactivated users churn again within 30 days at a high rate, your reengagement unit economics look worse than they appear at the cost-per-event level.
4. Audience size remaining in your lapsed pool Reengagement has a ceiling. Once you've exhausted your addressable lapsed audience, marginal returns collapse. If your lapsed pool is small or already heavily retargeted, acquisition has to carry more of the load regardless of CPI.
The Budget Split Framework
This table maps four common app situations to a recommended starting split. These aren't universal rules—they're starting points you adjust based on the four variables above.
| App Situation | Recommended Reengagement % | Recommended New Installs % | Primary Rationale |
|---|---|---|---|
| Early-stage app (< 6 months old, small lapsed pool) | 10–15% | 85–90% | Not enough lapsed users to move the needle; growth is the priority |
| Growth-stage app with high Q3 churn | 30–40% | 60–70% | Lapsed pool is large and recent; cost advantage is real |
| Mature app with plateauing organic growth | 40–55% | 45–60% | Acquisition costs are high relative to reactivation; efficiency matters |
| Apps with strong seasonal Q4 (e-commerce, gifting) | 20–25% | 75–80% | Use Q3 to build new install base ahead of Q4 conversion push |
The single biggest mistake teams make is applying a mature-app split to an early-stage app, or vice versa. Know which row you're in.
Not sure which row you're in? Our mobile app marketing team can audit your current funnel and give you a concrete recommendation based on your actual cohort data.
How to Structure Reengagement Campaigns That Actually Work
Reengagement isn't "send a push notification saying 'We miss you.'" That approach produces noise, not reactivations. Here's what actually moves the metric:
Segment by recency and last action. A user who churned 2 weeks ago after completing onboarding but never completing a core action is a different audience from a user who was active for 3 months and then went quiet. The message, channel, and offer need to match the segment.
Prioritize push + paid retargeting together. Push alone reaches only users who haven't disabled notifications—typically 40–60% of your lapsed base, and skewed toward less-churned users who still have the app installed. Paid retargeting on Meta or Google reaches the rest. Running both in coordination with consistent creative raises reactivation rates meaningfully compared to either channel alone.
Give them a reason to come back that's tied to product, not just promotion. "Here's 20% off" works once. A notification that says "Your [feature] data is ready" or "3 people you follow just joined" pulls users back into the product loop they left. If you don't have a hook like that, the deeper fix is a product one—not a marketing one.
Set a reengagement frequency cap. If a lapsed user hasn't reactivated after 3–4 touches across channels over 2 weeks, remove them from active reengagement for at least 30 days. Continued spend on unresponsive segments destroys ROAS and trains the platforms' algorithms on bad signal.
How to Structure New Install Campaigns for Q3
The lower-CPM environment in July and August creates a window. Use it deliberately.
Test creatives aggressively in July. Q3 is a low-cost environment for learning. Run more creative variants than you normally would, get statistically significant signal on what converts, and then scale the winners into August and into Q4 with a proven creative library. Teams that use Q3 purely as an efficiency quarter come into Q4 with stale creative—that's a costly mistake.
Tighten your acquisition funnel before scaling spend. More installs from paid acquisition don't fix a broken onboarding flow. If your D7 retention is below your category benchmark, adding install volume amplifies that problem. Audit the funnel first. Check out our post on 5 app marketing strategies to skyrocket user retention in 2026 for a concrete framework on the retention side before you scale acquisition.
Layer ASO with paid. Paid install campaigns lift branded search volume on the App Store and Google Play. If your ASO isn't optimized—screenshots, keyword metadata, ratings—you're leaving organic installs and conversion rate improvement on the table while paying for paid traffic to compensate. The two channels compound when they're aligned.
Don't ignore Apple Search Ads in Q3. With larger advertisers pulling back on ASA in summer, CPTs (cost per tap) tend to drop. That's a window for growth-stage apps that can't normally compete on high-volume keywords to get impressions and installs at better rates. Run exact-match campaigns on your highest-intent keywords in July and track blended CPA carefully.
For a broader look at how to build a sustainable acquisition engine, the 2026 mobile user acquisition strategy post covers the channel mix in more depth.
Tracking the Split Over Time
Set a checkpoint every 3 weeks in Q3 to reassess the split—not just the total budget. If your lapsed pool is draining faster than expected, shift dollars toward acquisition. If new install CPIs spike (they sometimes do in late August as Q4 planning kicks in early for larger advertisers), shift the other way.
This isn't about chasing every data point. It's about having a defined rule for when you'll move budget and what data triggers that move. Without that rule, you're making the decision emotionally under pressure.
FAQ
How do I know if my lapsed users are worth reengaging at all?
Run a small test before committing budget. Identify users who churned in the last 60 days, send a 2-week reengagement sequence across push and paid retargeting, and measure reactivation rate and 30-day retention of reactivated users. If reactivation rate is under 4–5% or reactivated users churn again quickly, redirect that budget to acquisition.
What counts as a "lapsed" user for targeting purposes?
The definition varies by app category. For a daily-habit app (fitness, meditation), 7–10 days of inactivity is lapsed. For a weekly-use app (budgeting, task management), 30 days might be the threshold. For a seasonal or transactional app, 90 days may make more sense. Set your window based on your own engagement patterns, not industry defaults.
Should we remarket to users who never completed onboarding?
Typically yes, but with a different message and a separate budget line. These users have high intent (they installed) but hit friction early. A reengagement ad that addresses that specific friction—"Need help getting started?"—outperforms a generic win-back message. In our engagements, this segment often has a higher reactivation rate than users who onboarded but later churned.
Does this framework apply to both iOS and Android?
The logic applies to both platforms, but the execution differs. iOS retargeting is constrained by ATT opt-in rates, which means your addressable retargeting audience on iOS is smaller than your full lapsed base. Factor that in when estimating reengagement reach. On Android, Google's Privacy Sandbox changes are still evolving—check current audience reach capabilities before allocating heavily to Android retargeting.
How much of our Q3 budget should go to creative production vs. media?
A reasonable starting point is 10–15% of total paid budget on creative production, with the remainder on media. In Q3, when you're testing more creative variants than usual, that ratio can shift toward 20% on creative. Skimping on creative to maximize media spend is a false economy—especially in reengagement, where creative relevance to the specific user segment drives most of the performance difference.
What's the biggest signal that our split is wrong?
If your blended CPA is rising while ROAS is falling over a 4-week window, you're either over-indexing on an exhausted audience or under-investing in the channel that has remaining headroom. Pull the data by channel, compare cost-per-reactivation vs. cost-per-install against their respective LTV projections, and move money toward the channel with the better margin—not the one that feels safer.
If you want a second set of eyes on your Q3 allocation before you lock the budget, our mobile app marketing team works through exactly this kind of analysis with growth-stage apps. Book a 30-minute call and we'll tell you straight where the leverage is.
