What nearly a decade at Litmus taught me about email deliverability

After nearly 2 decades in the world of email, there’s one question I still get asked more than any other: “What actually impacts deliverability?” It’s the email world’s equivalent of “what’s the meaning of life?”. Everyone’s got an opinion, but very few can explain it without looking like Charlie from It's Always Sunny.

This blog post covers everything I’ve learned after spending nearly a decade at Litmus and managing an email program for one of the toughest possible audiences: Lifecycle marketers.

Engagement is everything. Inbox providers are constantly monitoring how subscribers interact with your emails (opens, clicks, forwards, replies, or even dragging a message out of spam and into their inbox). All of those actions send strong signals that people actually want to hear from you.

The mistake a lot of marketers make is overthinking personalization. You don’t need to insert someone’s name in three places or reference their last purchase to improve deliverability. What matters more is relevance. The content simply has to feel useful and worth reading. If your subscribers value what you’re sending, inbox providers will notice.

Even with the right setup, deliverability issues can creep in during execution. If you don’t have it right, even great content will struggle. At a minimum, you need to properly configure SPF, DKIM, and DMARC. These records authenticate your email and prove to inbox providers that your messages are legitimate.

You also want to keep your email streams separate. Marketing and transactional emails should live on different subdomains, so one doesn’t damage the other. Always include an unsubscribe link, and make sure it works. Keep your image sizes small (GIFs under 1MB if possible) to avoid slow load times. And finally, don’t underestimate the importance of the sender name. It should be clear, consistent, and recognizable so recipients immediately trust who the email is coming from.

Length on its own isn’t the problem people think it is. Readers today are used to scrolling, so they’ll often go through longer emails if the content is engaging. The real issue is Gmail and Yahoo clipping messages once they pass about 102KB of HTML. When that happens, the bottom of your email gets cut off, and usually that means the unsubscribe link is hidden, which can create compliance risks.

That’s why it’s important to balance copy and design. Keep the code lean, watch your file sizes, and use testing tools to check when clipping occurs. Beyond the technical side, I always recommend looking at click maps. They show you where people stop engaging in the email, which is far more telling than word count alone. And you can use design to your advantage to tease something at the top and reveal it later down the email to encourage scrolling.

Here’s what I run through before every send:

It might sound like a lot, but in practice, it becomes second nature. The checklist is what prevents mistakes when you’re moving fast or under pressure.

Warming up is not something you can rush. I recommend an eight-week launch plan. You want to start by sending to your most engaged subscribers (e.g., the people who consistently open and click). That way, inbox providers immediately see positive engagement signals tied to your new domain.

Each week, slowly expand the audience by adding less engaged segments, but always keep an eye on performance. And be selective about the content you send. Choose messages that historically drive strong engagement. The goal during warm-up is to build a reputation for quality and relevance right out of the gate.

For lifecycle marketers, this also means factoring warm-up timelines into launch planning. If your execs want a new domain ready to go next week, you’ll need to push back (inbox trust simply doesn’t work that way).

If you’ve burned a domain, the first thing you need to do is stop sending. Don’t make the problem worse by trying to brute force your way through. Step back and figure out what caused the issue. Was it a sudden spike in volume? Poor data quality? Falling engagement? Once you know the cause, you can start to fix the problem. When you’re ready, restart slowly using the same warm-up principles: small volumes, engaged cohorts, and proven content.

And here’s something a lot of people overlook: inbox providers actually want to help you if you show accountability. Gmail, Yahoo, and Microsoft all have postmaster forms where you can submit an explanation. Tell them what happened, what you fixed, and how you’re going to prevent it in the future. They won’t always reply; Gmail especially is hit or miss, but Yahoo often gets back within a week or two. Even when you don’t hear back, the act of showing you’re taking it seriously can help the recovery process.

Yes, it’s definitely possible. A burned subdomain isn’t a death sentence. Inbox providers don’t blacklist forever, but they do remember history. That means you’ll be under closer scrutiny for a while.

The way back is patience and consistency. Treat it like warming up a brand-new domain: start with low volumes, stick with highly engaged subscribers, and send content that has a proven track record. If the subdomain had a good reputation before, you can usually recover it. It just takes time and discipline.

I always tell people to look beyond opens. Opens matter, but what’s more important is whether your subject line sets up the action that happens inside the email. That’s where the click-to-open rate comes in. It shows whether the promise of the subject line matched the reality of the content.

To maintain that balance, I’d rather build a smaller, more engaged core audience than inflate metrics with unengaged addresses. For the rest of your list, use re-engagement campaigns to either win them back or remove them. Engagement is the signal inbox providers value most, so keeping your list clean is just as important as the content you’re sending.

Generally, anything under 1% is healthy. Under 0.5% is even better. At Litmus, we averaged around 0.3%. That told us our content was hitting the right audience and that people were finding value in it.

If your unsubscribe rate starts climbing, that’s a warning sign. It usually means either the frequency is too high or the content isn’t aligned with what subscribers expected when they signed up. I’d much rather see people unsubscribe than quietly disengage and start marking emails as spam. This is especially critical in lifecycle, where frequency tends to be higher than in one-off campaigns. You need to watch unsubscribes as a leading indicator of fatigue.

There are a few common triggers. The first is sudden spikes in sending volume, which inbox providers see as suspicious. Another is poor list sourcing caused by buying or scraping email addresses. And then there’s ignoring negative signals, like an increase in unsubscribes or a dip in click-throughs.

Spam filters are increasingly behavioral. If they see a pattern of users deleting your emails without reading them, or worse, marking them as spam, your sender reputation will decline quickly. It’s not usually one big mistake that teams make; it’s a series of ignored warning signs.

I actually think Gmail’s new unsubscribe feature is a good thing. Anything that makes it easier for users to control their inbox is positive in the long run. That said, Gmail’s one-click unsubscribe is hidden in the toolbar, so adoption might be slower than people expect.

I’d love to see usage data in a year. But the reality is, inbox providers introduce features like this because too many marketers make unsubscribing hard. If you’re doing things right by offering clear value and making it easy to opt out, then this new feature isn’t something to worry about. It’s a safeguard against bad practices, which is a win for the industry overall.

There isn’t a magic formula to game deliverability and engagement. Inbox providers reward patience, relevance, and accountability. As lifecycle marketers, those same principles define great customer relationships; deliverability is just where they show up most starkly.