The Web Just Lost More Ground: Why I’m Telling Advertisers to Plan In-App First

This post was written by Ravit Ross, Chief Revenue Officer at Start.io.

Like most people in this industry, I spend a lot of my time talking with advertisers. Lately, one question keeps coming up: Where can you still reach real audiences with signals you can rely on? In October, Apple offered part of the answer.

Apple’s WebKit team explained that Safari and WebKit now block network access to the infrastructure behind cross-site identity resolution — the services that turn an email address typed into a login or newsletter form into a persistent tracking ID. This goes well beyond cookies. It targets the workarounds the industry built after cookies.

I lead revenue at an in-app advertising company, so I see this through a particular lens. But the numbers behind the argument aren’t ours. They come from Apple’s own announcement and from independent industry measurement. Judge the case on those.

What Apple actually did — and why it matters more than cookies

In its iOS, iPadOS, macOS, and visionOS 27 releases, Safari blocks these identity resolution services at the domain level. As WebKit put it, providers can’t restructure a network request to get around it. Email-based identity — the web’s leading cookie replacement — simply doesn’t function in Safari anymore.

Here is the scale. Safari carries nearly half of U.S. mobile and tablet browsing — 44.3% as of October 2026, according to independent market-share data. So for a U.S. advertiser, almost one in every two mobile web impressions now sits in an environment where email-based identity graphs cannot operate. That is not a future risk to plan around. It shipped.

And the direction of travel is one-way. Apple blocked third-party cookies in 2020, Link Tracking Protection followed, and now identity resolution. Every time the web rebuilds cross-site identity, a browser vendor takes it apart. Building an audience strategy on that foundation means rebuilding it every 18 months.

Follow the time, not the habit

While addressability erodes on the mobile web, consumers were never really there to begin with. Around 90% of mobile time is spent in apps, against roughly 10% in mobile browsers — a ratio independent measurement has shown for years, and one that keeps widening.

Independent app-market research puts the global total at 5.3 trillion hours spent in apps in 2025, with the average user in apps 3.6 hours a day. The mobile web gets minutes; apps get hours.

Yet many media plans still allocate as if the mobile web were where mobile audiences live, because that is where the identity plumbing — and the habit — was. Apple just removed the plumbing. The habit is the only thing left holding those allocations in place.

Why in-app signals hold up where web signals don’t

Let me address the obvious objection head-on (because sophisticated buyers will raise it): Didn’t Apple do the same thing to apps with App Tracking Transparency in 2021? Yes — and that is exactly the point. In-app advertising took its privacy medicine four years ago. The ecosystem that survived rebuilt around signals that don’t depend on tracking individuals across unrelated properties:

• The context of the impression itself. Apps are purpose-built — finance, travel, fitness, games. A session in a flight-tracking app the week before a trip says more about intent than any reconstructed browsing graph, and it says it without following anyone anywhere.
• Direct publisher integrations. When the SDK sits in the app, the signals — app category, placement, device model, language, session behavior — come from the environment where the impression occurs, not from a third-party graph a browser can switch off.
• Device-level signals that work without an advertising ID. Contextual and device-based approaches establish relevance even when an IDFA or GAID is unavailable — which, post-ATT, is the design assumption, not the edge case.

The web is only now losing its post-cookie workarounds, and each loss is a surprise that breaks campaigns mid-flight. In-app already made that transition. Its foundation is the one Apple’s announcements keep validating rather than dismantling.

What I tell advertisers to do now

Not “move everything in-app.” Web and search keep a role, particularly at the bottom of the funnel. What I push buyers to do is three things before the next planning cycle:

1. Audit your identity dependence. Line by line, how much of your planned mobile reach assumes browser-based identity — cookies, email-based IDs, fingerprinting? Price in that roughly half of U.S. mobile browsing already blocks the second category.
2. Match budget to attention. If consumers spend about 90% of their mobile time in apps and your mobile budget doesn’t remotely reflect that, the gap is your opportunity — and your competitor’s.
3. Interrogate your partners’ supply path. Ask any provider: How directly do you access your inventory, and where do your audience signals originate? A partner reselling signals built on browser identity inherits every browser restriction. A partner with direct publisher integrations doesn’t.

The advertisers who moved after ATT in 2021 had two years of learning before their competitors reacted. Apple just fired the same starting gun for the web.

Where Start.io fits

This is the strategy we built Start.io around: direct publisher integrations supporting audience-based buying across in-app inventory. Our device-based audience capabilities use app and device context to establish relevance — including when mobile advertising identifiers are unavailable — so campaigns don’t depend on identity infrastructure that a browser release can switch off.

Apple didn’t change my view of where mobile advertising is heading. It published the roadmap.

Talk to our team about building a mobile-first audience strategy on direct access to in-app signals.