Why Your Emails Stop Reaching the Inbox (and What to Do About It)

Your email platform says 98% delivered. So why did sales from the list fall? That delivered figure tells us the receiving servers accepted most messages. It does not tell us how many reached an inbox, landed in spam, sat in a Promotions tab or went to subscribers who no longer care. Before we change the sending domain, we need to find the gap between acceptance and attention.

For a direct-response publisher or list owner, the whole send has to work. Authentication matters, but so do the promise in the subject line, the offer in the copy, the domains behind every link and the people selected to receive it. We look at those parts together because a technical repair cannot rescue an offer that attracts complaints, and a sharper offer cannot repair rejected mail.

First, name the failure

We separate three conditions: rejected or deferred messages, accepted messages filtered away from the inbox, and accepted messages that reached an audience unlikely to act. A fourth possibility sits in the creative: subscribers see the message but do not trust the claim or want the offer. Each requires different evidence.

Start with the email service provider's acceptance, bounce and deferral logs, broken out by mailbox provider. Then compare complaints, unsubscribes, clicks, purchases and revenue per send for the same campaigns and cohorts. If Gmail acceptance changes but Yahoo does not, examine Gmail response codes. If acceptance stays stable while clicks and sales fall on one offer, read the copy and inspect the links before blaming DNS.

A Promotions tab is not a spam folder. Neither a delivered rate nor an open rate proves primary-inbox placement. Google says it does not track open rates and cannot verify the accuracy of third-party open reporting. Apple Mail Privacy Protection can load tracking pixels even when a reader has not opened the message. We treat opens as a secondary diagnostic, not a clean measure of human attention.

Check the sender setup on a real message

Google's sender guidelines require SPF or DKIM for all senders to personal Gmail accounts. Senders of more than 5,000 messages a day to those accounts need SPF and DKIM, DMARC, From-domain alignment with SPF or DKIM, and one-click unsubscribe for marketing and subscribed messages, with a visible unsubscribe link in the body. Google accepts p=none as a minimum DMARC policy for that group. It advises keeping the user-reported spam rate in Postmaster Tools below 0.10% and avoiding 0.30% or higher.

Yahoo's bulk-sender guidance likewise calls for SPF, DKIM and DMARC with at least p=none, plus functioning unsubscribe headers and a visible unsubscribe link for marketing messages. Yahoo says to honor unsubscribe requests within two days. If a platform claims to handle the header, we inspect a delivered message instead of taking the configuration screen at its word.

We review the actual SPF, DKIM and DMARC results in message headers and map every legitimate sender: the campaign platform, transactional system and follow-up automations. The visible From domain must align with the authenticated SPF or DKIM domain for DMARC. A green DNS check alone does not show whether the live message passes. Before tightening a DMARC policy, we identify legitimate sources and look at reports so we do not interrupt our own mail. p=none is a monitoring policy, not a guarantee of placement.

We also check volume history. Google recommends consistent sending, gradual increases and monitoring response codes rather than sudden spikes. Yahoo asks senders to respect the frequency people signed up for. A list used to a weekly email should not suddenly receive a daily run without a reason and a review of feedback.

Copy and subject lines: make the promise match the message

A subject line earns the open, but the email has to fulfill it. The FTC's CAN-SPAM guide says a commercial message's subject must accurately reflect its content. Google says not to start a subject with Re: or Fwd: unless it is an actual reply or forward. We do not dress a promotion as a personal response to win a click.

We check the complete path from subject to preview text, first paragraph, offer, call to action and landing page. If the subject promises a practical answer, put the answer in the email or clearly explain what the link provides. If the email sells a third-party offer, identify what the reader is being asked to visit. Make links visible and understandable; Google explicitly advises that recipients should know what to expect when they click. A vague button leading through several surprise destinations is a trust problem even if the message is authenticated perfectly.

Complaint patterns tell us which expectations we broke. Compare complaints by subject, offer, acquisition source and frequency. A spike after an aggressive subject does not by itself prove the copy caused it, but it tells us where to inspect. We would look at whether the claim was supported, whether the landing page matched it, and whether recipients had asked to get that sort of promotion. Test a clearer version on an appropriate cohort rather than sending the same mismatch harder.

There is a legal and an operating reason to keep the copy honest. Misleading subjects can violate CAN-SPAM in the United States; unwanted messages can also invite spam reports. Google says frequent spam reports can lower a domain's reputation and affect future placement. Clear copy, sensible frequency and a visible way to leave the list protect both the relationship and the channel.

Check every link before you send, especially affiliate offers

The link audit is part of the send, not a task to do only after a complaint. Spamhaus's Domain Blocklist can be used during content inspection to identify, classify or reject mail containing listed domains, including URLs in the body. Not every receiver uses this list in the same way. Still, a troubled affiliate, tracking or redirect domain can put an otherwise legitimate campaign at risk. Google's sender guidance also recommends checking the status of domains linked to yours when delivery goes wrong.

We open every link in the final message, including the unsubscribe route, on the version that will actually send. Follow the redirect chain, note each hostname, confirm the destination is live and consistent with the copy, and inspect the domains against relevant reputation tools. Check tracked links too: the visible offer domain may not be the first domain a filter sees. If a third-party destination changes after approval, stop and review it again. Record which offer owner approved the final URL and when the check happened.

For affiliate-driven programs, the mailing partner matters as much as the offer link. Google warns that a brand associated with affiliate spam may see its other messages marked as spam and recommends monitoring affiliates. The FTC says a business cannot contract away its CAN-SPAM responsibility by hiring someone else to handle email marketing; depending on the message, both the promoted business and the sender may be responsible. We set expectations for permitted subjects, claims, suppression and links before an affiliate mails, then check what actually went out. A contract is not a substitute for oversight.

A clean blocklist lookup is not an inbox-placement certificate. It removes one possible explanation. If a linked domain is listed, trace the exact domain and redirect, establish whether the list is used by the affected receiver, fix the source of the listing and only then retest. Swapping a tracking link to hide the issue while leaving the bad destination in place solves nothing.

Segment by engagement instead of mailing everyone equally

We split active readers and buyers from people who have stopped responding. Start with clicks, purchases and replies where they are available, along with recent signup or preference signals. Do not let an apparent open alone keep someone indefinitely in the active group: Apple Mail Privacy Protection can inflate opens. Equally, a buyer who reads without clicking every campaign is not the same as an unknown address imported years ago.

For the inactive group, send less frequently while you test whether the relationship can be renewed. A short re-engagement sequence can remind subscribers what they signed up for, offer a clear choice of topics or cadence, and ask them to confirm they still want mail. Suppress those who remain inactive after that attempt, and suppress hard bounces and opt-outs without waiting for a re-engagement campaign. Yahoo recommends periodically reconfirming inactive subscribers and removing invalid recipients promptly.

About 90 days since the last meaningful action is a common starting point for a review, not an official Google or Yahoo rule and not an automatic deletion threshold. Adjust it to the purchase cycle, the promised cadence and the history of each cohort. Someone on a monthly product update has had fewer opportunities to engage than someone receiving daily offers. Write down the definition, test the re-engagement group separately and look at complaints as well as clicks and sales before expanding volume again.

Segmentation is not just list hygiene. It determines what each subscriber receives next. Active purchasers may need relevant follow-up rather than the same acquisition pitch. A dormant segment may need a single clear reason to stay instead of another blast. If the frequency or offer changes, tell readers plainly. We want a smaller audience that chose to hear from us rather than a larger number in the ESP that neither buys nor wants the mail.

Check reputation, then test the fix

Review provider-level reputation reporting, including Google Postmaster Tools when Gmail is material to the list. Look at complaints, sending domains, IPs and relevant blocklists alongside actual SMTP responses. If a shared IP is involved, ask the provider what it sees: Google notes that activity from other senders on a shared IP can affect reputation. Moving to a dedicated IP is not an automatic cure, and changing infrastructure before diagnosing the offer or segment can make the investigation harder.

We built BlacklistGuard to watch your sending domains and IPs for blacklist listings. That watch helps surface a specific reputation issue to investigate; it does not prove that an accepted message reached the primary inbox. We compare it with provider reports and controlled placement tests using messages close to the real campaign. Seed tests are samples, so we also watch live responses and commercial results.

Change one meaningful variable at a time when possible. Log the date, audience, provider, message, links and action taken. If authentication, sending domain, offer and segment all change together, we cannot tell which intervention worked. After a repair, recheck several sends and give a named operator ownership of the follow-up. A single good campaign is not a recovery plan.

Decide what to do before the next campaign

• Rejects or deferrals: Read provider response codes and repair the specific authentication, policy or pacing issue before retrying traffic.
• Accepted mail with poor placement: Compare provider reputation, complaints, links and representative placement tests. Avoid declaring victory from delivered rate alone.
• Complaints concentrated in one offer: Pause that offer, inspect subject and body promises, destination, affiliate behavior and signup expectations.
• Low response from old cohorts: Reduce frequency, run a focused re-engagement sequence and suppress people who remain inactive.
• Healthy placement but weak sales: Look at offer fit, copy and the path after the click rather than changing DNS for a conversion problem.

On a large health list, we achieved a median 24% open rate and $0.38 EPC across nearly 6,000 sends. That is one list's result, not a universal benchmark. We judge a program by the placement evidence, subscriber response and commercial outcome together.

Our email and deliverability work connects reputation, the send calendar, segmentation, suppression and list monetization. If your platform reports strong delivery while revenue slips, book a 30-minute diagnostic and bring the provider-level report, recent campaign creative, live URLs and cohort results. For a broader operating view, our free Growth Blueprint is a guide to seven systems for scaling affiliates, ads and email without becoming the person who runs them.

Sources

• Google, Email sender guidelines
• Yahoo, Sender Best Practices
• FTC, CAN-SPAM Act compliance guide
• Spamhaus, Domain Blocklist
• Litmus, Apple Mail Privacy Protection for marketers