Email deliverability: why your emails land in spam (and how to fix it)
Only 83.5% of legitimate marketing emails make it to the inbox — down from 87% in early 2024, according to Validity's data cited by Litmus. Roughly 6.7% land in spam and 9.8% disappear entirely. If you're sending from a clean domain with good copy and still wondering why nobody's replying, the math says you're not imagining it.
Authentication gets the headlines, but it's only one piece of the puzzle. We've covered SPF, DKIM, and DMARC in a separate guide on email authentication — this one is about everything else inbox providers are quietly judging you on.
Deliverability isn't a setting — it's a verdict
Your delivery rate and your inbox placement rate are not the same thing. Mailgun's research found that 88% of senders can't correctly define what "delivery rate" measures. The catch: a "delivered" email can still land in spam. Spam folders are part of the recipient's mail server, so the mail technically arrived — it just arrived where nobody looks.
Inbox placement is what actually matters. Industry average sits around 83.5%, per Validity. Senders aiming to be competitive in 2026 target 90% or higher.
Inbox providers — Gmail, Yahoo, Microsoft, Apple, Fastmail — each run their own filtering models, but they all weigh the same general signals:
- Who you are (authentication and reputation)
- Who's getting the mail (list quality)
- What's in the message (content and structure)
- How recipients react (engagement)
Get one of those wrong and the others can't carry you.
Sender reputation: your invisible credit score
Every domain and every IP that sends email accumulates a reputation score. Warmy describes it as a 0–100 scale: above 90 means reliable inbox placement, 70–89 means inconsistent results, anything below 70 effectively means the spam folder by default.
What feeds the score? Pretty much everything. Litmus and Postmark both highlight the same factors:
- Spam complaint rate — keep it under 0.1% if you can. Anything above 0.3% triggers Gmail and Yahoo's bulk-sender penalties; above 0.5% you're heading toward a block.
- Bounce rate — over 2% signals you're sending to a list you don't really know.
- Volume consistency — sudden spikes look like compromise or list buying. Steady cadence reads as legitimate.
- Spam trap hits — these are dormant addresses ISPs use as bait. One hit on a recycled trap is bad news; one hit on a pristine trap (an address that never opted in) can land you on a blocklist.
Reputation is sticky. Damaging it takes a single bad week. Repairing it takes weeks of careful sending — which is why most experienced marketers treat it the way you'd treat a credit score, not a metric.
What to do when reputation is already damaged
Litmus recommends a four-step recovery:
- Identify which ISP is filtering you — usually one provider goes bad before the others.
- Cut volume to your most engaged subscribers only at that provider.
- Pause low-engagement automations like cart-abandonment or six-month re-engagement flows.
- Increase volume gradually as engagement recovers.
You're rebuilding a signal, not flipping a switch.
Warmup: why new domains can't sprint
A brand-new domain has no reputation. None. Sending 5,000 emails on day one looks identical to a phisher who just bought a domain and started blasting.
Warmup is the answer — gradual volume ramp paired with positive engagement signals. Mailivery and Postmark both recommend the same general shape: start with 5–10 sends a day from a new domain, increase by roughly 30–50% per week, and keep volume predictable. A typical warmup takes 4–6 weeks before you're sending at full scale.
Two warmup mistakes that kill reputations before they exist:
- Skipping it for cold outreach. Sending fifty cold emails a day from a fresh domain is the fastest way to a permanent blocklist.
- Inconsistent ramp. 50 emails Monday, 800 on Tuesday, 30 on Wednesday — providers read that as suspicious activity, not a marketing schedule.
If you're using a tool like Doxiefy to run sequences, treat the first month on any new sending domain as warmup time. Send to your warmest subscribers first. Build the signal before you need it.
List hygiene: the cheapest way to fix deliverability
Most deliverability problems are list problems wearing a costume.
Mailgun reports that only 32.7% of senders actively monitor blocklists, and the #1 cause of getting blocklisted is hitting spam traps. A spam trap is an email address that exists only to catch senders who don't have permission. They get on lists through scraped data, expired addresses recycled by ISPs, and lists bought from "B2B data providers."
A clean list does three things at once: it lowers your bounce rate, raises your engagement rate, and keeps you off blocklists. That's three reputation wins from one decision.
Practical hygiene rules
- Never buy a list. Every reputable source — Mailgun, Postmark, Litmus — agrees. Purchased lists are spam-trap minefields.
- Use confirmed opt-in for new subscribers when possible. The extra friction filters out garbage signups.
- Remove hard bounces immediately. They're permanent failures — sending to them again signals incompetence.
- Sunset inactive subscribers every 6–12 months. If someone hasn't opened or clicked in that window, they're either gone or unhappy. Either way, mailing them is dragging your engagement rate down.
- Validate your list periodically. Tools that ping for SMTP validity will catch dead addresses before you do.
The instinct to keep every email address you've ever collected is understandable. It's also expensive — bigger lists with worse engagement perform worse than smaller lists with strong engagement. Every time. No exceptions.
Content triggers: what filters actually flag
Modern spam filters don't work off a fixed word list. They use machine learning models trained on patterns of bad mail. That said, some patterns reliably trigger them.
Postmark's spam playbook calls out the obvious ones:
- ALL CAPS subject lines
- Excessive emojis (think 🔥🔥🔥 or 💰💰💰)
- URL shorteners like bit.ly — phishers love them, so filters distrust them
- Mismatched subject lines and body content
- Broken HTML, missing body tags, image-only emails with no text
Less obvious but just as damaging:
- Single-image emails. A picture with no text reads like an attempt to hide content from filters.
- Heavy link-to-text ratios. Five links and twenty words looks promotional in the worst way.
- Hidden text or white-on-white tricks. This is 1998 spam tactics — modern filters catch it instantly.
- Attachments on cold mail. Spreadsheets and PDFs from unknown senders are filter bait.
For small businesses and solo creators, the practical version is simple: write the email like a human, keep the link count reasonable, use full branded URLs, and don't send a single image with "click here" pasted on top.
Engagement signals: what recipients teach the filter
This is the part most senders underestimate. Inbox providers watch how recipients behave, and they use those behaviors to decide where future mail goes.
Positive signals build reputation:
- Opens (less weighted than they used to be — Apple Mail Privacy Protection broke that)
- Clicks
- Replies — replies are gold
- Forwards
- Marking as "not spam"
- Adding the sender to contacts
- Starring or flagging the message
Negative signals tear it down:
- Deleting without opening
- Marking as spam
- Ignoring repeatedly
- Unsubscribing (less harmful than complaints, but still negative)
The deliverability gap between an engaged list and a stale one is huge. Gmail in particular routes mail differently for senders whose recent recipients consistently engaged versus senders whose recipients consistently ignored.
That's why segmentation isn't just a marketing nicety — it's a deliverability tool. Mailing your most engaged 20% builds reputation. Mailing your least engaged 20% destroys it. Treating both segments identically is how senders end up in spam without understanding why.
ISP behavior: not all inboxes filter alike
Quick mental model that helps when you're triaging issues:
- Gmail weighs engagement heavily — replies and clicks matter more here than almost anywhere.
- Yahoo / AOL is stricter on complaint rates and shared-IP reputation.
- Microsoft (Outlook, Hotmail, Live) leans hard on authentication alignment and is the toughest on cold outreach.
- Apple Mail doesn't filter much itself but its Mail Privacy Protection makes open-rate data noisy.
If your deliverability tanks, check provider-by-provider before assuming a global problem. A 95% inbox rate at Gmail and 40% at Microsoft tells you a very different story than a flat 60% across the board.
A 30-day deliverability tune-up
If your numbers are slipping and you don't know where to start, here's a sequence that actually works:
Week 1. Audit authentication. SPF, DKIM, DMARC at minimum. Run mail-tester or Google Postmaster Tools and fix any red flags.
Week 2. Clean the list. Remove hard bounces, suppress anyone with no open or click in 9+ months, run a validation pass.
Week 3. Segment by engagement. Send only to your most engaged tier while reputation recovers. Pause re-engagement campaigns until basic deliverability is back.
Week 4. Audit content. Pull your last ten campaigns and check for the triggers above. Rewrite anything image-heavy or link-stuffed.
The order matters. Cleaning a list before fixing authentication wastes effort. Auditing content before solving reputation is rearranging deck chairs.
Frequently asked questions
How do I know if my emails are landing in spam?
Use Google Postmaster Tools for Gmail data, Microsoft SNDS for Outlook, and a tool like mail-tester or GlockApps for inbox-placement testing across providers. Don't rely on your own inbox — your account is biased toward your own mail.
What's the single biggest factor in deliverability?
Engagement. Authentication gets you considered; engagement decides whether you stay in the inbox. Senders with high reply and click rates outperform senders with perfect technical setup but stale lists, every time.
Can I recover from a spam-folder drop quickly?
Quickly is relative. Expect 2–6 weeks of disciplined sending — to engaged subscribers only, with consistent volume — to see meaningful recovery. Trying to "test" your way out by sending more makes things worse.
Does sending volume affect deliverability?
Yes, both ways. Too low and you have no signal — providers don't know you. Too high too fast looks like a compromised account. Steady, predictable volume aligned to subscriber expectations is what providers reward.
Do I need a dedicated IP to fix deliverability?
For most small businesses and solo creators — no. Postmark and Mailgun both note that dedicated IPs only help senders pushing 300,000+ emails a month. Below that, you'll struggle to maintain enough volume to keep the IP warm, and shared IPs from a reputable provider perform better.
Final thoughts
Deliverability isn't one problem. It's four — authentication, reputation, list, content — interacting with how recipients react. Fix one in isolation and the others will still drag you down.
The good news: the senders who get this right aren't using secret tools. They're authenticating properly, warming up patiently, mailing only people who asked, writing like humans, and watching engagement instead of vanity metrics.
If you're building email sequences for a small business or as a solo creator, that's exactly the discipline Doxiefy is designed for — AI-assisted campaigns that respect inbox rules and your subscribers' attention. Join the waitlist and we'll let you know when it's your turn to try it.