Breaking
UTM Parameters Turn $338B in Social Ad Spend Into ProofP&N Bank flips the customer-owned banking scriptHow to Get Products Found in Google and ChatGPT AI ShoppingWhy PageSpeed Scores Might Jump After Lighthouse 13.6TikTok Search Ads Adds Impression Share and Keyword PacksUTM Parameters Turn $338B in Social Ad Spend Into ProofP&N Bank flips the customer-owned banking scriptHow to Get Products Found in Google and ChatGPT AI ShoppingWhy PageSpeed Scores Might Jump After Lighthouse 13.6TikTok Search Ads Adds Impression Share and Keyword Packs

Mobile ad IDs: the fragile backbone of app marketing

What a MAID is, how IDFA and GAID power in-app targeting and measurement, and why regulators and platform changes are reshaping mobile ad tracking.

What the mobile ad ID means for your campaigns

A mobile advertising ID may be the least glamorous piece of ad tech, but it is the reason in-app targeting, frequency capping and install attribution work at all. Apps do not share a browser cookie jar. Each app runs in a sandbox, so a network cannot natively connect an ad seen in a puzzle game to a later install of a retailer’s app. The MAID closes that gap.

One ID, two platforms

Apple calls its version the Identifier for Advertisers (IDFA); Google’s is the advertising ID, often written GAID or AAID. Both are 32-character UUIDs that every app on a device can read through an OS call. In programmatic auctions, supply-side platforms copy the value into the OpenRTB device object’s ifa field, alongside a limit-ad-tracking flag.

The value is resettable, and since iOS 14.5 an app must ask permission through App Tracking Transparency before reading it. If a user refuses, deletes the ID or opts out of personalisation, the OS returns a string of zeros. The ID is pseudonymous, not anonymous, and European law treats it as personal data.

From hardware serials to consent prompts

Before MAIDs, mobile advertising leaned on hardware identifiers such as UDID, IMEI and MAC address. Apple shipped the IDFA in 2012 and dropped UDID access for new App Store submissions in 2013. Google introduced its advertising ID in 2013, replacing the Android ID for ads. The real shock came in April 2021, when ATT made IDFA access opt-in. Research cited by economist Reinhold Kesler put the share of trackable US users at 72.6% before ATT and 17.9% after. Meta’s CFO told investors in early 2022 that the iOS changes would cost around $10 billion that year.

Google zeroed the GAID for opted-out users rather than forcing a prompt, then in October 2025 retired the Android Privacy Sandbox APIs that were meant to replace it. The advertising ID remains in service.

Why regulators keep hitting it

The same persistence that makes a MAID useful also makes it sensitive. The FTC’s Kochava complaint described a dataset pairing MAIDs with GPS coordinates for 61 million unique devices and said it could trace a device from a reproductive health clinic to a family home. In December 2024, the agency said Mobilewalla had collected more than 500 million advertising identifiers with precise location. Apple’s ATT design is also contested: France fined Apple 150 million euros in March 2025, Italy followed with 98.6 million euros, and Germany accepted a prompt redesign in August 2026.

Yet demand is not fading. Adjust measured average ATT opt-in at 38% in Q1 2026, up from 35% a year earlier, and OpenAI began accepting the Google Advertising ID for ChatGPT custom audiences in August 2026.

What to do about it

Treat the MAID as a regulated identity signal, not free tracking infrastructure. That has three practical consequences:

  • Run separate attribution logic for iOS and Android, because iOS now leans on SKAdNetwork-style install confirmation while Android keeps deterministic matching.
  • Audit SDK and bidstream data, especially anywhere MAIDs meet location signals.
  • Design consent prompts for transparency, not maximum opt-in; regulators are watching the difference.

For performance teams, the MAID is still the join key for in-app measurement. The new job is managing consent, legal exposure and measurement across two very different mobile ecosystems.

Source: PPC Land