Email Newsletter &
Email Marketing Tips
- Bounce Basics Part 2: The Reckoning
- DMARC Changes
- So Fresh and So Clean (The Email Version)
- How to Test Sending Times
- Bounce Basics
- The Email Frequency Fallacy
- Below the Fold: More Important Than You Might Imagine
- Reasonable Automation
- Now verified: Gmail image caching affects MNB opens (but barely)
- Making the Scene
Email Marketing Tips
So, after our last highly informative blog post, you’ve got the skinny on email bounces. But, what do you do with all of that sweet info? Let’s take a look.
First: Watch Your Numbers
Bounce rates vary from business to business, but industry experts suggest that a successful campaign should return less than 5% of emails bounced, with closer to 1% being ideal. Now, that may seem like an awfully low number, but it’s actually quite reasonable (and very attainable) for an effectively run campaign.
If you’re seeing a bounce rate much higher than 5%, there are a number of easily rectifiable issues that could be contributing. Let’s go through each, and talk about how to fix them.
No list maintenance. An unmaintained list could include several derelict email addresses with full inboxes. Log your soft bounces for 3-5 consecutive sends. If you’re seeing the same addresses appear in each report, it might be time to give ‘em the boot.
Additionally, watch your reports, keep track of which subscribers are active, and then start eliminating those who’ve shown no response over your past five sends. If they’re not opening, they may not be worth the subscriber space.
Unconfirmed subscribers. A list full of unconfirmed subscribers can affect sending reputation, forcing some ESPs to bounce your incoming messages. Make sure your new subscribers are completing a confirmation process, and issue a Confirmation Message to any existing unconfirmed subscribers. Those who don’t respond might be worth leaving behind.
Content problems. Senders too often ignore the impact content has on campaign efficacy. If you’re sending newsletters with content that offends SPAM filters, links to pages with poor reputations, or simply has no value, you’ll likely see an increase in bounces. Junk content often yields an increase in SPAM complaints, which can lead to more bounces in the future. Focus on clean, simple, effective content, and you’re much more likely to make it into the inbox.
Dangerous subject lines. Subject lines that are irrelevant to the content or touch on any of those dreaded SPAM filter trigger words can force bounces more often than you might imagine. Don’t get too cute. Keep those subject lines simple and safe.
Second: Leave No Sub Behind
So, now you’ve got a few tips to help you bring down that bounce rate, but what about those subscribers returning bounces that still have some value to your campaign?
Well, on some level, this requires some individual evaluation on your part. As we mentioned above, repeat bounce-offenders would likely benefit from the old Dear John letter, but that may not always be the case. If you’re at all concerned your content may be the problem, try making your adjustments before making your cuts.
If, after several consecutive sends of your leaner, meaner newsletters, those same email addresses are still bouncing back, you might want to consider issuing a special newsletter to those subscribers to try and gauge their interest.
To do so, use the Bounce Details tool on your Bounce Report page (found by clicking the bounce number for any sent newsletter on the Reports page), and then click Export Soft. This will allow you to download a list of all your soft bounces from the newsletter you selected. Next, re-upload it to a new subscriber category.
With your soft bounces list in place. Create a quick newsletters with an eye-catching headline, like “Where did you go?” or “Do you still want to hear from us?” Use the body of the email to ask the subscriber if they’re still interested in receiving your newsletters, and then include a link to confirm their subscription. You can even add a quick teaser for some of the (very valuable) content you’ll be putting in future newsletters.
This ultra slim format should help reach a few more inboxes, hopefully giving those subscribers a chance to see the message, respond, and re-engage.
You can also use this process to reach out to those subscribers who aren’t opening your newsletters. This will help weed out any disinterested parties, increasing your open rate, and likely limiting future bounces.
Finally: Stay Active!
The steps above may sound like a lot of work, but after you’ve gone through the process once or twice, you’ll see how easy it actually is. However, the results won’t be permanent. As your list grows and as time passes, you’ll need to repeat these steps with some regularity. Even if you’re bounce rates are standing pat in that sweet 1-2% range, running some quarterly maintenance will always help build your reputation, and make your marketing more effective.
As of April 23, 2014
AOL’s DMARC policy has just followed-suit with YAHOO and will now also be rejecting emails sent from third party platforms with sender’s email addresses of their own domain being sent from NOT their own domain. Its a bit much to digest and understand but please do know, don’t use an AOL or YAHOO email address in your Sender’s Email Address inside MNB. We are working on a viable solution to this that does NOT reproduce what happened on or near the 8th.
We expect other providers will be doing the same thing in the near-ish future. As we hear about them or find out about them, we’ll be posting here.
As of April 8, 2014
In the very recent past [few days], email service providers have been implementing stricter policy on sending/receiving emails (namely free service providers like GMAIL and YAHOO). In your MNB account, if your “Sent From” email address is a Yahoo or GMAIL address, you very well may be seeing an inordinately large amount of bounces. If you look into the details of these bounces, you will notice something related to “DMARC policy”.
DMARC isn’t a brand new authentication standard; what is new is the live implementation of it as a means to blocking emails. For the most part, this change was completely unannounced by any of the service providers using the new standards - we got no “heads up” (nor did any of our competitors).
In response, MNB is increasing our DMARC compliance (starting today and likely finishing over the next 72 hours as DNS replicates). Part of this compliance requires the sending domain to match the domain in the sent from address. MNB users will no longer be able to provide a “Sent From” email address as MNB will now control that (which will now be something like “email@example.com”). The “Sent From” name and “Reply To” information (name and email address) will be unchanged.
Gosh, you’re thinking, didn’t you just write a spring cleaning blog post last year? Why yes, yes we did and we’ll probably write one next year because guess what—having a clean email list is ALWAYS relevant! Spring cleaning has two real elements: one is the literal removal of dirt and the other is the infusion of the indoors with fresh air from outside. Both of these apply to email marketing as readily as they do to one’s home.
The Literal Cleaning
Keeping a clean email list starts with bounces but it doesn’t stop there. Successful delivery into an inbox is a great first step. The next step is engagement. Your open rate will give you a good idea of what level of engagement you are receiving. Average open rates vary depending on industry; a 15-25% open rate is within the normal for most.
Here’s my proposal: a smaller interactive subscriber list is more beneficial than a large subscriber list with a smaller percentage of engaged readers. Annually clearing out the email addresses of subscribers who consistently fail to engage with your email list will help you to cull out those who are uninterested in your content or who are not longer using an email address but keep it activated, etc. As a general practice, pulling a list of non-openers who haven’t opened your newsletter(s) in six months or more runs a low risk of accidentally booting valuable contacts from your list. Learn more about exporting this data on MNB. If you’d rather have us do it for you, that’s called “Subscriber Services,” and it’s a great list hygiene option for those who’d like to get their spring cleaning done pronto.
Sometimes people feel hesitant about simply dumping subscribers who have ceased to be interactive. In those cases it’s no skin off anyone’s teeth to give a warning shot across the bow informing those subscribers that you’ve missed them but you’re going to stop sending if they don’t renew their subscription or opt back in. It’s probably obvious, but this is a good time to mention your value proposition.
Freshening It Up
We can’t really talk about open rates and non-openers without at least touching on the topic of quality and content. Around the same time that you’re cleaning your list (or running a win-them-back campaign to see if you can re-interest readers who have stopped engaging) you might want to look at your content and assess where you are. Are you still delivering the content your readers signed up for? Are your interests still well-represented by that content? If no, it might be time to throw open the windows and freshen things up. Getting rid of stale content and adding fresh new features might actually be enough to bring fatigued readers back on board and helps to keep active readers interested.
Testing the best sending times for your newsletter is a valuable way to learn more about the behaviors of your readers and the best times to reach them. We’ve already talked about techniques to utilize in creating a simple case study and those rules all apply here. In order to test sending times, try to keep everything else as constant as you can; don’t try sending new types of content at this time and keep your subject lines and calls to action consistent with your usual email marketing practices.
Step One: Examine Potential Windows
Consider your demographics. Who is opening your emails and where? This information is vital. If you are primarily sending to professionals engaging with your emails as a course of business, then you want to identify times during their work week when they will be apt to open, read, and respond to calls to action in emails. On the other hand, if you are sending emails that are being opened by individuals in their leisure time, your tactic will need to be different. Keep in mind that most emails are opened during the day on weekdays, although this is certainly subject to variability based on industry and unique subscriber lists. Unless you have a strong reason to suspect that yours is an exception, stick to the more common email-opening time slots.
Step Two: Design and Execute the Test
If you’re a science geek like me, this part is quite exciting. Now, you don’t have to go brush up on the scientific method, just decide on a basic approach. Stick to one of two options: run an A/B test or design a longer-term test in which you send to the entire group.
Let’s back up. First pick a few (not more than four) times during the day/week that you’d like to test. Then determine your test method.
For an A/B Test you’ll need to divide your readership into two groups. Make sure that you are sending to an active readership when doing this. (You want to know that you have active engaged subscribers on both the A and the B side of the testing). Send to the A group at one of your test times and the B group at the other. Continue with each of the times you are testing for.
NOTE: if you pick more than two times, you will be doing an A/B/C or A/B/C/D test.
For a trial that uses your entire readership, simply lengthen the time span and schedule sends for each of your test times while sticking to your normal sending schedule. Sending more emails than usual could skew your data. To check for consistency, send a few times at each of the times you have selected.
Step Three: Adjust As Necessary
If you experienced clear and measurable results, then adjust course and continue to actively observe your metrics to ensure that you continue to see the results you expect. If you didn’t see a difference, then you may choose to test at more unconventional times. Of course, this might also be a sign it’s time to give closer examination to other aspects of your approach such as subject line and call to action.
Other potential factors that will require you to adjust course will be unexpected or confusing results. Say you start sending at 3pm in the afternoons on Thursday and see a fast increase in the number of opens, but the number of click-throughs decrease and fewer readers respond to the call-to-action. This lets you know that you’ve found a viewing time, but you’ve yet to hit on your subscribers’ action time. All isn’t lost here - you know exactly when to send if you want to prime readers for a future call to action… you just have to find out what time to send that email. All information that you gain is useful once you see how it applies to the bigger picture.
P.S. If you want to get fancy….
Once you’ve conducted such a test, you have a wealth of information at your fingertips that, if you so choose, can be used to further research and segment subscriber behavior. For example, if you collect personal data such as gender, you could examine the behaviors of your male and female readers to determine if there were any significant differences in open rates between the two during the times you tested.
Here at MNB, we get a lot of questions about bounces and bounce rates. In an effort to clear things up, we thought we’d take a quick look at some of the basic information you need to know about bounces, and explore how they affect your email marketing campaign. We’ve got a lot to cover, so let’s get right to it.
What is a Bounce?
A bounce (or bounce message) is a non-delivery report issued by a recipient email server after a send (usually) from an email service provider. In short, it means that your email newsletter didn’t reach its destination. Bounces appear in two different forms: hard and soft, but we’ll get back to that in just a minute. First let’s take a look at why a bounce occurs.
Why Does an Email Bounce?
Usually, a bounce is the result of something like a full inbox, incorrect domain, or deleted address. These issues force a non-delivery and result in a bounce report returned to the sending client.
However, bounces can also stem from recipient host restrictions.
Every email domain server host (like gmail.com or mynewsletterbuilder.com) has an established set of parameters that email service providers must meet in order to successfully deliver a message to a recipient inbox. Those parameters include very specific caps for sender statistics like reputation, spam reports, bounce history and even recent sending history.
If a recipient server’s particular maxims are exceeded in any of those categories, the domain may determine the message undeliverable, and issue a bounce. Unfortunately, those specific guidelines are unique to each domain (although, they’re usually fairly similar to industry norms). And, what’s worse is that they’re not always consistent throughout the year. Many host servers will adjust their protocols during those parts of the year when inbound email is higher than average.
Finally, host domains don’t make their guidelines public as that information could help spammers penetrate their security protocols. Add all that up, and it’s easy to see why determining an exact formula for how to avoid bounces isn’t easy – for senders, or for sending clients (like us).
Once a host domain determines that an incoming email will be reported as a non-delivery, it is required (adherent to Internet law) to attach a six digit code to the report, identifying it as a hard (code starts with a 5) or soft bounce (code starts with a 4). Hard and soft bounce designations are made available to the sender, offering a little insight into why the email wasn’t delivered.
So, what separates a hard bounce from a soft bounce?
Well, a hard bounce is issued in the event of a permanent problem with deliverability. This can include deleted addresses, incorrect domains or a domain established block. MyNewsletterBuilder reads into every hard bounce to determine whether or not it’s legitimate, and if it’s really permanent. If the system validates the permanent bounce, it will remove the address from your subscriber list automatically. We don’t rely on a simple 5xx code to determine a hard bounce as some domain admins report codes improperly. This helps you conserve valuable list space, rather than wasting it on dead addresses.
A soft bounce, on the other hand, is the result of a temporary issue with deliverability. That can include a domain service interruption, a full inbox or a host identified problem with content or issuer’s sending reputation. Soft bounces can occasionally overlap with spam filter redirects, but not always (it usually depends on server protocol).
After receiving an initial soft bounce report, MyNewsletterBuilder’s servers will repeat the send attempt over a period of 24 hours to try and complete delivery. If, after the sending period is over, the recipient host is still issuing soft bounce reports, MNB will record the send as a soft bounce in your report data.
We’re Not Done Yet
So, what do you do with bounced emails? What should your bounce report look like? How can you reduce the number of bounces you’re reporting in the future?
We’ll get to all those questions and more in two weeks when we return with part 2 of Bounce Basics blog breport – err, report.