Home/Email Deliverability/Inbox Placement

Inbox placement: delivered is not inboxed

Your log says the message was accepted. It does not say a human ever saw it. Inbox placement is the difference, and it is the only delivery number worth reporting.

Verified By George Hartley, CEO · Updated July 28, 2026

What inbox placement measures

Inbox placement is the proportion of accepted mail that reaches the primary inbox, as opposed to the promotions tab, another filtered tab, or the spam folder. It is a placement measurement, not a transport one.

Every one of those destinations reports the same thing to your sending infrastructure: 250 OK, delivered. That is the single most important fact about deliverability reporting, and it is why a delivery rate of 99% can sit alongside an inbox rate of 55% without anything appearing broken.

Delivered is not placed
acceptedthe server said yes
receiver filtering
inbox
spam folder
the split is invisible to the sender
Your logs stop at accepted. The inbox-or-spam decision happens after, where you cannot see it.

How you actually measure it

You cannot derive placement from your own logs, because the information is not in them. The only reliable method is a seed panel: a set of monitored mailboxes spread across Gmail, Outlook, Yahoo and Apple, included in the send, then inspected to see which folder the message reached at each provider.

Roughly twenty seeds gives a usable read for a single send. It is a sample, not a census, so treat a two-point move as noise and a ten-point move as real. Run an inbox test against the same segment you actually send to, because placement varies by recipient engagement, not just by domain.

The Promotions tab is not the emergency

People spend enormous energy trying to escape the Promotions tab. Recipients who want that mail go and read that tab.

Spam-folder placement is the real failure, and it is a different problem with different causes. Ed has looked at tools like Mailmend that exist specifically to push mail from Promotions into Primary, and there is a whole category of products doing this, with Instantly running a warm-up network of 4.2M+ accounts that rotates inboxes and emulates human reads. That approach treats a filter as an adversary to be gamed, and the providers adjust faster than the tools do.

We do not do that, and would not. The durable version of the same outcome is engagement: mail people opted into and actually open lands in Primary on its own.

Content that filters cannot read

One concrete and very common cause of bad placement: image-only mail. If you send nothing but images, Gmail, Yahoo and Outlook cannot read an image, so they see a message with no interpretable content and treat it with suspicion. A readable text-to-image balance is not an aesthetic preference, it is a placement input.

The other structural one is the sending domain. When you add a sending domain you own, mail goes out from your address on our infrastructure, and the reputation accrues to you. Operating without domain verification is possible to start, but it is needed for real inbox delivery, and our guides/sending-domains material exists because this is the step most new senders skip.

If placement is already bad, work the diagnostic order: authentication first, then blacklist status, then list hygiene, then warmup.

Placement is not only a marketing problem. The difference between transactional and marketing mail changes what good placement even means, and delivery events from the webhooks feed are how you detect a placement problem without waiting for someone to complain.

Go deeper

References

The primary sources behind the rules on this page. Provider policy and the underlying standards, not vendor marketing.

Common questions

What is inbox placement?

Inbox placement is the share of accepted mail that lands in the primary inbox rather than a filtered tab or the spam folder. Delivery only measures acceptance by the receiving server.

How do I measure inbox placement?

With a seed panel: a set of monitored mailboxes across Gmail, Outlook, Yahoo and Apple that you include in a send, then check to see where the message landed at each provider.

Is the Promotions tab bad for deliverability?

Less than people assume. Spam-folder placement is the real failure.

Delivered right, by default, not by testing

An inbox test tells you about a problem after you have already sent to real subscribers. Nitrosend's AI agent authenticates, paces and formats every send correctly before it goes out, so inbox placement stays high because nothing was done wrong, not because someone caught it with a test first. That is the case for running email through an agent instead of a dashboard: the checks above become unnecessary rather than routine.

Run your whole email stack from your AI agent, with deliverability built into every send.

Simple pricing. Unlimited contacts.

Every plan includes full stack emailing: Flows, Newsletter Campaigns and Transactional Email, plus our NitroWheel LLM and all agent integrations (Claude, ChatGPT, Codex, Cursor and others). Pay for what you send, not who you store.

Daily allowances depend on your plan and sender standing. Strong list, domain and delivery evidence can raise standing, including on day one. Trusted receives the full plan allowance; available email credits, safety checks and delivery pacing still apply.

Free
$0
forever
  • Emails 8,000then 500/mo
  • Email types Transactional & Marketing
  • AI actions 20/mo
  • Contacts Free & Unlimited
  • Brands 3
  • Seats 1
  • Recipients / rolling 24h 100–5,000
  • Email validation Prepaid only
Start free
Ultra
$100
per month
  • Emails 125,000/month
  • AI actions 5,000/mo
  • Brands 10
  • Seats 10 · Domains 10
  • Frontier AI Included
  • Dedicated IP Available
  • Recipients / rolling 24h 1,000–625,000
  • Email validation Prepaid only
Get started
Enterprise
$300
per month
  • AI actions Unlimited
  • Unlimited brands & domains Included
  • SSO / SAML Included
  • 99.9% SLA Included
  • Recipients / rolling 24h Contracted
  • Email validation Prepaid only
Get started

Free forever. No credit card required. See full comparison →