Where to start, by what you're trying to do
A 233-article blog can be paralysing. Most senders who arrive here have a specific problem rather than a general curiosity, so the right entry point depends on what kicked off the search. The map below points to the right category for the most common arrival reasons — and the FAQ at the bottom of this page covers the meta-questions that every sender eventually asks.
Email Deliverability88 articles
The largest category by volume, because deliverability is where most senders eventually find themselves. Articles cover the root causes of inbox-versus-spam classification: list hygiene, engagement signals, complaint thresholds (the 0.1-0.3% tolerances at Gmail and Microsoft), why emails go to spam at the level of headers and content, and the diagnostic process for figuring out which of a dozen possible causes is actually responsible for a specific drop in performance.
Email Authentication21 articles
Since February 2024 (Google + Yahoo) and May 2025 (Microsoft), proper SPF, DKIM, and DMARC are no longer best practices — they are required for any domain sending more than 5,000 messages per day to those mailbox providers. Yet, according to the EasyDMARC 2026 Adoption Report, only 7.6% of internet domains enforce DMARC. This category covers staged DMARC rollout (p=none → p=quarantine → p=reject), DKIM key rotation, SPF record troubleshooting (the 10-lookup limit is where most senders fail), and how to read DMARC aggregate reports without getting lost in the XML.
MTA Administration19 articles
The configuration layer most ESPs hide from their users. PowerMTA tuning (vMTAs, per-ISP queueing, accounting log analysis), Postfix in production (main.cf, master.cf, transport_maps for per-ISP routing), KumoMTA — the modern Rust-based open-source competitor built by PowerMTA's original creator — and the trade-offs between commercial MTAs (Halon, MailerQ) and open-source alternatives. This category is for operators running their own infrastructure, not consumers of managed services.
IP Reputation Management18 articles
The phase nobody warns new senders about: the first six weeks on a new IP, when reputation has not yet stabilised and inbox placement settles into a 70-80% valley before climbing back to 90%+. This category covers IP warming protocols (week-by-week schedules calibrated by destination ISP), feedback loop setup (JMRP for Microsoft, Yahoo FBL, Google's Postmaster Tools), the dedicated-versus-shared-IP decision (the 50K/month volume floor), and how to recover when an IP gets listed on Spamhaus.
Email Infrastructure17 articles
Architecture-level guides for senders past the ESP tier. Stream separation (the most impactful single architecture decision in email — keep marketing complaints from damaging transactional reputation), multi-IP scaling, dedicated domain strategy, the cost comparison at high volume between self-hosted PowerMTA and managed ESPs (the break-even is around 500K/month), and ecommerce-specific infrastructure patterns (abandoned cart isolation, seasonal volume planning, transactional-marketing separation).
Cold Email Strategy15 articles
Cold email is structurally different from marketing email: different ISP risk profile, different infrastructure requirements (domain rotation, isolated sending IPs), different legal frameworks (CAN-SPAM tolerates unsolicited B2B with proper unsubscribe; GDPR effectively prohibits it without prior consent in the EU). This category covers the three-tier infrastructure pattern that separates cold from marketing from transactional, the inbox-warming SaaS landscape (Folderly, MailReach, Warmy, Warmbox), and the major sending platforms (Instantly, Smartlead, Apollo, Lemlist).
Deliverability Recovery12 articles
What to do after the incident — a Spamhaus listing, a Gmail spam folder placement spike, a complaint rate above 0.3%, an ISP-specific block. Recovery is rarely fast and never linear. These guides cover the diagnostic steps that come before remediation (delisting without root-cause analysis produces re-listing within days), the data sources every incident response procedure depends on (PowerMTA accounting logs, Gmail Postmaster Tools, Microsoft SNDS), and the recovery timelines you should actually expect.
Compliance & Legal10 articles
PCI DSS 4.0 added DMARC as a requirement in March 2025 for any organisation handling cardholder data, with monthly fines from $5,000 to $100,000 for non-compliance. NIS2 and DORA in the EU recognise email authentication as a required cybersecurity control. GDPR enforcement remains active. This category maps the legal frameworks (CAN-SPAM, CASL, GDPR, LGPD) onto operational practice — what consent looks like, what a one-click unsubscribe must do under RFC 8058, and how regulated industries (healthcare, financial services) configure email infrastructure to meet HIPAA and FINRA requirements.
Transactional Email9 articles
The category where deliverability problems are highest-stakes — a marketing campaign in spam annoys subscribers; a password-reset email in spam locks users out of the product. The architectural rule that matters most: transactional and marketing must run on separate IPs and ideally separate domains. Marketing complaints damage reputation; if transactional shares that reputation, the product breaks. This category covers stream isolation, transactional API providers (Postmark, Mailgun, SendGrid, AWS SES), and ecommerce lifecycle email patterns.
Email Security9 articles
Authentication for security, not just for deliverability — the same SPF, DKIM, and DMARC configuration that gets emails into the inbox also protects the brand from spoofing and Business Email Compromise. This category covers BEC defences (technical controls beyond DMARC enforcement), MTA-STS and TLS-RPT (transport-level encryption enforcement), DANE (DNSSEC-bound TLSA records for stronger transport security), and phishing prevention at the infrastructure level rather than the user-training level.
SMTP Configuration8 articles
Protocol-level configuration for senders running their own SMTP infrastructure. Port 25 versus 587 (server-to-server versus client submission), FCrDNS configuration (the silent gate at Layer 5 that operators routinely overlook), per-ISP transport maps in Postfix, and the SMTP error code reference — particularly 5xx codes, where 550 has at least seven distinct variants and the wrong remediation makes the situation worse.
Bulk Sending Operations7 articles
The operational practice at the highest end of volume — 1M+ daily, where the rules change. Per-ISP throttling matters more than aggregate sending rate. List cleaning has different stakes (a 0.1% bad-address rate is 10,000 bounces at 10M sends, enough to trigger reputation events). PowerMTA's vMTA configuration becomes the central tuning surface. This category is small in count but covers the operational reality of senders past the threshold where managed services stop being economical.
Frequently asked questions
Where do I start if I'm new to email deliverability?
What's the difference between this blog and operational notes?
Are there guides for cold email specifically, or only marketing email?
Is the content current with 2026 ISP requirements?
Do you cover specific MTA configuration in detail or just concepts?
Need help applying any of this to a real programme?
Articles answer general questions. The right answer for your specific sending programme depends on volume, list quality, current deliverability state, and operational capacity. We run infrastructure assessments that map your situation to the right architecture — without a sales pitch built into the diagnosis.
Schedule an infrastructure assessment →