In September 2021, Apple shipped iOS 15 and macOS Monterey with a feature called Mail Privacy Protection (MPP). It was a single toggle in Settings, buried under Mail > Privacy Protection. Most users turned it on without thinking about it.
That toggle broke email open tracking for roughly half the internet.
How open tracking used to work
Email open tracking relies on a 1x1 transparent pixel — a tiny image embedded in the HTML of your email. When the recipient opens the email, their mail client fetches the image from a tracking server. The server logs the request: this recipient opened this email at this time from this IP address.
It's been the standard for 20+ years. Every ESP uses it. Mailchimp, HubSpot, SendGrid, Postmark — they all report open rates based on pixel loads.
What Apple Mail Privacy Protection does
When MPP is enabled, Apple Mail downloads all remote content — including tracking pixels — at the time the email is delivered, not when the user reads it. Apple routes these requests through its own proxy servers, masking the recipient's IP address and location.
The result: every email delivered to an Apple Mail user with MPP enabled registers as "opened," regardless of whether the recipient actually read it.
As of 2025, Apple Mail accounts for approximately 49% of all email opens globally. On some B2C lists, it's 60%+. That means up to half your "opens" are Apple's servers prefetching pixels, not humans reading emails.
How broken the numbers actually are
Consider a list of 10,000 subscribers where 50% use Apple Mail with MPP enabled.
| Metric | Reported | Reality |
|---|---|---|
| Total opens | 6,000 (60%) | Unknown |
| Apple Mail opens (all fake) | 5,000 | ~0 real signal |
| Non-Apple opens | 1,000 | ~1,000 real |
| Actual open rate | 60% | Somewhere around 10-15% |
A 60% open rate looks incredible. It's also fiction. You have reliable data for about half your list, and that half is showing you a 20% open rate at best.
Some segments are worse. B2C lists skew heavily toward iPhone users. If 75% of your list uses Apple Mail, your open rate is almost entirely noise.
Mailchimp and HubSpot still push open rates
This is the part that frustrates developers. Mailchimp's campaign reports still show open rate as the primary metric. Their "engagement" scoring — which determines things like audience segmentation and send-time optimization — still factors in opens. HubSpot's email health dashboard features open rate front and center.
Both platforms acknowledge MPP exists. Neither has meaningfully adjusted their default dashboards or scoring models. You're making decisions based on a metric that's been unreliable for over four years.
What to track instead
Here are the metrics that still work, ranked by signal quality.
Click-through rate
Clicks require a human to tap or click a link. Apple Mail doesn't prefetch link destinations. A click is a real action by a real person.
Track click-through rate (CTR) as your primary engagement metric. If you're A/B testing subject lines, measure by clicks, not opens. If you're segmenting your audience by engagement, segment by click recency.
Reply rate
A reply is the strongest email engagement signal possible. Someone read your email, thought about it, typed a response, and hit send. For drip sequences and onboarding flows, reply rate tells you whether your content is resonating.
This matters most for transactional and relationship-building emails. If you're sending a welcome sequence and nobody replies to your "hit reply and tell me what you're building" email, the sequence isn't working.
Conversion rate
The metric that actually pays the bills. Track conversions — signups, purchases, upgrades — attributed to email sends. This requires linking your email click tracking to your conversion events, but it's the only metric your business should optimize for.
Open rate never told you whether an email made money. Click-through rate gets you closer. Conversion rate gives you the answer.
Bounce rate
Bounces tell you about list hygiene and deliverability health. A hard bounce rate above 2% means your list has bad addresses. A rising soft bounce rate might indicate reputation problems.
This isn't an engagement metric — it's a deliverability metric. But it's reliable, and it matters.
Spam complaint rate
Google requires senders to keep spam complaint rates below 0.1% (1 complaint per 1,000 emails). Exceed 0.3% and you risk being blocked entirely. Yahoo has similar thresholds.
Track this obsessively. A rising complaint rate is the most urgent signal in email — more important than any engagement metric.
Google Postmaster Tools
If you send to Gmail recipients (you do), Google Postmaster Tools is the only reliable source for inbox placement data. It shows you:
- Domain reputation (High, Medium, Low, Bad)
- IP reputation
- Spam rate (from real user reports)
- Authentication pass rates (SPF, DKIM, DMARC)
- Delivery errors
No ESP can give you this data because Google only shares it with verified domain owners. Set it up. Check it weekly. It's free.
How PushMail handles this
We don't report open rates as a primary metric because we know the data is unreliable. Instead, PushMail tracks the signals that actually work:
Click tracking — Every link in your emails is tracked through our redirect service. Click events are recorded with timestamps and delivered via webhooks in real time.
Bounce processing — Hard bounces and soft bounces are captured via SendGrid event webhooks, logged per-send, and used to automatically suppress bad addresses.
Complaint tracking — Spam complaints from ISP feedback loops are captured, logged, and used to automatically unsubscribe the complainant. Your complaint rate stays visible in your dashboard.
Webhook-first architecture — All events (clicks, bounces, complaints, deliveries) are available via webhooks. Pipe them into your analytics system, your data warehouse, or your own dashboards. You own the data.
We don't surface vanity metrics because vanity metrics lead to bad decisions. Track what's real.