Email deliverability benchmarks

Use deliverability benchmarks as a diagnostic sequence. First check whether mail reaches the primary inbox, then inspect bounces and complaints, and use opens as a later clue about attention. Roughly half of the opens in a typical B2B dashboard are machine events, so an open can appear without a human reading it.¹ Placement and complaint data give firmer sender health checks than opens alone.

Use a matched benchmark

Benchmarking works when the comparison matches the sending context and metric definition. Pick the threshold before inspecting performance.

For sender health analysis, use a cold email deliverability guide and provider-specific response data.² Reported B2B email open rates range from 15.1% to 55.71%, a spread wide enough to make the metric difficult to use without context.³

A deliverability rate above 89% is considered good.⁴ A rate above 95% is considered excellent.⁵ These labels describe a broad deliverability rate, so check inbox placement separately before calling a sender healthy.

Run the check in order

Use this sequence when a send looks weak or when you are setting a baseline. Each step answers a different question, so clear each earlier gate before moving on.

Start with placement. Delivery does not show where the message landed, so check the share reaching the primary inbox and break it out by provider where possible.

Use 90% or higher primary inbox placement as a healthy benchmark.⁶ Check the aggregate figure alongside provider-level results, then continue only when placement is at or above that threshold across the providers that matter to your list.

Next separate failed addresses from messages that reached a mailbox. This shows whether the list itself is putting pressure on sender health.

Use a bounce rate below 3% as a healthy benchmark.⁷ Record hard and soft bounces separately, because a temporary failure needs a different investigation from a permanently invalid address.

When the rate rises, inspect the list source, verification process, and recent address changes before increasing volume. Resume scaling only after the cause is clear and the rate is below your working threshold.

Complaint rate is the fastest warning that recipients or mailbox providers may reject future mail. Treat it as a stop signal even when placement and opens look acceptable.

Keep spam complaints below 0.1%.⁸ As the rate approaches that boundary, review targeting, expectations, and unsubscribe handling before sending more.

Authentication gives you a basic sender health check before you interpret engagement. Review the records for the sending domain and confirm that they match the stream being measured.

Improved SPF, DKIM, and BIMI adoption helps deliverability, while stricter spam filtering and ISP policies have tightened inbox placement.⁹ Nine percent of senders still lack basic SPF records.¹⁰ Verify these records before changing copy or judging the audience response.

Use opens after placement, bounces, complaints, and authentication have been checked. An open-rate threshold can point to a problem, though it cannot settle the diagnosis by itself.

Open rates below 40% signal a potential deliverability or list-quality problem worth investigating.¹¹ Compare the result with a similar audience, provider mix, and sending context. If opens hold while downstream engagement weakens, inspect the message's relevance after the open. The effective click rate in the DACH region is dropping by 1.7 percentage points.¹²

What not to do

These mistakes can turn a benchmark into false reassurance or an unnecessary alarm.

• Keep opt-in bounce ranges tied to their sending context and check the fit before reusing them. Opt-in guidance puts total bounces below 2%, with hard bounces around 0.3% to 0.5% and soft bounces around 1% to 1.5%.¹³
• A hard bounce rate of 1% to 5% is excessive.¹⁴ A rate above 5% is very excessive, and either level indicates a list problem that needs to be addressed.
• Act before complaint rates move past the boundary. At 0.1%, mailbox providers may treat the sender as a bad sender and filter email as spam for all recipients.¹⁵
• Treat aggregate placement as a starting point. Average inbox placement is 83.1%, and the gap between providers is wide.¹⁶

Make the result usable

A benchmark is useful when it points to the next check. Keep one baseline for each sending context, with placement, delivery, hard bounces, soft bounces, complaints, authentication status, opens, and downstream engagement in separate fields.

When a threshold trips, pause expansion and change one condition at a time. Isolate the list, sender setup, provider mix, or message so the next result shows what caused the movement. Keep the benchmark attached to the same context until you have enough comparable sends to replace it.