In this article
1. The Two Worlds of Mobile Advertising
Mobile advertising accounts for over 70% of all digital ad spend globally, yet the mobile ecosystem operates across two fundamentally different environments: in-app and mobile web. Understanding the technical, behavioral, and economic differences between these two channels is essential for any advertiser or publisher building a mobile strategy.
In-app advertising refers to ads served within native applications downloaded from app stores (Google Play, Apple App Store). These ads are delivered through advertising SDKs integrated into the app's code, giving advertisers access to rich device-level data and a controlled rendering environment. Mobile web advertising, by contrast, refers to ads served within a mobile browser (Chrome, Safari, Samsung Internet) on web pages rendered by the browser engine — the same infrastructure that powers desktop web advertising, adapted for mobile screens.
The split between these two channels is significant. Globally, approximately 80-85% of mobile time is spent in apps versus 15-20% in mobile browsers. However, this distribution is heavily concentrated: a small number of "super apps" (Facebook, Instagram, YouTube, TikTok, WhatsApp) capture the majority of in-app time, while mobile web browsing remains more distributed across millions of websites.
2. In-App Advertising: How It Works
In-app advertising relies on an SDK (Software Development Kit) embedded in the application during development. When the app loads an ad slot, the SDK communicates with an ad server or exchange, receives a creative, and renders it within the app's native interface.
SDK Integration
The advertising SDK is the technical backbone of in-app monetization. Popular mobile ad SDKs include Google AdMob, Meta Audience Network, AppLovin, ironSource (Unity), InMobi, and Pangle (ByteDance). The SDK handles:
- Ad request generation: Collecting device signals (GAID/IDFA, device model, OS version, connection type, screen size) and packaging them into an OpenRTB bid request
- Auction participation: Communicating with ad exchanges to solicit bids from multiple demand sources
- Creative rendering: Displaying the winning creative within the app, handling format-specific rendering (banner, interstitial, rewarded video, native)
- Event tracking: Reporting impressions, clicks, video starts, completions, and other engagement events back to the exchange and verification providers
- Mediation: Managing multiple ad networks within a single integration, selecting the highest-paying network for each impression
App Store Ecosystem
In-app advertising exists within the app store ecosystem, which creates both advantages and constraints. Apps must pass app store review, adhere to platform policies (Apple's App Tracking Transparency, Google's Play Store policies), and share revenue with the platform (typically 15-30% on in-app purchases, though ad revenue is not directly subject to platform fees). The app store ecosystem provides a trusted distribution channel but also gives Apple and Google significant control over how advertising operates on their platforms.
Common In-App Ad Formats
- Banner ads: Static or animated banners displayed at the top or bottom of the app interface
- Interstitials: Full-screen ads that appear at natural transition points (between game levels, after content loads)
- Rewarded video: Full-screen video ads that users opt into watching in exchange for in-app rewards (extra lives, premium content, virtual currency)
- Native ads: Ads designed to match the visual style and content format of the surrounding app experience
- Playable ads: Interactive mini-game or demo experiences that let users try an app before installing
- Offer walls: Curated lists of incentivized actions (install an app, complete a survey) that reward users with in-app currency
3. Mobile Web Advertising: How It Works
Mobile web advertising uses the same infrastructure as desktop web advertising — JavaScript ad tags, header bidding wrappers, and ad exchanges — but is adapted for mobile browser environments. When a user visits a website on their mobile device, the publisher's ad scripts execute in the browser, triggering an auction that fills available ad slots.
Technical Architecture
Mobile web ads are served through JavaScript-based ad tags embedded in the publisher's HTML. The process typically involves:
- Prebid.js wrapper: A client-side header bidding implementation that solicits bids from multiple demand partners simultaneously before calling the primary ad server
- Ad server call: The primary ad server (typically Google Ad Manager) receives the prebid results alongside direct-sold and network demand and selects the winner
- Creative rendering: The winning creative is rendered in an iframe within the publisher's page
- Tracking: JavaScript-based impression and click tracking, often supplemented with verification pixels from IAS, DoubleVerify, or similar
The Cookie Limitation
Mobile web advertising has historically relied on third-party cookies for user identification and targeting. With Safari's Intelligent Tracking Prevention (ITP) already blocking third-party cookies and Chrome moving toward cookie deprecation, mobile web targeting faces significant challenges. Publishers are responding with first-party data strategies (login walls, newsletter signups) and contextual targeting approaches that do not depend on persistent user identifiers.
4. Head-to-Head Comparison
| Dimension | In-App | Mobile Web |
|---|---|---|
| User Identifier | GAID (Android) / IDFA (iOS) — persistent, device-level | Cookies — ephemeral, browser-level, increasingly restricted |
| Targeting Depth | Deep: device data, app usage, in-app behavior, SDK-derived signals | Moderate: contextual, first-party data, limited cross-site tracking |
| Creative Control | High: SDK renders in native container, supports rich formats | Moderate: rendered in browser iframe, limited by web standards |
| Viewability | 75-90% (native rendering, full-screen control) | 50-70% (iframe rendering, scroll-dependent) |
| IVT Rate | 2-5% (device-level verification) | 8-15% (browser-level, more susceptible to bots) |
| CTR (Banner) | 0.3-1.2% (higher but includes accidental clicks) | 0.1-0.4% (lower but more intentional) |
| CPM Range (US) | $3 - $20+ (varies by format and app quality) | $1 - $12 (varies by publisher quality) |
| Inventory Scale | Concentrated in top apps; long tail of smaller apps | Distributed across millions of mobile websites |
| Brand Safety | App store review provides baseline; bundle ID targeting | Contextual analysis, blocklists, inclusion lists |
5. Targeting Capabilities
The targeting capabilities of in-app and mobile web differ substantially due to the technical differences between native apps and browser environments.
In-App Targeting Advantages
In-app advertising benefits from access to the device's advertising identifier (GAID on Android, IDFA on iOS), which enables deterministic cross-app user identification. This allows advertisers to build behavioral profiles based on the user's complete app usage patterns — which apps they use, how frequently, at what times, and in what sequence. SDK providers like AppsFlyer, Adjust, and Branch aggregate this data to power sophisticated audience segmentation and retargeting.
Additionally, in-app environments provide rich contextual signals: the app's category (game, news, fitness), the user's in-app behavior (level reached, items purchased, content consumed), and device-level data (battery level, available storage, network type) that is unavailable in browser environments.
Mobile Web Targeting Advantages
Mobile web advertising excels at contextual targeting. Because the browser has direct access to the page URL, page content, and referrer data, contextual signals are rich and immediately available. The open web also provides access to the full breadth of content — from news articles to blog posts to e-commerce product pages — creating contextual targeting opportunities that don't exist in the more siloed app ecosystem.
First-party data strategies are also more mature on the mobile web. Publishers with authenticated user bases (news sites, e-commerce platforms) can target known users directly, leveraging CRM data, purchase history, and subscription status for precise targeting that does not depend on third-party identifiers.
6. Viewability & Engagement
Viewability — the percentage of served impressions that were actually visible to the user — is one of the most significant differentiators between in-app and mobile web traffic.
In-App Viewability
In-app environments typically achieve higher viewability rates (75-90%) because the SDK has native-level control over ad rendering. The SDK can ensure ads are rendered only when they are fully on-screen, pause video ads when the app is backgrounded, and prevent creative refresh before the minimum viewability threshold is met. Rewarded video formats, where users explicitly opt in to watch an ad, achieve near-100% viewability and completion rates above 90%.
Mobile Web Viewability
Mobile web viewability is lower (50-70%) because ads are rendered in iframes within the browser, and the publisher's page layout, scroll behavior, and lazy loading implementation all affect whether the ad is actually visible. Ads placed below the fold, in sidebars on narrow mobile screens, or in auto-refreshing slots that cycle before the user scrolls to them generate served impressions that never meet the MRC viewability standard (50% of pixels in view for at least 1 continuous second for display, 2 seconds for video).
7. Fraud Rates & Quality
Ad fraud manifests differently in in-app and mobile web environments, with distinct risk profiles for each channel.
In-App Fraud Patterns
- Click injection: Malicious apps detect when a user organically installs another app and inject a fraudulent click attribution to claim the install credit
- Click spamming: Apps generate a high volume of fake clicks in the background, statistically claiming credit for organic installs
- SDK spoofing: Fraudsters simulate SDK events (impressions, clicks, installs) without any real user interaction, generating fake traffic directly from bot farms
- Device farms: Physical racks of real devices running apps to generate genuine-looking ad impressions and engagement
Mobile Web Fraud Patterns
- Bot traffic: Automated scripts that crawl websites and trigger ad impressions without human involvement
- Domain spoofing: Fraudsters misrepresent low-quality or non-existent inventory as premium publisher sites in bid requests
- Ad stacking: Multiple ads rendered on top of each other in a single ad slot
- Pixel stuffing: Ads rendered at 1x1 pixel size, generating served impressions that are never visible
- Traffic redirects: Users are involuntarily redirected through ad networks, generating impressions on pages they never intended to visit
Overall, in-app traffic tends to have lower invalid traffic (IVT) rates (2-5%) compared to mobile web (8-15%), primarily because device-level identifiers and SDK-based verification make traditional bot traffic more difficult to generate. However, sophisticated in-app fraud (SDK spoofing, click injection) can be more damaging per incident because it bypasses many standard detection mechanisms.
8. Monetization Strategies
In-App Monetization
App developers typically use a waterfall + mediation approach, where multiple ad networks compete for each impression through a mediation layer. The mediation platform (AppLovin MAX, ironSource LevelPlay, Google AdMob Mediation) solicits bids from all connected networks and selects the highest bidder for each impression. This has largely replaced the legacy sequential waterfall, where networks were called in priority order.
The most successful in-app monetization strategies combine multiple ad formats: rewarded video for high eCPM, interstitials for fill rate, banners for consistent baseline revenue, and native ads for user experience preservation. Games, in particular, have mastered the rewarded video format, where ad engagement is voluntary and users receive tangible in-app value in exchange for their attention.
Mobile Web Monetization
Mobile web publishers monetize through header bidding (Prebid.js or server-side alternatives like Prebid Server), which allows multiple demand sources to compete simultaneously for each impression. The publisher's ad server (typically Google Ad Manager) serves as the final decision point, comparing programmatic bids against direct-sold campaigns and guaranteed deals.
Key mobile web monetization strategies include responsive ad units that adapt to screen size, lazy loading to improve viewability, sticky footer banners for consistent revenue, and interstitial ads between page navigations (common in news and content sites).
9. When to Use Each Channel
Choose In-App When:
- App install campaigns: In-app is the primary channel for driving app installs, with direct store integration and deep linking
- Gaming audiences: Mobile gaming inventory is almost exclusively in-app, offering unique access to engaged gaming audiences
- Rewarded engagement: When campaign goals include video completion, interactive engagement, or opt-in experiences
- Device-level targeting: When precise device targeting (specific models, OS versions, network conditions) is required
- Emerging markets: In many emerging markets, app usage dominates and mobile web penetration is lower
Choose Mobile Web When:
- Content-driven campaigns: When targeting users consuming specific editorial content (news, sports, finance)
- Contextual targeting: When page-level context is the primary targeting signal
- Direct response to web conversions: When the conversion path is a web form, purchase, or signup — keeping the user in-browser reduces friction
- Brand safety priority: Premium web publishers offer well-established brand safety frameworks and content classification
- Cross-device consistency: When campaigns need to reach users across desktop and mobile in a unified environment
In practice, most sophisticated mobile advertising strategies use both channels in combination, with budget allocation determined by campaign objectives, target audience, and performance data. The in-app channel typically drives higher engagement metrics and lower fraud rates, while the mobile web offers broader contextual targeting and a more transparent supply chain. The most effective approach is to test both channels, measure downstream conversion quality, and allocate budget based on incremental ROI rather than surface-level engagement metrics.