Understanding ACOS and Key Advertising Metrics for Apps

Running paid user acquisition campaigns without understanding your advertising metrics is like driving with your eyes closed — you might move forward, but you have no idea whether you're heading toward profitability or a cliff. Mobile app advertising has its own vocabulary of metrics, each telling a different part of the story about campaign performance.

ACOS (Advertising Cost of Sales) is one of the most important yet frequently misunderstood metrics in app advertising. Originally popularized by Amazon advertising, ACOS has become a standard measure for evaluating the efficiency of ad spend relative to revenue generated. But ACOS is just one piece of a larger metrics ecosystem that includes CPI, CPA, ROAS, LTV, and more.

This guide explains every advertising metric that matters for mobile app campaigns, how to calculate them, what benchmarks to target, and how to use them to make better spending decisions.

ACOS: Advertising Cost of Sales

What Is ACOS?

ACOS measures the percentage of revenue spent on advertising:

ACOS = (Ad Spend ÷ Revenue Generated) × 100

Example: You spend $1,000 on ads and generate $4,000 in revenue from acquired users.

ACOS = ($1,000 ÷ $4,000) × 100 = 25%

A 25% ACOS means you spend $0.25 in advertising for every $1.00 of revenue.

ACOS vs. ROAS

ACOS and ROAS (Return on Ad Spend) are inverse measurements of the same thing:

ROAS = Revenue ÷ Ad Spend
ACOS = Ad Spend ÷ Revenue × 100

ROAS = 1 ÷ (ACOS / 100)
ACOS = (1 ÷ ROAS) × 100

Which to use? ROAS is more common in mobile app advertising. ACOS is more common in Amazon and e-commerce. Both convey the same information — choose whichever your team and tools report by default.

What's a Good ACOS?

Target ACOS depends on your margins:

Target ACOS = (Revenue - COGS - Operating Costs) ÷ Revenue × 100

For subscription apps: Revenue is recurring but has platform commission (30% year 1, 15% year 2+). If your gross margin after commission is 70%:

• Target ACOS: Under 70% to be profitable
• Healthy ACOS: 20-40% (leaving room for operational costs)

For ad-supported apps: Revenue is lower per user but has no direct COGS per user:

• Target ACOS: Under 100% (breakeven)
• Healthy ACOS: 30-60%

For IAP-driven apps: Revenue varies widely by user:

• Target ACOS: Depends heavily on whale economics
• Healthy ACOS: 25-50% for gaming, 15-35% for non-gaming

ACOS by Time Window

ACOS changes dramatically depending on your measurement window:

Key insight: Don't judge campaign profitability on Day 7 ACOS. Subscription apps in particular need 30-90 days for ACOS to reach its steady state.

The Complete Advertising Metrics Framework

Acquisition Metrics

CPI (Cost Per Install)

CPI = Ad Spend ÷ Number of Installs

What it tells you: How much you pay for each new app install.

Benchmarks (US market):

CPI limitations: CPI tells you acquisition cost but nothing about user quality. A $1 CPI that delivers users who never open the app is worse than a $5 CPI that delivers users who subscribe.

CPA (Cost Per Action)

CPA = Ad Spend ÷ Number of Target Actions

Common target actions:

• Registration/sign-up
• Tutorial completion
• First purchase
• Subscription start
• Level 10 reached (gaming)

Why CPA > CPI: CPA measures the cost of acquiring a user who actually does something valuable. It's a better indicator of campaign quality than CPI alone.

Example:

• Campaign A: CPI = $2, but only 20% register → CPA (registration) = $10
• Campaign B: CPI = $4, but 60% register → CPA (registration) = $6.67
• Campaign B is more efficient despite higher CPI.

CPT (Cost Per Tap / Cost Per Click)

CPT = Ad Spend ÷ Number of Taps/Clicks

What it tells you: How much you pay for each user who interacts with your ad. Used primarily in Apple Search Ads.

Benchmarks: $0.50-$3.00 for most categories (Apple Search Ads).

Engagement Metrics

TTR (Tap-Through Rate)

TTR = Taps ÷ Impressions × 100

What it tells you: What percentage of users who see your ad interact with it. Indicator of ad creative effectiveness.

Benchmarks: 1-3% for display ads, 5-15% for search ads (Apple Search Ads).

CVR (Conversion Rate)

CVR = Installs ÷ Taps × 100

What it tells you: What percentage of users who tap your ad actually install. Indicator of store listing effectiveness and ad-to-listing alignment.

Benchmarks: 20-40% from search ads, 5-15% from display/social ads.

IPM (Installs Per Mille)

IPM = (Installs ÷ Impressions) × 1,000

What it tells you: How many installs you get per 1,000 ad impressions. Combines ad creative performance (TTR) with store listing performance (CVR) into a single efficiency metric.

Why IPM matters: Ad networks optimize toward IPM. Higher IPM = better ad performance = lower CPI from the algorithm.

Revenue Metrics

ARPU (Average Revenue Per User)

ARPU = Total Revenue ÷ Total Active Users

What it tells you: Revenue generated per user across your entire user base (including free users who generate $0).

Monthly ARPU benchmarks:

• Ad-supported apps: $0.50-$4.00
• Subscription apps (blended): $0.20-$2.00
• IAP-driven games: $0.10-$1.00 (casual), $0.50-$5.00 (mid-core)

ARPPU (Average Revenue Per Paying User)

ARPPU = Total Revenue ÷ Number of Paying Users

What it tells you: How much each paying user spends on average. Useful for understanding monetization depth.

ARPDAU (Average Revenue Per Daily Active User)

ARPDAU = Daily Revenue ÷ DAU

What it tells you: Revenue per active user per day. The most granular revenue metric, especially useful for ad-supported apps.

Benchmarks: $0.02-$0.15 for ad-supported apps, higher for games with IAPs.

Lifetime Value Metrics

LTV (Lifetime Value)

LTV = ARPU × Average Lifetime (in months)

Or more precisely:

LTV = Σ (Monthly Revenue per Cohort User × Retention Rate at Month N) for N = 0 to ∞

What it tells you: The total revenue a user will generate over their entire relationship with your app. THE most important metric for sustainable growth.

Why LTV matters for advertising: Your LTV determines your maximum sustainable CPI:

Max CPI = LTV × Target Profit Margin

If LTV = $30 and you want 50% margins: Max CPI = $15.

LTV by acquisition source: Not all users have the same LTV. Organic users typically have 1.5-3x higher LTV than paid users. Apple Search Ads users often have higher LTV than social media acquired users (higher intent).

LTV:CAC Ratio

LTV:CAC = Lifetime Value ÷ Customer Acquisition Cost

Benchmarks:

• 1:1 → Breakeven (unsustainable long-term)
• 2:1 → Minimum viable (covers operational costs)
• 3:1 → Healthy (standard target for most apps)
• 5:1+ → Excellent (or potentially under-investing in growth)

Efficiency Metrics

ROAS (Return on Ad Spend)

ROAS = Revenue from Ad-Acquired Users ÷ Ad Spend

Time-windowed ROAS:

• D0 ROAS: Revenue on install day ÷ spend (typically 5-20% for subscription apps)
• D7 ROAS: Cumulative 7-day revenue ÷ spend (20-50%)
• D30 ROAS: Cumulative 30-day revenue ÷ spend (50-120%)
• D90 ROAS: Cumulative 90-day revenue ÷ spend (80-200%)

Target D30 ROAS by model:

• Subscription apps: 80-120% (break even by day 30, profit from renewals)
• IAP games: 60-100% (longer payback period due to whale dynamics)
• Ad-supported apps: 100%+ by D30 (must be profitable quickly due to lower LTV)

eCPI (Effective Cost Per Install)

eCPI = Total Marketing Spend ÷ Total Installs (organic + paid)

What it tells you: Your blended acquisition cost including the organic multiplier effect. If paid campaigns drive organic installs too (through improved rankings), your eCPI is lower than your paid CPI.

The organic multiplier: For every paid install, well-optimized apps generate 0.5-2.0 additional organic installs. This dramatically improves blended eCPI:

• Paid CPI: $3.00
• Organic multiplier: 1.5x
• Blended eCPI: $3.00 ÷ 2.5 = $1.20

Building a Metrics Dashboard

Essential Daily Metrics

Weekly Metrics

Monthly Metrics

Common Metrics Mistakes

Optimizing for CPI instead of LTV. The cheapest installs are rarely the most valuable. A $1 CPI from an incentivized network delivers users worth $0.50 in LTV. A $5 CPI from Apple Search Ads might deliver users worth $30 in LTV.

Measuring ROAS too early. D7 ROAS for a subscription app is almost always below 100%. That doesn't mean the campaign is unprofitable — it means subscription revenue takes time to accumulate. Wait for D30 or D90 ROAS before making decisions.

Ignoring the organic multiplier. Paid campaigns that improve your app's ranking generate organic installs you're not paying for. Your true acquisition efficiency is better than your paid CPI suggests.

Comparing CPI across platforms without context. Android CPI is typically 40-60% lower than iOS CPI, but iOS ARPU is typically 2-3x higher. Lower CPI ≠ better deal.

Not segmenting by geography. US users might have $40 LTV while Indian users have $3 LTV. Blending these together masks the true profitability of each market.

Treating all installs equally. An install from a brand keyword search (high intent) is fundamentally different from an install from a broad social media campaign (low intent). Segment your metrics by campaign type.

Not accounting for platform commission. Revenue reported by your analytics tool is gross revenue. Net revenue after Apple/Google's 30% (or 15%) commission is what matters for ROAS and ACOS calculations.

Using Metrics to Make Decisions

When to Scale a Campaign

Scale when:

• D30 ROAS > 100% (or trending confidently toward it based on D7 data)
• LTV:CAC > 3:1
• Retention of paid users is within 80% of organic user retention
• You have at least 2 weeks of stable data

When to Pause a Campaign

Pause when:

• D7 ROAS < 20% with no improvement trend
• Retention of acquired users is <50% of organic benchmark
• CPA for target events is >3x your target
• Creative fatigue has pushed CPI up 50%+ from baseline

When to Optimize (Not Scale or Pause)

Optimize when:

• Metrics are close to targets but not quite there
• Performance varies significantly by creative, audience, or placement
• Some cohorts within the campaign perform well while others don't

Optimization levers:

1. Creative refresh (new ad variants)
2. Audience refinement (exclude underperforming segments)
3. Bid adjustment (increase for high-performing keywords/placements)
4. Dayparting (concentrate spend during high-conversion hours)
5. Geographic focus (shift budget to highest-ROAS markets)

Conclusion

Advertising metrics are your compass for navigating paid user acquisition. ACOS and ROAS tell you whether campaigns are profitable. CPI and CPA tell you what you're paying for users and actions. LTV tells you what those users are worth. And the LTV:CAC ratio tells you whether your growth is sustainable.

Master these metrics and you can make confident, data-driven decisions about where to spend, when to scale, and when to cut. Ignore them and you're gambling — spending money with no reliable way to know if it's working.

Start with the basics: track CPI and D7/D30 ROAS by channel. Graduate to LTV-based optimization as your data matures. Build a dashboard that surfaces the right metrics at the right cadence. And always remember: the goal isn't to minimize CPI — it's to maximize the gap between what you pay for a user and what that user is worth to your business.