Deliverability problems almost never announce themselves. Nothing bounces, no error appears, and the campaign report looks normal — opens are just quietly lower than they used to be, because a growing share of your mail is landing in spam.
This is a practical checklist for small teams: the authentication that is now effectively mandatory, the list habits that matter more than any setting, and how to diagnose a drop.
Key takeaways
- Authenticate first: SPF, DKIM and DMARC on your sending domain. Without these, bulk mail to major providers is increasingly rejected outright.
- Engagement is the main signal. Mailbox providers watch whether people open, reply and don't complain — not your content.
- A large dormant list actively hurts you, and on contact-priced platforms it costs money too.
- Warm up new domains and IPs over two to three weeks. Never send full volume on day one.
- Use a subdomain for bulk sending so a marketing problem cannot damage your business email.
- Complaints are the fastest way to get filtered. Make unsubscribing easier than reporting spam.
The authentication baseline
These are DNS records on your sending domain. If you do nothing else on this page, do these, and verify them rather than assuming your provider handled it.
| Record | What it does | If it's missing |
|---|---|---|
| SPF | Lists which servers may send for your domain | Mail is treated as unauthenticated and filtered harder |
| DKIM | Cryptographically signs your messages | Receivers cannot verify the mail was not altered |
| DMARC | Tells receivers what to do when SPF/DKIM fail, and where to report | No policy, no visibility, and bulk mail is often rejected |
| One-click unsubscribe | A standard header letting recipients opt out in one action | More complaints, which is the worst possible signal |
Start DMARC at p=none so you receive reports without affecting delivery, read what those reports show for a couple of weeks, then tighten. Going straight to a strict policy is how teams accidentally block their own invoices.
Why engagement matters more than content
Small teams spend a lot of effort avoiding "spam words" and very little on the thing that actually decides placement. Mailbox providers care primarily about how recipients behave: do they open, reply, move mail out of spam — or delete without reading and click report.
That has a practical consequence which feels counterintuitive: sending less mail to fewer people usually improves delivery to everyone, because your aggregate engagement rate rises.
List hygiene, which is most of the job
Sunset the dormant
Contacts who have not opened anything in six to twelve months are dragging your engagement rate down. Run a re-engagement sequence, then stop mailing the ones who ignore it. On contact-priced platforms this is also the largest single saving available.
Never buy or scrape a list
Beyond the legal exposure under GDPR, CAN-SPAM and equivalents, purchased lists contain spam traps — addresses that exist purely to catch senders who did not get consent. Hitting one can damage your domain reputation for months.
Use confirmed opt-in where it is practical
It costs you some list growth and buys a list that engages. For B2B with a long sales cycle the trade is usually worth it.
Clean at import, not later
Validate syntax and remove role addresses (info@, sales@, admin@) before the first send rather than after the bounce rate has already been recorded.
Want to see what your own numbers look like? Start a workspace and price it against the table above.
Start free14-day free trial · No credit card required
Domain strategy
Send bulk marketing from a subdomain — news.yourdomain.com or similar — rather than your root domain. Reputation is tracked at domain level, so this keeps a bad campaign from affecting the mail your team sends to customers and suppliers.
For cold outreach, go further: separate domains entirely, each warmed independently. Cold outreach generates complaints at rates that will damage any domain doing double duty, and recovering a burned primary domain is far harder than buying another sending domain.
Warming up, properly
A new domain or IP has no reputation, and volume from an unknown sender looks exactly like a spam run. Ramp deliberately.
- Week one: small volumes to your most engaged contacts only — people who opened something recently.
- Week two: roughly double, still weighted to engaged segments.
- Week three: approach normal volume, watching bounce and complaint rates at each step.
- Throughout: if complaints rise or opens fall, stop increasing. The ramp is a response to signals, not a schedule.
- After a migration: treat the new platform as a new sender even if the domain is old, because the sending infrastructure has changed.
Diagnosing a drop
When open rates fall, work through this in order rather than rewriting subject lines first.
| Check | What it tells you |
|---|---|
| Did open rates fall for one provider or all? | One provider means a reputation issue with that receiver |
| Did bounce rate change? | Rising hard bounces point at list quality |
| Did complaint rate change? | The fastest route to filtering |
| Did volume or frequency change recently? | A sudden spike looks like a spam run |
| Are SPF, DKIM and DMARC still passing? | DNS changes break these silently |
| Did you add a new list source? | One bad source contaminates the whole sender |
Send a test to seeded addresses across the major providers to see where mail is actually landing. Your own inbox is not evidence — you have engaged with your own mail more than anyone.
Since Apple began pre-loading images for Mail users, a significant share of "opens" are automated rather than human. That makes open rate unreliable as an absolute number — but still useful as a trend against itself, provided you compare like with like. Where a decision matters, weight clicks and replies instead.
Questions to ask before you commit
Whichever way you are leaning, these are the questions that change the answer. Work through them against your own numbers rather than anyone's feature matrix — including this one.
- Are SPF, DKIM and DMARC verified on the exact domain you send from?
- What is your complaint rate? If you do not know it, find it today.
- When did you last remove a dormant contact?
- Do you send bulk mail from your root domain?
- Where did each list source come from, and did those people ask?
- Is one-click unsubscribe present, and is it easier than the spam button?
If you cannot answer the first three, you are not ready to choose between Scalry and your own sending setup yet — you are still deciding what you need, which is a cheaper problem to solve first.
Cold outreach is a different discipline
Everything above applies to people who asked to hear from you. Cold outreach breaks several of those assumptions at once, and treating it as "marketing email but colder" is how domains get burned.
Separate infrastructure, always
Dedicated sending domains, warmed independently, never your primary. Assume some will eventually be unusable and plan for replacing them rather than rescuing them.
Volume per mailbox stays low
Spreading sends across several mailboxes at modest daily volume looks far more like normal correspondence than one mailbox sending hundreds. The constraint is per-mailbox behaviour, not total volume.
Replies are the signal that saves you
Cold mail that generates genuine replies builds reputation; cold mail that generates deletions and complaints destroys it. That makes relevance a deliverability control rather than just a conversion one — a tighter list mailed less often outperforms a broad one.
The legal position varies more than people assume
Unsolicited commercial email is treated very differently across jurisdictions, and the rules generally follow the recipient rather than the sender. Before running cold outreach into a new market, check what applies there — this is one of the few areas where getting it wrong is a regulatory problem rather than a performance one.
Frequently asked questions
Why are my emails going to spam?
Most often engagement rather than content: too much mail to people who never open it, which teaches mailbox providers your sends are unwanted. The other frequent causes are missing SPF, DKIM or DMARC records, a sudden volume increase from a cold domain, and a list source nobody consented to.
Do I need SPF, DKIM and DMARC?
Yes. Major mailbox providers now require authentication for bulk senders, and unauthenticated mail is increasingly rejected rather than filtered. Start DMARC at p=none to collect reports without affecting delivery, then tighten once the reports look clean.
How do I warm up a new sending domain?
Over two to three weeks, starting with small volumes to your most engaged contacts and roughly doubling each week while watching bounces and complaints. Treat the ramp as a response to those signals rather than a fixed schedule, and stop increasing if anything worsens.
Should I send marketing email from my main domain?
Better not to. Use a subdomain for bulk marketing so reputation problems cannot affect the mail your team sends to customers and suppliers. For cold outreach, use entirely separate domains — complaint rates there will damage any domain doing double duty.
Is open rate still a useful metric?
As a trend against itself, yes. As an absolute number, no — Apple's Mail Privacy Protection pre-loads images, so a meaningful share of recorded opens are automated. Where a decision depends on it, use clicks and replies instead.
Verify authentication and check your complaint rate first — both take minutes and cause most problems. You can rebuild one real campaign inside Scalry before moving anything else.
Start your free trialSee pricing14-day free trial · No credit card required
