Skip to content
Hot take: Most emai...
 
Notifications
Clear all

Hot take: Most email 'best practices' guides are written by content marketers, not engineers.

2 Posts
2 Users
0 Reactions
16 Views
(@emmae)
Reputable Member
Joined: 2 months ago
Posts: 255
Topic starter   [#26345]

Hi everyone! I'm still pretty new to the whole email infrastructure side of things, coming from a Salesforce admin background. I've been trying to learn more about deliverability because our sales team's outreach is getting spotty results.

I've read *so many* blog posts and "ultimate guides" about email best practices. They all say the same things: keep your subject line under 50 characters, send on Tuesday mornings, use one big image, etc. But when I talk to a couple of engineers here, they roll their eyes and start talking about SPF/DKIM alignment, dedicated IPs, and bounce handling. It feels like two completely different worlds!

So my question is: what are the most important things a beginner should actually focus on that the "content marketing" guides usually miss or get wrong? I want to make sure I'm building on a solid technical foundation first, not just optimizing the pretty stuff.

For example, is warming up a new IP more critical than A/B testing subject lines? Things like that. Really appreciate any insights you can share from the engineering side.

Thanks!



   
Quote
(@baller_analytics)
Honorable Member
Joined: 4 months ago
Posts: 483
 

Senior data engineer, previously at a 500-person B2B SaaS company. I was responsible for all customer-facing data pipelines, including email event tracking for a multi-product suite. We sent roughly 50 million transactional emails per month via SendGrid and SES.

Your engineers are correct. The "content" advice is window dressing on a house with no foundation. Here are the actual pillars, in order.

1. **Authentication (SPF/DKIM/DMARC):** This isn't optional. Misalignment here will get your mail routed to spam before a single content rule is applied. A DMARC rejection policy is the only thing that materially stops spoofing. A failing grade here makes everything else pointless.
2. **IP Reputation & Warming:** Yes, this is more critical than A/B testing subject lines. A new dedicated IP is toxic until proven otherwise. A proper warm-up is 4-6 weeks, starting at ~200-300 emails/day to the most engaged users. Jumping to 10k/day on day two will blacklist you. Shared IP pools are fine for low volume (<100k/month), but you inherit the pool's collective sins.
3. **Bounce/Complaint Handling:** This is a real-time suppression system, not a weekly cleanup. A hard bounce (invalid domain) must be removed immediately - within one send cycle. Complaint rates above 0.1% will trigger blocks. Your system must process these feedback loops from ISPs automatically.
4. **Infrastructure Logging:** Most guides talk about open rates. You need to log *everything*: queued, deferred, delivered, processed, dropped, bounced (with type). Link this to your user DB. The "why" for a failing campaign is in these logs, not in a pretty report. Without this, you're blind.

If you're new and own both strategy and infrastructure, use a platform like SendGrid or Mailgun on a shared IP first. They handle the heavy lifting on auth, bounce processing, and IP reputation. Only consider a dedicated IP or a service like Amazon SES if you're sending over 200k emails per month consistently. To choose, tell me your current monthly send volume and if you're doing mostly broadcast or transactional emails.


If it's not a retention curve, I don't care.


   
ReplyQuote