Google Partner Utah SEO, AEO & digital marketing agency
General Marketing

Will iOS 17 Update Kill Your Ads: Here's What You Need To Know

Apple's privacy changes removed most of the signal advertisers were quietly relying on. Here is what actually broke, what server side tracking recovers, and how to measure campaigns now.

4:10 watchMay 12, 2026General Marketing
5POINTS TO KEEP A LOOK AT
1Video
AllSkill Level
4:10MINUTES ONLY!
What you'll learn

3 things you can act on after watching

  • Safari
  • Apple Mail
  • iMessage
Overview

What Apple actually broke

When Apple introduced App Tracking Transparency and tightened Intelligent Tracking Prevention, a lot of advertisers watched their reported conversions fall off a cliff and concluded their campaigns had stopped working.

Mostly they had not. What stopped working was the ability to observe them. That distinction matters enormously, because the two problems have opposite responses. One calls for changing the campaign, the other for fixing the measurement.

What actually changed

  • App tracking requires permission. Most people decline, so the identifier that linked an ad impression in one app to an action in another largely disappeared.
  • Third party cookies got restricted hard in Safari, and first party cookies set by script had their lifetime cut dramatically.
  • Attribution windows shortened. Conversions that happen days after the click became much harder to connect back to it.
  • View-through attribution largely went away. People who saw but did not click, then converted later, are now mostly invisible.

Notice what is not on that list. Ad delivery, targeting breadth and the existence of demand were not removed. Measurement of outcomes was.

Who felt it most

The damage was not evenly spread. Businesses with long consideration windows felt it most, because theirs are exactly the conversions that happen outside a short attribution window. Businesses relying on view-through credit saw the biggest apparent drops. Ecommerce with same-session purchases felt it least.

If your reported performance collapsed but revenue did not, that is the signature of a measurement problem rather than a performance one. Comparing platform-reported conversions against actual revenue over the same period is the quickest way to tell which you have.

What recovers the signal

  1. Server side tracking. Meta's Conversions API, Google's enhanced conversions, and server side tagging generally. Events sent from your server cannot be blocked in the browser, so a large share of the lost matching comes back.
  2. First party data collection. Email addresses and phone numbers you have a lawful basis to use are what make matching possible. This is now the core asset rather than a nice extra.
  3. Consent handled properly. Consent mode and equivalent mechanisms preserve more signal than a blunt block does, and they keep you on the right side of the rules.
  4. Offline conversion uploads. If the deal closes in your CRM weeks later, upload that outcome so the platform can learn from the conversions that actually matter.

How to measure now

Platform reported numbers are no longer a complete count and never will be again. The practical approach is to stop expecting reconciliation and build a picture from several angles.

  • Track total revenue or booked work against total spend over a period, rather than trying to attribute each sale.
  • Watch the direction of platform numbers rather than their absolute value. The trend is still meaningful even when the level is understated.
  • Ask people where they heard about you, at the point of enquiry. Crude, unfashionable, and often the most reliable source you have.
  • Use holdout tests where volume permits. Turning a channel off for a period remains the only genuinely clean measure of its contribution.

The part worth accepting

Some fraction of your marketing will never be attributable again. Businesses that handle this well set a tolerance for unattributed revenue, measure what they can at the aggregate level, and stop making decisions on a per-click report that no longer describes reality.

Key takeaways

  • Nothing stopped working. The measurement of it did, which looks identical on a dashboard and is a completely different problem.
  • The loss is concentrated in view-through and cross-site attribution, not in whether ads reach people.
  • Server side tracking recovers a meaningful share of the conversions that stopped being matched.
  • Modelled conversions are estimates. Treat them as directionally useful and never as a reconciled number.
  • First party data is the asset that survives all of this, and most businesses still are not collecting it deliberately.

Step-by-Step: How to Prepare for LTP

  1. Step 1: Standardize UTM Parameters

    UTM parameters are still essential and fully usable.

    Example structure:

    • utm_source
    • utm_medium
    • utm_campaign
    • utm_content
  2. Step 2: Adapt to New Tracking Parameters (GCLID Alternatives)

    Apple’s update encourages newer tracking methods such as:

    • gBraid
    • wBraid

    These are designed to improve tracking between:

    • App → Web
    • Web → App
    • App → App

    They help fill gaps when traditional URL tracking is restricted.

  3. Step 3: Use Custom Landing Pages

    Instead of relying only on shared tracking links, create:

    • Dedicated landing pages per campaign
    • Clean URLs without unnecessary parameters
    • Platform-specific funnels (Google Ads vs Facebook Ads)

    This improves:

    • Attribution accuracy
    • Conversion tracking reliability
    • User experience
  4. Step 4: Improve Conversion Tracking Setup

    Make sure your conversion tracking is robust and not dependent on fragile URL data.

    Focus on:

    • Server-side tracking where possible
    • First-party data collection
    • CRM-based attribution
    • Form tracking events
  5. Step 5: Strengthen First-Party Data Strategy

    Since third-party tracking is getting weaker, prioritize:

    • Email collection
    • CRM integration
    • Lead forms
    • Phone tracking
    • Direct customer data capture

    This ensures you still retain accurate performance insights even when tracking is limited.

In depth

Google Trend FAQs

FAQs

Questions people ask about this

What did Apple's privacy changes actually break?

Measurement rather than delivery. App Tracking Transparency removed the identifier linking an ad in one app to an action in another, Safari's protections cut cookie lifetimes sharply, attribution windows shortened and view-through credit largely disappeared. Ads still reach people and demand still exists, but connecting the two became much harder.

Did iOS changes actually reduce ad performance?

For most advertisers the reported performance fell further than the real performance. The quickest diagnostic is to compare platform reported conversions against actual revenue over the same period. If revenue held while reported conversions collapsed, you have a measurement problem rather than a campaign problem.

How do I recover lost conversion tracking?

Server side tracking is the main fix: Meta's Conversions API, Google's enhanced conversions, and server side tagging generally. Because events are sent from your server rather than the visitor's browser, blocking in the browser does not remove them. Offline conversion uploads recover outcomes that happen later in your CRM.

Are modelled conversions real?

They are statistical estimates of conversions the platform can no longer observe directly. They are useful for direction and for feeding optimisation, and they should never be treated as a reconciled count or compared like for like against your accounting figures.

What is the best way to measure campaigns now?

Stop expecting per-click reconciliation. Track total revenue against total spend over a period, watch the direction of platform numbers rather than their absolute level, ask people where they heard about you at the point of enquiry, and use holdout tests where volume allows, since turning a channel off remains the cleanest measure of its contribution.

Why does first party data matter more now?

Because matching is what was lost, and first party data is what makes matching possible again. Email addresses and phone numbers you have a lawful basis to use are what let server side tracking reconnect a conversion to the click that caused it. It has gone from a nice extra to the core measurement asset.

Next step

Need Help Building a Future-Proof Paid Ads Tracking System?

Our in-house team works directly with US-based marketers. Tell us what you are trying to grow and we will tell you what we would do first.

ENES