Have you received an email like this?

Or noticed something like this in a newsletter?

This is likely because you haven’t opened and//or clicked an email from that Sender in a certain amount of time.

The Sender is likely attempting to “clean” their email list of unengaged subscribers. This is oftentimes done through “sunsetting”.

Sunsetting is a common email strategy to identify unengaged subscribers and stop sending them emails.

The goal is to improve engagement rates, deliverability and overall health of your email list and email marketing efforts.

Better engagement → better deliverability.

Sending emails to subscribers who do not engage with your content will impact your Sender Reputation and deliverability.

If deliverability is impacted, emails might not arrive in a reader’s inbox, or be considered Spam.

💌 Were you forwarded this email? Subscribe here! 💌

📧 I want my emails to be delivered. Where do I begin?

Start by figuring out how you’re going to define engagement.

You can define engagement in a variety of ways:

• Opens

• Clicks

• Conversions

• Replies

We are including Opens in the list because it is a commonly tracked metric, but we are not going to consider it for this exercise.

Why?

It’s not reliable due to Apple Mail Privacy Protection. Read more here.

For this example, let’s go with Clicks.

It’s a more reliable metric, and a good indicator that readers are engaging with your content.

📊 Explore your engagement numbers

Your definition of superfan, or most engaged subscriber, is unique to you.

You may define it as a subscriber that clicks at least once in each newsletter.

Or a subscriber that's clicked in the last 60 days.

It's up to you.

It also depends on the frequency of email sends, the types of emails you’re sending and more.

Here is an exercise to walk through

For this exercise: We send out a weekly newsletter with ample opportunity of links to click.

When outlining a Sunset Policy, your audience is least engaged subscribers.

We’re going with Clicks as our definition of engagement and how we’re deciding who is engaged and who isn’t.

We need to get a better understanding of our Click engagement, and would like it mapped over time.

Find the Click-Through Rate (# of Clicks / # of Emails Delivered) for subscribers that signed up:

• In the last 30 days

• Between 31-60 days

• 61-90 days

• 91-120 days

• 120+ days

For example, you might see a Click-Through Rate of:

• 10% in the last 30 days  

• 7% between 31-60 days.

• 1.5% between 61-90 days.

• 0.25% between 91-120 days

• 0.10% after 120 days

Looking at these buckets helps you identify various levels of engagement over time.

There’s a significant drop off in engagement after 60 days, with very little engagement after 120+ days.

We can also take a look at the “raw” number of clicks for each cohort. This helps us understand the degree of magnitude.

• If we continue sending to subscribers that are not very engaged: We risk deliverability issues.

• If we stop: We risk not receiving a dozen or more clicks.

Are the dozen of additional clicks you may receive from sending to a not very engaged audience worth risking the potential for your emails to land in Spam for emails sent to your engaged audience?

✏️ Draft up a plan

We now have a better understanding of the scale we’re operating on, and the tradeoffs of continuing to send to a not very engaged audience.

Let’s draft up a Sunset Policy, and use Clicks as our measure of engagement.

Continuing with our example, we know engagement drops significantly 60 days post signup.

Let’s try an audience of: Subscribers who signed up 60+ days ago and haven’t clicked.

We discover this accounts for ~25% of our email list.

If we were to put in place a Sunset Policy with the above findings, we could remove up to 25% of your email list.

Getting stakeholder buy in might be a little difficult.

Marketing, Sales, Editorial, Executives discover 25% of the email list is not engaged. The proposed Sunset Policy means removing up to 25% of the email list.

“Wait, you want to remove 25% of our readers from our email list?”

“This impacts our ability to sell ads in the newsletter,”

“How does this impact readership?”

Etc.

🔍 Start with a subset of subscribers, review the data and iterate if needed

Implementing a Sunset Policy can happen incrementally.

You identified your least engaged subscribers: They signed up 60+ days ago and haven’t clicked an email.

Maybe they’re just not “clickers”? They might be reading the emails, but not clicking?

They wouldn’t want to stop receiving emails from us, right?

There are some unknowns, so let’s conduct an experiment! Get some data to help us make informed decisions.

🔬 Conducting an experiment

When you're trying to see the impact of a new strategy, like a Sunset Policy, it can be helpful to conduct an experiment.

It’s important to maintain a Control during an experiment to be able to directly attribute any impact to metrics to the experiment.

A Control means status quo, or what your current workflow or strategy is. For this example, it would mean not having a Sunset Policy.

Rolling out a Sunset Policy to 100% of eligible subscribers without a Control makes it difficult to attribute impact on metrics. It might’ve been seasonality, other marketing efforts, etc.

You want to be able to test the status quo against a new variable and see its impact.

• Begin by identifying a cohort of subscribers for the test. For this example: Least engaged subscribers. Subscribers that signed up 60+ days ago and haven't clicked in an email.

• Select a percentage of eligible subscribers to receive the Control and a percentage to receive the Variant. This might be 80/20, 50/50, etc.

• Add subscribers to a static list in your ESP. One list to receive the Control, and another list to receive the Variant.

You’ll be able to see how subscribers who received the Control performed and how subscribers who received the Variant performed.

✅ We have an outline for testing.

Now… what is the Sunset Policy?

☀️ Sunset Policy examples

Below are examples featuring a callout from Morning Brew and a dedicated email from 1440.

The screenshots are for demonstration purposes, and are not reflective of the actual workflow for Morning Brew or 1440.

👀 Depending on stakeholders and business priorities, you might not actually remove subscribers from the email list

You might test out one of the above examples, but stop at removing subscribers from the email list.

When you roll out an experiment, you gather more data and when you gather more data you make more discoveries. And sometimes, stakeholders or business priorities get factored into the decision-making as well.

You may discover that the Variant did not yield a statistically significant result.

Or the Sunset Policy would increase Click-Through Rate, but Management doesn't want to decrease list size because Reach is important to advertisers.

Regardless, this process will result in learning more about your readers.

If a Sunset Policy is not feasible, start with removing email addresses that are unverified, or “bouncing”.

There are “hard bounces” and “soft bounces”. We won’t get too deep into the details for this post, but as an overview:

A soft bounce occurs when a user’s inbox is full, or a server is down.

It also might occur if a large number of users mark your email as Spam.

A hard bounce occurs when you send an email to an email address and it could not be delivered, it was rejected.

It could be because the email address does not exist, or isn’t valid.

For both types of bounces, figure out your Email Service Providers processes.

For example:

• If an email could not be delivered, is delivery reattempted? How many times?

• What happens to email addresses resulting from a soft bounce? A hard bounce? Are they added to a suppression list?

• Is there a report or somewhere in the dashboard to monitor this data?

Maintaining list hygiene is one of the many things to keep track of when you send emails.

You want to send emails that are going to be delivered, and hopefully engaged with.

This is also important to Internet Service Providers (ISP) like Gmail and Yahoo.

This factors into whether your emails are marked as Spam, and can contribute to your Sender Reputation.

If you’re not doing so already, removing email addresses that are not receiving your emails is a great introduction to cleaning your email list. You might also get a glimpse into the impacts of maintaining a healthy, engaged email list.

⚙️ Maintenance

Depending on your Email Service Provider (ESP), you may be able to build out a workflow to automate a Sunset Policy or list cleaning process.

Or you may have to generate a new list each time you want to remove unengaged subscribers or not valid email addresses from your email list.

Creating a recurring calendar event may be helpful.

🤔 But I’ll be sending emails to fewer recipients…

Having a Sunset Policy is up to you. It’s also not a one-size-fits-all policy.

It’s based on your readers and your goals.

Just like a real sunset, there are levels. A gradient and gradualness to it.

If you choose to continue to send to inactive or not engaged email addresses, this might:

• Reduce your engagement rates.

  • For example: Sending to 100,000 emails with a 2% Click Rate. With a Sunset Policy, you might send to 85,000 emails with a 10% Click Rate.

• Increase the likelihood of being marked as Spam.

• Impact your Sender Reputation.

These can result in a decreased chance of your emails making it to the inbox, and in front of your subscribers.

It’s hard to purge an email list. You want everyone to read and engage with your content.

If it’s helpful, try to identify the benefits of continuing to send to unengaged subscribers.

You may have a larger email list size which helps with advertisers, but… what else?

🧪 Testing and iterating

We walked through various approaches to a Sunset Policy and list cleaning.

Keep in mind these will be unique to you.

Every business, newsletter, audience is unique. We wanted to provide some thought starters.

… as always, it’s about testing and iterating.

❓ Question

Thank you to everyone who answered the question in the previous newsletter!

Y’all 👏 are 👏 testing 👏 your 👏 sign 👏 up 👏 flow! That is awesome!

🔑 Key takeaways

• It’s important to identify both your most and least engaged subscribers. This outlines the “edges”. The floor and ceiling you’re working within.

• Not having a Sunset Policy or list cleaning process may damage your deliverability. You continue to send emails to unengaged subscribers. This does not help your engagement or deliverability.

• Internet Service Providers (ISP) like Gmail and Yahoo don’t like this either. An ISP may begin to filter and send your emails to Spam. Unopened or deleted emails are viewed as not relevant content. This hurts your reputation as a sender.

• Bucket subscribers based on engagement and drill down on timeframes. This helps you see more of the subscriber journey over time.

• Hesitation with a Sunset Policy. Various stakeholders have different priorities with a newsletter or email marketing program. Try proposing a “toe-in-the-water” approach. Look at subscribers that haven’t clicked an email in the last six months. What does that cohort look like? Is it 5% of your email list? Then, test with that cohort. This helps improve Sender Reputation, but doesn't remove a large percentage of your email list.

• Again, if a Sunset Policy is not feasible, start with removing email addresses that are not valid. This means email addresses that are ”bouncing” or not being delivered to the inbox.

• Oftentimes, a Sunset Policy yields an increase in Click-Through Rate. One of the benefits of maintaining a healthy and engaged email list. This provides a positive indication to ISPs, and hopefully, internal stakeholders.

🎉 Have a great weekend!

• ⏩ Forward this email to a colleague or someone you know that has a newsletter or works in email marketing.

• 👋 Say hi on Twitter, LinkedIn.

• 💭 Reply to this email or send us an email with what you thought about today’s send. Feedback is welcome!