Email delivery rate and email deliverability rate answer two different questions, and conflating them hides real problems. Delivery rate is the share of sent messages a receiving server accepts: (emails accepted ÷ emails sent) × 100, per Postmark’s deliverability glossary. It counts a message as a success the moment a mail server responds “250 OK,” even if a spam filter buries it afterward. SendGrid’s Event Webhook documentation defines its delivered event the same narrow way: the message reached the receiving server, nothing more. Deliverability, or inbox placement rate, measures what happens next: emails delivered to the inbox divided by emails sent, times 100, according to Litmus. A sender can post a 99% delivery rate and a 60% inbox placement rate in the same week. The server accepted the mail. The recipient never saw it.
Delivery Rate: What It Measures and Where It Falls Short
Delivery rate is the easiest deliverability number to pull, because every ESP reports it natively. Twilio SendGrid fires a delivered event the moment the message has been “successfully delivered to the receiving server,” logged separately from bounce and block events. Postmark uses the same accepted-versus-sent math: send 5,000 messages, have 30 fail as hard bounces or timeouts, and the delivery rate is 4,970 ÷ 5,000 × 100 = 99.4%, per Postmark’s own worked example. Postmark reports maintaining roughly “99% delivery rates for both broadcast and transactional emails” across its platform.
The catch is what delivery rate does not track. Acceptance by a receiving server says nothing about whether the message landed in the primary inbox, a promotions tab, or a spam folder nobody opens. A high delivery rate rules out one class of problem (bounces, blocks, DNS misconfigurations) but says nothing about the one that actually costs opens and revenue: filter placement.
Deliverability (Inbox Placement Rate): What Happens After Delivery
Deliverability, most often reported as inbox placement rate, measures where an accepted message actually lands. No ESP can report this number directly, because an ESP only sees what happens up to the point its message leaves its own infrastructure; it has no visibility into a recipient’s spam folder. That is why inbox placement rate has to be measured indirectly, through seed testing or mailbox-provider tools, instead of being read off a sending dashboard the way delivery rate is.
The distinction matters because delivery rate and inbox placement rate point to different problems. A dropping delivery rate usually means bad addresses, a broken MX record, or a block from the receiving server. A dropping inbox placement rate, with delivery rate holding steady, points to sender reputation, an authentication gap, or content that trips a spam filter, the territory covered in Email Deliverability Best Practices.
Delivery Rate vs. Deliverability Rate at a Glance
| Metric | What It Measures | Formula | Where You Get the Number |
|---|---|---|---|
| Delivery rate | Receiving server accepted the message | (Emails accepted ÷ Emails sent) × 100 | ESP dashboard (SendGrid, Postmark, etc.) |
| Deliverability / inbox placement rate | Message landed in the inbox, not spam or promotions | (Emails delivered to inbox ÷ Emails sent) × 100 | Seed list test or mailbox-provider tools |
A 99% Delivery Rate Can Still Mean a 60% Inbox Placement Rate
Take a hypothetical send of 5,000 messages, the same volume as Postmark’s worked example above. If 30 bounce, delivery rate lands at 99.4%: 4,970 messages accepted by receiving servers. Now run a seed test across those same providers. If only 3,000 of the 4,970 accepted messages land in the primary inbox, with the rest routed to spam or a promotions tab, the inbox placement rate is 3,000 ÷ 5,000 × 100 = 60%. Both numbers are true at once. A campaign report showing 99.4% delivered tells a sender nothing is obviously broken, while 40% of the list never saw the message land anywhere they would check it.
How Each Rate Gets Measured in Practice
Delivery rate comes straight from ESP event logs. Every major provider exposes delivered, bounced, and blocked events, either in a dashboard or through a webhook a development team can consume directly; SendGrid’s Deliverability Insights and Postmark’s Bounce API both report this in near real time. For teams building against these events directly, the way Coldletter exposes delivery and bounce activity through its own webhooks, delivery rate is just a query against events already being received.
Inbox placement rate needs a different method, because no ESP has visibility past the receiving server. Seed list testing sends a campaign to a set of test addresses spread across Gmail, Outlook, Yahoo, and other providers, then reports where each copy landed. GlockApps describes the mechanic directly: the test analyzes each message “to make sure they get to the main folder and don’t end up in spam before you send them to your real subscribers,” reporting inbox, promotions-tab, and spam placement per provider. The other primary source is Google Postmaster Tools, which reports Gmail-specific spam complaint rates from real recipient behavior rather than a seed panel, making it a complement to seed testing rather than a replacement for it.
What Counts as a Good Rate in 2026
Delivery rate benchmarks are tight, because the number excludes filter placement entirely. Litmus sets 90 to 98% as the healthy range, with 95% or higher as the internal target, and treats anything below 80% as a sign of a broken list or a blocked sender.
Inbox placement benchmarks move more, and vary by provider. Validity’s 2026 Email Deliverability Benchmark Report found “the average global inbox placement rate reaching 87.2%, a 3.7% uplift year over year,” with meaningful spread by mailbox provider: Gmail’s average inbox placement rate rose to 89.8%, while Microsoft’s stayed comparatively low at 77.4%, with Yahoo at 87.3% and Apple at 82.0%. A sender averaging around 87% across a mixed list is roughly at the global norm; a sender whose Microsoft-domain recipients land in the 70s should not assume the same list health holds at Gmail.
The Levers That Actually Move Deliverability Rate
Delivery rate problems are usually mechanical: a stale list, a typo-heavy signup form, or a receiving server rejecting a misconfigured domain. Email Bounce Rate covers the specific thresholds and fixes for that half of the problem.
Deliverability problems trace back to levers that a clean delivery rate does nothing to fix: authentication (SPF, DKIM, and DMARC all passing), sender reputation built on low complaint rates and consistent sending behavior, and list hygiene that keeps engagement high. Fixing bounces without checking reputation or authentication keeps delivery rate clean while inbox placement quietly erodes underneath it.
Frequently Asked Questions
What is a good email delivery rate?
Litmus considers 90 to 98% a healthy delivery rate, with 95% or higher as the target for well-maintained sending programs. A delivery rate under 80% signals bounced or rejected mail at a scale worth investigating immediately, usually a stale list or a domain-level block.
What is a good inbox placement rate?
Validity’s 2026 benchmark report puts the global average inbox placement rate at 87.2%. Rates vary by mailbox provider: Gmail averaged 89.8% and Microsoft averaged 77.4% in the same report. A sender tracking meaningfully below these averages on a given provider likely has a reputation or authentication gap specific to that provider.
Why is my delivery rate high but my inbox placement rate low?
Delivery rate only confirms a receiving server accepted the message; it says nothing about spam-filter placement. A high delivery rate paired with low inbox placement usually points to sender reputation, incomplete SPF, DKIM, or DMARC alignment, or content that trips spam filters, not a list or infrastructure problem.
How can I check my inbox placement rate?
Seed list testing tools like GlockApps send a campaign to test addresses across major providers and report where each copy landed. Google Postmaster Tools is a free, Gmail-specific complement that reports real recipient spam complaints rather than a seed panel, though it only covers Gmail traffic.
Does a high bounce rate lower delivery rate or deliverability rate?
Bounces lower delivery rate directly and immediately, since delivery rate is calculated as accepted messages over sent messages. Deliverability responds more slowly: a sustained high bounce rate damages sender reputation over weeks, which then drags down inbox placement even on messages that were accepted without a bounce.
How often should I run an inbox placement test?
Run a seed test before any major campaign and after any change to sending infrastructure, such as a new IP, a new domain, or an ESP migration. Ongoing sending programs benefit from a recurring test on a monthly or per-campaign cadence to catch reputation drift before it shows up in open rates.
I’ve spent my career building software at scale with a soft spot for email: deliverability, lifecycle campaigns, and getting messages to actually land. I started Coldletter to fix what bugged me about transactional and marketing email tools. I’m based in Vancouver.