In short: A low inbox rate is not a reason to send more. First separate rejected messages, temporary deferrals, accepted mail that lands in spam, and messages that reach the inbox but get no response. Then compare the affected recipient providers, sending streams, and recent changes. Fix the cause, confirm the signals improve, and only then consider a gradual increase.
When a B2B newsletter or campaign starts underperforming, teams often look at one blended delivery rate or open rate. That can hide the problem. A provider may be rejecting one sending domain while another accepts the same campaign; a mailbox may accept a message but place it in spam; and a weak response rate may reflect the audience or offer rather than delivery.
Treat deliverability as a diagnosis across the sending system and the recipient’s provider. A useful review connects the mail platform’s event logs with domain authentication, provider-specific reputation data, list expectations, and the business outcome the campaign was supposed to support.
1. Define which delivery failure you are seeing
Start by naming the observable symptom. These cases call for different checks:
- Rejected: the receiving provider refused the message. Record the SMTP response, provider, sending domain, IP, and time.
- Temporarily deferred: the provider accepted the connection but asked the sender to retry later. Look for repeated deferrals, queue age, and whether a retry eventually succeeded.
- Accepted but not in the inbox: a successful server handoff does not prove inbox placement. The message may be in spam, filtered, or routed to another folder.
- In the inbox but not acted on: investigate audience fit, timing, content, and the next step. A low response rate alone does not establish a delivery problem.
Do not use opens as the deciding signal. Google says it does not track open rates for Gmail and cannot verify the accuracy of third-party open-rate reporting. Privacy features and measurement differences also make opens a weak proxy for whether a person saw or valued a message.
2. Narrow the problem by provider and sending stream
Break results out before changing a campaign. At minimum, compare:
- receiving provider or domain family;
- marketing versus transactional mail;
- sending domain and, where available, IP;
- campaign, audience source, and send date;
- accepted, rejected, deferred, bounced, unsubscribed, and complained-about messages.
Compare a campaign that is affected with a recent campaign sent to a similar audience. If the decline is limited to one provider or one sending stream, a site-wide copy change is unlikely to be the first place to look. If every provider worsened after a shared DNS or platform change, inspect that shared dependency.
Keep the provider scope clear. Gmail’s published requirements and Postmaster Tools data concern personal Gmail accounts. Microsoft’s high-volume sender requirements describe Outlook.com consumer addresses, including Hotmail and Live. Business mail hosted on Google Workspace or Microsoft 365 can have organization-specific filtering and should be examined through the recipient organization’s administrator or available message trace.
3. Verify authentication for every system that sends
Inventory the services that send from your domain: marketing automation, CRM, event tools, support systems, billing, and any agency platform. A new vendor or forgotten subdomain can create a gap even when the main marketing platform passes its tests.
Check the actual DNS records and message headers, not only a setup screen:
- SPF: the published record authorizes the services that send for the domain and stays within the provider’s limits.
- DKIM: messages carry a valid signature for the intended signing domain.
- DMARC: the From domain aligns with a passing SPF or DKIM identity, and reports go to a monitored destination.
- DNS and transport: confirm the sending IP’s forward and reverse DNS where relevant and that the service uses a secure TLS connection.
Requirements depend on the recipient provider and volume. For mail sent to personal Gmail accounts, Google requires all senders to use SPF or DKIM; senders exceeding 5,000 messages a day to those accounts must use SPF, DKIM, and DMARC, with alignment for direct mail. Yahoo publishes separate requirements for bulk senders, including SPF, DKIM, DMARC, and one-click unsubscribe for marketing and subscribed messages. Microsoft’s published high-volume rule applies to domains sending more than 5,000 messages per day to Outlook.com consumer accounts. Treat these as provider-specific rules, not one universal threshold for every business mailbox.
4. Read provider signals alongside your mail-platform logs
Use the mail service’s delivery logs as the event-level record. Group the original SMTP response and retry outcome by destination provider. A hard bounce, a temporary deferral, and a complaint are different events; don’t combine them into a single “bad email” percentage.
For personal Gmail recipients, Google Postmaster Tools provides dashboards for authentication, domain and IP reputation, spam rate, and delivery errors. Google notes that this data is not real time and may be incomplete at low volumes. Yahoo’s Complaint Feedback Loop can help identify messages reported as spam when the sending mail is DKIM signed. Use the available provider data with your ESP logs; neither view is a complete measurement of every inbox.
Track a small set of signals on the same time window:
- authentication pass rates and alignment;
- provider rejection and deferral responses;
- hard-bounce and complaint rates;
- unsubscribes and delivery volume by audience source;
- qualified replies, meetings, or other campaign outcomes.
Provider dashboards may use different populations and definitions. Keep their metrics separate. A low complaint number is not reassuring if most messages are being filtered before recipients can report them.
5. Check whether recipients expected this message
A technically authenticated sender can still damage trust by contacting people who did not expect the message. Review how each audience entered the list, what they agreed to receive, how often the team has written to them, and whether a CRM or event import changed the audience unexpectedly. Remove purchased or unverified lists from the diagnosis and do not use them as a recovery tactic.
Make sure the unsubscribe path works and that opt-outs, hard bounces, and complaints suppress future sends across connected tools. The guide to keeping marketing suppression rules in sync across campaign tools covers that handoff. If a broken preference sync, import, or authentication change caused the incident, record the cause, owner, and recovery check in a marketing incident log.
6. Change one cause, then watch a controlled recovery
Pause or reduce the affected stream while you fix a confirmed issue. Avoid increasing volume to “push through” a filtering or reputation problem. Make one change at a time when possible, record when it went live, and compare the same provider-level signals before and after.
When the evidence shows a stable recovery, return gradually to the normal audience and cadence. Start with people who recently and clearly engaged or requested the message. Do not suddenly combine an old dormant audience with the regular campaign. A successful test to a handful of internal inboxes is useful for checking rendering and authentication, but it does not establish provider-wide inbox placement.
Set a stop condition before resuming. For example, pause again if rejection or deferral responses rise, authentication fails, complaints increase, or the sending platform reports a growing queue. The appropriate threshold depends on provider, volume, and your normal baseline; use the recipient provider’s current guidance and your own verified history.
Deliverability diagnosis worksheet
- Observed symptom: rejected, deferred, accepted but missing, or low engagement
- Affected providers and time window: ______
- Sending domains, IPs, and services in scope: ______
- SMTP responses and retry outcomes: ______
- SPF, DKIM, DMARC, and alignment checks: ______
- Provider dashboards and their coverage limits: ______
- Recent list, volume, DNS, template, or vendor changes: ______
- Confirmed cause, change owner, and recovery signal: ______
- Stop condition and next review time: ______
Email delivery improves when teams diagnose the receiving provider’s response, authentication, recipient expectations, and campaign outcome together. A larger send can amplify a healthy system, but it cannot repair a broken one.
If email performance changed after a CRM, vendor, or campaign update, request a marketing diagnostic to map the delivery evidence back to the operational change.
Sources and scope
- Google: Email sender guidelines — requirements and recommendations for mail sent to personal Gmail accounts, including authentication, complaint rates, unsubscribe, and open-rate limitations.
- Google: Email sender guidelines FAQ — provider scope and enforcement updates.
- Google: Postmaster Tools dashboards — available Gmail-specific reputation, authentication, spam, and delivery-error data and its limits.
- Yahoo: Sender Best Practices — authentication, complaints, list expectations, and bulk-sender unsubscribe guidance.
- Yahoo: Complaint Feedback Loop — requirements for receiving complaint reports.
- Microsoft: Outlook.com high-volume sender requirements — SPF, DKIM, and DMARC requirements for high-volume senders to Outlook.com consumer addresses.
- IETF RFC 8058 — one-click unsubscribe protocol.
These references describe provider-specific sending requirements, not a guarantee of inbox placement. Recheck them before changing a live sending system. Accessed October 9, 2026.
How did this article land?
Choose one reaction. You can change it anytime.