Your open rate has climbed past 40%. Replies are still zero.
That gap isn't a coincidence. Apple Mail Privacy Protection (MPP), launched in September 2021, prefetches images regardless of whether anyone actually opens the email. Industry-wide open rates jumped after that — not because writing got better, but because the metric started counting machines as readers.
So if open rate is half-machine, what should you watch instead?
Five metrics. With public benchmarks. None perfect, but together they tell you what actually happens after you hit send.
Why open rate stopped meaning anything
A quick refresher on what MPP broke:
- Apple Mail accounts for ~51% of all email opens globally, according to Litmus.
- MPP inflated reported open rates by 35% or more across the industry.
- "Real" iPhone opens dropped from 37% of opens to under 4% — the rest moved into the new "MPP" client column, where Apple prefetched the tracking pixel without anyone actually opening.
After MPP, what platforms like Beehiiv now report as "average" reflects the new baseline. Beehiiv's State of Email puts the 2024 overall average open rate at 38.7%, with 2025 reports trending above 41%.
That doesn't make open rate useless — relative changes within your own list still matter — but you can't trust the absolute number to mean "people read this." You need supporting metrics.
Here are five.
1. CTR (Click-Through Rate)
The percentage of all recipients who clicked at least one link.
- Mailchimp global average: 2.62%.
- Beehiiv top quartile: ~5%.
CTR is honest in a way open rate isn't. Clicks require a deliberate action. Image prefetchers don't click.
The catch: a click isn't endorsement. A great subject line plus an underwhelming body still produces clicks. Use CTR as a check on body strength, not a verdict on whether readers liked what they found.
2. CTOR (Click-to-Open Rate)
Clicks ÷ unique opens. This is the metric Litmus and Mailchimp both now recommend as the post-MPP primary KPI.
Why it survives MPP: the inflation hits the numerator and denominator together. If MPP fakes an extra 100 opens, those phantom recipients aren't clicking anything — so the ratio between clicks and opens stays roughly true to behavior.
Industry benchmark sits near 10%, but the absolute number matters less than your own trend. Comparing CTOR across your last ten issues is one of the cleanest reads of body strength you can get.
3. Reply / Response rate
The strongest signal you can collect — and the rarest.
- Cold email reply rate (2026): 3.43%, top quartile 5.5%. Newsletter reply rates aren't tracked publicly, but they sit somewhere in this range.
- A specific question can earn 1 in 5 replies (~20%), according to Inbox Collective.
The gap between "ask vaguely" and "ask specifically" is roughly 10x. Reply rate isn't a personality test of your readers — it's a function of how cheap you make replying.
But the reply channel itself is partly broken. We covered why in Why readers don't reply to your newsletter: identity exposure, format weight, and email deliverability all gate the one-line reaction before it ever lands in your inbox. That's why publishers increasingly add a separate response channel that doesn't require hitting reply.
4. Unsubscribe rate per issue
Mailchimp's global average: 0.26% per send.
The mistake is looking at unsubscribe rate as an aggregate. The signal is per-issue.
A normal issue lands at 0.2–0.3%. An issue that spikes to 0.8% the same day you sent it did something different — maybe wrong topic, wrong tone, wrong length. The issues with your highest unsubscribe rates are your strongest fit data, because they cost real subscribers to learn from.
Track it per-issue or you're throwing the signal away.
5. Forwards and shares
The hardest metric to measure, and the strongest signal that exists.
You can't reliably count forwards (most email clients don't expose them). What you can do:
- UTM-tagged share links in the footer, unique per issue
- "How did you hear about us?" as the first signup field
- Reader screenshots and DMs — count them manually if you have to
A 1% forward rate on a 1,000-subscriber list is more meaningful than 41% opens. One person re-sending your email to a colleague is a vote on every metric simultaneously: the open, the read, the trust, the recommend.
The 5-metric cheat sheet
| Metric | Benchmark | Funnel position | MPP-affected? |
|---|---|---|---|
| Open rate | ~38–41% (Beehiiv) | of delivered | Yes — inflated 35%+ |
| CTR | 2.62% (Mailchimp) | of delivered | No |
| CTOR | ~10% | of opens | No (ratio-stable) |
| Reply | 1–3% baseline, ~20% with a specific ask | of delivered | No |
| Unsubscribe | 0.26% per send | per issue | No |
| Forwards | qualitative | — | No |
What about list growth rate?
Watch it. Quarterly, not per-issue.
It's an important metric — but it doesn't tell you about this send. The five above do, which is why they're in the working set.
Read the five together
None of these five tells the whole story alone. Together, they triangulate.
A normal issue looks like: open rate stable, CTOR within your usual range, unsubscribes at baseline, maybe one or two replies, maybe a forward. A worth-replicating issue: CTOR spikes, unsubs stay normal, a handful of replies show up. A misfire: CTOR holds but unsubs jump — same body, wrong audience for that topic.
The first four come from your ESP dashboard. The fifth — reactions and replies — you have to build a channel for, because the default reply path is too expensive for casual readers to use.
That's the part the dashboard won't give you. The rest, it will.
References
- Apple — Mail Privacy Protection
- Litmus — Identifying real opens after MPP
- Litmus — Current state of MPP, Q&A
- Beehiiv — 2024 State of Newsletters · 2025 State of Email Newsletters
- Mailchimp — Apple Privacy FAQ & email benchmarks
- Inbox Collective — The secret strategy to ramp up your reply rate · Ways to measure success
