iOS Privacy Update: Conversion Tracking Broken—Quick Fixes

Why Apple's iOS Privacy Rules Broke Your Conversions

In September 2021, Apple shipped App Tracking Transparency (ATT) with iOS 15. It forced apps—including Facebook's in-app browser—to ask users: "Allow [App] to track your activity across other companies' apps and websites?" Around 70–85% of users hit "Ask App Not to Track."

When a user opts out, the Facebook pixel stops firing. Facebook's pixel is a JavaScript snippet that lives on your website and sends data back to Facebook's servers: "This visitor added a product to cart," or "This person called the business." Without the pixel firing, Facebook sees zero signal that the ad worked.

The result: conversion tracking loss of 40–60% for iOS traffic. Since iOS accounts for roughly 25–35% of all web traffic in the US (higher in affluent areas), you're losing data on 10–20% of your total conversions. Your cost-per-lead appears 30–50% higher than reality, and Facebook's algorithm can't optimize toward the winning audience.

A plumber in Portland, OR running $2,000/month in Facebook ads saw conversions drop from 18 to 11 in June 2021. His actual cost-per-lead was $111, but the platform showed $176 because it couldn't track 7 calls from iOS users who had opted out of tracking.

The Difference: Pixel Tracking vs. Conversion API

Pixel Tracking (what you've been using): JavaScript runs on your website. When someone clicks your ad, Facebook sets a cookie or ID. When they convert (buy, call, fill a form), the pixel fires and tells Facebook. The problem: on iOS 15+, the pixel fires after the user has decided to opt out of tracking. Facebook never gets the message.

Conversion API (the fix): Instead of relying on pixels on the user's device, you send conversion data directly from your server to Facebook's servers. Your backend code logs the event—say, "Customer called at 2:14 PM on Jan 15"—and posts it to Facebook's API endpoint. Apple's privacy rules don't block server-to-server communication.

The difference is critical: with Conversion API, you control the data. You know when the lead came in, what they said, whether they converted. You send that to Facebook after it happens, not hoping a pixel fires on the user's device.

Real Numbers: What You're Losing

A HVAC contractor in Atlanta running lead-generation ads lost visibility into this breakdown:

• Before Conversion API: 40 conversions reported / month, $1,250 ad spend, $31.25 CPC.
• After Conversion API: 64 conversions reported / month, same $1,250 ad spend, $19.50 CPC.

The actual leads didn't change. The tracking did. Once Conversion API was live, Facebook could see 24 additional conversions it had been blind to. The platform immediately re-optimized: the algorithm shifted budget toward the audiences and placements driving the best ROI.

In this case, the contractor's actual ROAS (return on ad spend) was 2.8:1, but the platform was showing 1.6:1. He was cutting budgets thinking they weren't working, when in fact he had tracking blindness.

Step 1: Understand Your Traffic Loss

Not all traffic is equal. Desktop and web-only browsing have minimal tracking loss (5–10%). Mobile Safari and in-app browsers have high loss (40–60%). Here's the breakdown:

• Desktop: 85–95% of conversions tracked (pixel works fine).
• Mobile Safari (iOS in-app browser): 30–50% of conversions tracked (pixel blocked on many users).
• Android: 90–95% of conversions tracked (no ATT equivalent).

If you run Facebook ads for electricians in a dense metro area, your audience is 60–70% mobile. You're losing data on 25–40% of your actual leads.

To see your own loss: open Facebook Ads Manager, go to "Conversions" in your campaign, toggle between "Device OS." You'll see iOS traffic has 20–40% fewer conversions than Android or desktop. The gap is your tracking blindness.

Step 2: Set Up Conversion API (with a Developer)

Conversion API requires backend code. If you have a developer, the setup takes 4–8 hours:

1. Get your Conversion API token. In Business Manager → Settings → Data Sources → Web → Conversions API, generate an access token. Store it securely (don't commit it to GitHub).
2. Identify events to track. Start with the 3 highest-value events: Purchase (if e-commerce), Lead (form submission), and Contact (phone call). Add these first; don't boil the ocean.
3. Map your backend to send events. When a user submits a form, your server posts JSON to Facebook's endpoint with the event data: phone number (hashed), timestamp, currency, value, etc. Your developer will add this in 2–3 lines of code per event.
4. Hash Personally Identifiable Information (PII). Phone numbers, emails, and names must be hashed (SHA-256) before sending to Facebook. This is not optional and is why developer help is needed.
5. Test the integration. Use Facebook's Conversion API Test Event Tool (in Ads Manager, Test Events button) to fire a test conversion and confirm it arrives at Facebook within seconds.
6. Monitor for 7 days. Watch Ads Manager's conversion reporting. You should see a 15–30% increase in conversions within 48 hours if tracking was your problem.

Cost: $500–$2,500 one-time for developer setup. If your ad spend is over $1,500/month, this pays for itself in the first week.

Step 3: Conversion API Without a Developer

If you don't have a developer, use a third-party integration:

Shopify / WooCommerce: Install the official Meta Conversions API plugin (free). It auto-sends all purchase events to Facebook. Setup takes 15 minutes, no code required.

Zapier / Make (formerly Integromat): Connect your form, CRM, or phone system to Zapier. When a new lead comes in, Zapier triggers a webhook that posts the conversion to Facebook's Conversion API. Cost: $20–$99/month depending on volume.

HubSpot / Salesforce Integration: If your CRM is HubSpot or Salesforce, add a Facebook app that auto-syncs conversions. Cost is usually bundled in your CRM subscription.

Trade-off: third-party tools are slower and less reliable than native integrations. A Zapier delay of 30 seconds to 5 minutes means Facebook gets the event later, which can impact its ability to attribute it back to the correct user session. But 30 seconds late is vastly better than never.

Step 4: Keep Your Pixel Live (Don't Remove It)

A common mistake: teams assume Conversion API replaces the pixel and delete it. Don't do this.

The pixel still works for 35–40% of your traffic (desktop, Android, opted-in iOS users). Deletion means you lose tracking on those segments. The optimal setup is both pixel and Conversion API running simultaneously.

Here's the coverage:

• Pixel fires: 40% of iOS users (who opted in) + 90% of desktop + 95% of Android.
• Conversion API catches: 100% of server-side events (forms, calls, purchases).
• Overlap: 50–60% of conversions are tracked twice, which is fine; Facebook deduplicates automatically.

Together, you're now tracking 85–95% of actual conversions. Alone, you're tracking 40–60%.

When Conversion API Does NOT Fix Your Problem

Conversion API is powerful, but it's not a cure-all. If your issue is one of these, Conversion API won't help:

• You're not getting leads at all. Conversion API doesn't fix bad targeting, poor ad copy, or wrong audience. If you're running ads to the wrong people, API won't change that. See why your Facebook ads aren't getting leads first.
• Your landing page is slow. If your mobile landing page takes 5+ seconds to load, 50% of iOS users bounce before converting. The pixel can't track a conversion that never happened. Mobile page speed is a silent killer; fix that first.
• Your call-tracking system is broken. If your phone number isn't ringing, Conversion API can't help. Check your call-tracking setup before assuming it's a tracking issue.
• Your conversion definition is wrong. If you're tracking "page view" instead of "purchase," Conversion API will dutifully report 1,000 page views. Make sure the event you're tracking is something that actually makes money. A roofing contractor should track "qualified lead call" (5+ min duration, homeowner property), not "contact form submission." See how to score and define quality leads.
• Your business genuinely has poor product-market fit. If 1,000 people call and 2 buy, Conversion API will prove it—and prove you should change your pitch, not your tracking.

Conversion API is a data pipe. It doesn't fix the tap at the other end.

The Economics: When to Invest in Setup

Conversion API setup costs $500–$2,500 plus $0–$100/month for third-party tools. Here's when it's worth it:

If you spend $1,500+/month on Facebook ads: Yes, set up Conversion API immediately. You're losing 15–25% of optimization data, which translates to $225–$375 in wasted budget every month. The setup pays for itself in 2–3 weeks.

If you spend $300–$1,500/month: Use a free integration (Shopify, WooCommerce plugin, or native CRM sync). Skip paid third-party tools and developer hire. You get 70% of the benefit with zero setup cost.

If you spend under $300/month: Your tracking loss is small ($30–$50/month). Don't hire a developer. Use a free integration if available. If not, wait until your budget grows.

A small plumbing company in Nashville running $400/month in ads was losing 5–10 conversions per month to tracking blindness (about $50 in wasted budget). The owner decided not to hire a developer. Instead, she switched to Facebook Lead Ads, which collect conversions natively in Leadria and send them directly to her CRM—no pixel or API needed. This is a simpler path for very small budgets.

Testing and Monitoring Your Setup

Once Conversion API is live, give it 48–72 hours before declaring victory. Facebook's learning algorithm needs time to reweight audiences based on new data.

What to watch:

• Conversion count should jump 15–40%. If it doesn't, check that events are firing (use the Test Event Tool). If they are, your problem may not be tracking—see the "Does NOT Work" section above.
• Cost-per-conversion should drop 20–30%. Not because conversions are cheaper, but because the true cost becomes visible. If CPC was $15 when you were missing 40% of conversions, it's really been $9–$10 all along. Now you see the truth.
• Conversion delay should be under 10 minutes. In Ads Manager, click a conversion → Conversion Details → Event Time. If the timestamp is 1–2 hours old, your API integration is slow (likely a Zapier delay). This is liveable but not ideal.
• Check for duplication. Open Ads Manager → Conversions → Conversion Quality. Facebook shows a "Low Event Quality" warning if it's seeing the same conversion twice in the same second. This is usually fine and gets auto-deduplicated, but too much duplication can confuse the algorithm.

Next Steps: Optimizing After API is Live

Once you have full visibility into conversions, you can actually optimize. With tracking blindness, you were flying partially blind. Now:

• Identify your winning audiences. With complete data, you can see which 3–5 audience segments convert best. Double budget to those.
• Cut underperforming creatives. If a video ad is getting clicks but zero conversions, you can now see it. Kill it and replace it. Before, you thought conversions were low because you weren't seeing them.
• Test landing pages. A/B test two versions to the same audience and measure conversion rate. With complete tracking, you'll see the winner in 7–14 days instead of guessing.
• Expand to sister platforms. Now that you know your Facebook ROAS, test Google Local Services Ads or Instagram Reels ads with confidence. You have a benchmark.

If you're generating your own leads through Leadria, Conversion API integrates seamlessly: as soon as a lead lands in your account with a phone number, your system can fire a "Lead" event to Facebook. This closes the loop: you see the ad → you get the lead → you track it back → Facebook optimizes. No guessing.