SPF, DKIM and DMARC without the tears

Setting up email authentication records is one of those subjects that looks trivial until you ship it to production. In this post we unpack the practical trade-offs we've learned running Host Tonic infrastructure at scale, and share the exact playbook our team uses today.

Why this matters

Every millisecond of latency and every avoidable outage compounds into lost conversions, angry support tickets, and eroded trust. The teams that grow fastest are the ones that treat performance and reliability as a first-class product feature.

• Faster page loads directly correlate with higher conversion rates.
• Predictable infrastructure lets developers ship features instead of fighting fires.
• Modern tooling makes best practices cheaper to adopt than ever before.

Our recommended setup

We prefer a layered approach: a fast origin, aggressive caching, and observability at every hop. The pattern is boring on purpose — boring means predictable, and predictable means sleeping through the night.

# Provision a new project
$ tonic init my-app --stack node --region sgp
$ tonic deploy --scale auto

"The best infrastructure is the one you never think about. That's what we ship, every day."

Common pitfalls to avoid

Under-provisioning is cheap in the short term and expensive during the launch spike. Over-provisioning burns money quietly. Autoscaling with sensible ceilings — and alerts on both directions — is the middle ground we recommend to almost every customer.

Wrapping up

If you take one thing away, let it be this: measure before you optimise. Set baselines, define budgets, and ship small changes with clear feedback loops. Your future self, and your users, will thank you.

Written by

Sana Islam

Developer Advocate at Host Tonic