In Mailgun’s 2025 survey of 1,100 senders, 39% said they rarely or never conduct list hygiene—a gap that can affect both campaigns and product email. These email deliverability best practices give SaaS teams a concrete checklist for protecting critical messages while improving campaign operations.
Email deliverability is the likelihood that a message accepted by a receiving mail server reaches the inbox rather than spam or another folder. Improve it by authenticating each sending domain, mailing people who expect your messages, ramping volume carefully, and monitoring complaints, bounces, and engagement. Mailbox providers decide where each message lands.
Distinguish delivery from inbox placement
Sending is not the same as delivery, and delivery is not the same as inbox placement. A server can accept a message while a mailbox provider filters it into spam; a successful API response alone does not prove that a person saw the email in their inbox.
Look for changes across several signals. A rise in hard bounces can point to invalid addresses or a list-source problem. More spam complaints, weaker engagement, or a sudden drop in inbox placement can indicate that recipients do not recognize the mail—or that authentication and reputation need investigation.
Mailbox providers, including Gmail, make their own filtering decisions. In the State of Email Deliverability 2025 report, 48% of surveyed senders named staying out of spam as a top challenge. Treat deliverability as an operating practice shared by engineering and marketing, not a DNS task you complete once.
Email deliverability best practices: authenticate every sending domain
SPF identifies which mail servers are authorized to send for a domain. DKIM adds a cryptographic signature that receiving servers can use to check that a message was authorized and its signed content was not changed. DMARC tells receivers how to handle mail that fails authentication and provides a way to receive reports; alignment connects the authenticated domain to the domain in the visible From address.
Set up and verify SPF, DKIM, and DMARC for the domains and subdomains you actually use. If product mail comes from notify.example.com and campaigns use news.example.com, check both sending paths and the visible From domain. A record on the root domain does not prove that every subdomain or sending service is configured correctly.
These examples show the shape of records; replace the example domain and provider-specific values with the values your email service provider supplies:
example.com TXT "v=spf1 include:mail.example.net -all"
selector1._domainkey.example.com TXT "v=DKIM1; k=rsa; p=<public-key-from-your-ESP>"
_dmarc.example.com TXT "v=DMARC1; p=none; rua=mailto:dmarc@example.com"
Start DMARC in a monitoring mode such as p=none while reviewing reports, then decide on enforcement after identifying legitimate senders. Google’s email sender guidelines specify SPF or DKIM for all senders to Gmail, and SPF, DKIM, and DMARC for senders sending more than 5,000 messages a day to Gmail accounts. Google also specifies one-click unsubscribe for bulk marketing and subscribed messages, and advises keeping the reported spam rate below 0.1% and preventing it from reaching 0.3% or higher. Those figures describe Google’s guidance, not a universal inbox-placement formula.
For our own domain setup, Volanea provides guided SPF, DKIM, and DMARC configuration and automatically checks domain authentication. That supports the setup step; teams still need to confirm each application and campaign uses the intended From domain.
Separate transactional email from marketing campaigns
A password reset and a product announcement have different urgency and recipient expectations. Separate the streams by sending domain or subdomain, provider configuration, or another appropriate identity boundary. This makes it easier to investigate a campaign complaint spike without mixing it into password-reset and receipt metrics.
Separation does not erase reputation risk. Do not introduce a fresh subdomain to evade complaint or blocking problems. Instead, keep campaign consent and suppression management strict, and limit transactional messages to the event the user requested. Document which system sends each message type, its From address, and where complaints and unsubscribes are handled.
We built Volanea to combine transactional email, campaigns, and automations on one API, using Amazon SES for sending. That shared platform still benefits from clear stream boundaries in your application and reporting. For developers, the documented sending endpoint is POST /v1/send; the API also supports batch sending, templates, contacts, segments, webhooks, and suppressions. Your message type and recipient rules should remain explicit in the code that calls it.
Build a permission-based, healthy contact list
Send campaigns only to people with a clear reason to expect them. A signup form should say what someone will receive; double opt-in adds a confirmation step that can document intent and catch mistyped addresses. It adds friction, so choose it based on the signup flow and the value of stronger confirmation—not as a substitute for clear consent.
Use this list-maintenance routine:
- Remove hard-bounced addresses promptly and suppress unsubscribe requests across campaign sends.
- Investigate bounce spikes by signup source, rather than repeatedly retrying invalid addresses.
- Segment inactive contacts before re-engagement, then stop mailing those who do not respond.
- Preserve the source and permission context for each contact so your team can explain why they receive mail.
Never treat purchased or scraped addresses as a shortcut to audience growth. Those recipients did not necessarily ask for your messages, and poor engagement or complaints can damage sending reputation. Mailgun’s 2025 survey found 9.4% of respondents admitted purchasing lists or scraping contacts in the prior two years. Our guide to automatic email opt-in risks explains why adding people without meaningful consent creates avoidable risk.
Send consistently and warm up new infrastructure carefully
A new sending domain or IP has little established history. Start with recent opt-ins or people actively using your product, and increase volume only while responses remain healthy. For example, a SaaS team preparing to send a new opt-in digest to a small, engaged audience might try 50 messages on day one, 100 on day two, 200 on day three, 400 on day four, and 800 on day five. This is an example sequence, not a prescribed schedule: stop increasing if bounces or complaints rise, or engagement drops.
Keep volume and frequency predictable. If a weekly digest becomes daily, explain the change and watch unsubscribes, complaints, and engagement. When signals worsen, pause the affected campaign or reduce its volume while you investigate; maintaining the same rate can compound the problem.
A dedicated IP is not automatically a better fit. It offers more control over the IP’s reputation, but the sender must establish a consistent history on that IP. Shared infrastructure can suit lower or variable volumes. Volanea sends over shared Amazon SES infrastructure and does not offer dedicated IPs; teams choosing it do not manage a separate dedicated-IP warm-up.
| Sending setup | What it means for the sender |
|---|---|
| Shared IP infrastructure | Sending reputation is shared across traffic on the infrastructure; the provider manages the IP pool. |
| Dedicated IP address | The sender has an IP-specific reputation to establish and maintain through consistent sending. |
| Volanea | Email is sent through Amazon SES shared infrastructure; dedicated IPs are not offered. |
Write messages people recognize and want
Use a From name and address recipients can connect to your company, and make the subject accurately describe the message. Keep transactional content focused on the requested action, such as confirming an order or resetting a password. Target campaigns to relevant segments instead of sending every offer to every contact.
Make unsubscribe controls easy to find in subscribed and promotional mail, and honor requests promptly. Google’s sender guidance calls for one-click unsubscribe on bulk marketing and subscribed messages. An easy opt-out gives recipients a clear alternative to marking unwanted mail as spam.
Do not rely on a list of forbidden words or a fixed text-to-image ratio to control placement. Content matters in context, alongside sender identity, consent, recipient behavior, and sending history. Test rendering and links in real clients, and make sure the message still communicates its purpose when images are not displayed.
For application messages sent through Volanea, the from field accepts either a bare address such as receipts@example.com or a formatted value such as Acme Receipts <receipts@example.com>. That lets teams present a recognizable sender without changing the envelope’s purpose. Volanea also provides an SMTP relay for transactional sending; its SMTP username is volanea, and the password is the project’s secret API key.
Monitor reputation and deliverability signals
Review trends, not isolated totals. Track hard and soft bounces, spam complaints, unsubscribes, delivery outcomes, and engagement by message type and recipient provider. If Gmail is a significant destination, Google Postmaster Tools can provide sender-related data, including spam-rate information for eligible traffic; use it alongside your provider’s event and suppression data.
Treat a complaint rate of 0.10% as a signal to investigate, not a target or a promise of inbox placement. Google advises bulk senders to keep reported spam rates below 0.1% and prevent them from reaching 0.3% or higher; see its sender guidelines for the requirements. Review transactional and campaign performance separately so a complaint spike on a newsletter is not hidden by healthy password-reset volume.
Volanea’s reputation guard slows sending when a list may harm the sender domain. That maps to a practical control in your checklist: monitor list risk and intervene before continuing at full campaign volume. Our email deliverability and sending guides cover related operational topics.
Troubleshoot a sudden drop in inbox placement
When a stable stream starts landing in spam or bouncing, isolate whether the problem affects one message type, domain, provider, or all sending. Check in this order:
- Confirm SPF, DKIM, DMARC, and alignment for the exact From domain and sending service.
- Compare bounce codes, complaints, unsubscribes, and engagement with the period before the drop.
- Identify changes in volume, frequency, templates, audience segments, or signup sources.
- Check the relevant domain or IP against email blacklist tools, then investigate the listing and its cause.
- Correct the cause, test with a small and relevant audience, and increase volume only when signals recover.
For example, an SMTP response of 550 5.1.1 commonly indicates a recipient address does not exist; suppress that address rather than retrying it indefinitely. A temporary 421 response indicates a temporary failure, so use your provider’s retry behavior and inspect the accompanying response text before treating it as a permanent bounce. Codes help classify an event, but the full response and provider logs provide the context.
A blacklist listing is a clue, not a complete diagnosis. Identify which list reported the domain or IP and whether the receiving provider cites it. Fix the source of unwanted mail, compromised credentials, or poor list quality before seeking removal or resuming normal volume; changing domains to keep sending the same traffic does not address the cause.
Choose an email service provider that supports good practices
A provider cannot guarantee inbox placement, but its tools can make responsible sending easier to operate. Compare authentication support, bounce and complaint events, suppression handling, reporting, and the fit for transactional messages and campaigns.
| Capability | Why a SaaS team needs it | Volanea detail |
|---|---|---|
| Domain authentication | Confirms sending setup for the actual domains and subdomains in use. | Guided setup for SPF, DKIM, and DMARC, with automatic authentication checks. |
| Sending and suppression | Connects application events to sending rules and prevents mail to suppressed recipients. | POST /v1/send processes suppression checks, contact upserts, template rendering, tracking, and delivery in one pipeline. |
| Multiple message types | Keeps product mail and campaigns manageable within one email operation. | Transactional email, campaigns, and automations are available on one API. |
| Test behavior | Prevents a test configuration from being mistaken for a real delivery. | sk_test_ keys render and log messages but never deliver them. |
| Infrastructure fit | Determines whether the team manages an IP-specific reputation. | Volanea uses shared Amazon SES infrastructure and does not offer dedicated IPs. |
Before rollout, map each message to its trigger, recipient permission, From identity, and suppression behavior. If using a test key, verify the resulting render and logs without expecting delivery; switch to the project’s secret API key for live sending. That distinction catches a common integration mistake before it is mistaken for an inbox-placement issue.
FAQ
Why are test emails not reaching recipients through Volanea?
A Volanea key beginning with sk_test_ renders and logs messages but does not deliver them. Use the project’s secret API key for live sending, then confirm the domain is verified before sending to real recipients. A successful test render is useful for checking content, but it is not evidence of recipient delivery.
When should a SaaS team consider a dedicated IP?
Consider one when sending volume is consistent enough to establish and maintain a distinct IP reputation and your team can operate the warm-up and monitoring process. It is not a default upgrade for every SaaS sender. Volanea uses shared Amazon SES infrastructure and does not offer dedicated IPs, so teams that require a dedicated IP need a different infrastructure fit.
What should engineers log for each send?
Record a message identifier, application event, message type, recipient domain, sending identity, and the provider’s outcome or error. These fields let a team distinguish a password-reset failure from a campaign issue and connect bounce codes to the originating workflow. Avoid logging secret API keys or message content that your application does not need to retain.
How do I start sending campaigns and transactional mail from one platform?
Map transactional events and campaign audiences separately, then define each stream’s From identity, consent rules, and suppression behavior before sending. Volanea combines both message types with automations on one API. Start with a small, permission-based audience and use the resulting bounce, complaint, and engagement trends to guide the next send.
Start free with 1,000 emails a month