Category: Email Delivery

A/B Testing in Email Marketing

Every marketer dreams of high Open and Click through rate and does everything to achieve it. There exists many strategies to get better open and click through rates but testing your emails before starting the email campaign.

The best way to test your emails is running A/B tests.  A/B testing is the best choice to test you email marketing campaign and get the best version of your email.

In this article, we will show you what A/B testing is. Its types and how to run A/ tests on your campaign to get the desired results.

What Is Email A/B Testing?

A/B Testing is a method where two different versions of a website pages are sent out to users at the same time. After some period, the number of people who click on each version of the page is counted. Based on these results, the best performing version of the page is chosen and implemented across the site.

Web designers to determine which design performs better commonly use this technique. In email marketing, A/B testing is often used to compare the effectiveness of various email campaigns.

A/B testing is a great way to test different variables at once. You can use it to compare two things at once, or you can use it to compare one thing over time.

Types of A/B Testing

There are three types of A/B testing:

A) Split Testing

Split testing is where you have two versions of something (e.g., two websites), and you want to know which version performs better. To do this, you randomly assign users to either group 1 or group 2. Then, you track how many people click each link. If one version gets more clicks than the other does, then you know which version performed better.

B) Time-Based A/B Testing

Time-based A/B testing is similar to split testing except that instead of having two groups, you only have one. In this case, you want to find out if one version of something performs better over time. To do this, again, you randomly assign users either to group 1 or group 2, but this time, you do not stop tracking them after they click the first link. Instead, you keep track of their behavior until you reach some predetermined end. At that point, you calculate whether the version-performing better was the original one or the one you switched to.

C) Multivariate Testing

Multivariate testing is a combination of both split testing and time-based testing. You start with a single variable (e.g., website design). Then, you randomly assign users into one of two groups. Next, you measure how well each group does on that particular variable. Once you have measured performance, you switch the assignment method and repeat the same steps. Keep doing this until you have tested all possible combinations of variables.

Things to Note before A/B Testing

• You should always test your email before starting the campaign.
• Always use the same strain and grow method.
• Use the same amount of emails.
• Do not overfeed.
• Test at least three times.
• Make sure you have enough time to wait for results.
• Be patient!
• Do not forget to document everything.
• If you do not know what to do, ask someone who does.
• Have fun!

How to Set Up Your A/B Test Email Campaign for Accurate Results

1. Decide Which Variable to Test

The first step is deciding what variables you want to test. You need to know what you are testing before you start collecting data. If you do not have any idea about how to choose the right variable, then you should ask yourself some questions. What is the purpose of your experiment? Why are you doing it? What is the goal of your experiment? Do you want to find out if the price of your product affects its sales? On the other hand, maybe you want to find out whether the quality of your product affects its popularity. Whatever the case may be, you need to make sure that you understand the purpose of your experiment and the question you want to answer.

2. Pick the Right Sample Size

Once you have decided what you want to test, you need to figure out how many samples you need to collect. There are two ways to go about picking the right sample size. One way is to use a statistical formula called the “rule of thumb”. Another method is to use a simple rule of thumb. The rule of thumb says that the number of samples you need to collect is equal to the square root of the number of factors you are testing. Therefore, if you were testing three different variables, then you would need nine samples. However, this method is not always accurate. That is why we recommend using the statistical formula instead.

3. Choose Your Timing Window

Timing windows are the timespan over which you collect data. You need to consider the following things when choosing your timing window. First, you need to know how long you want to keep collecting data after you begin collecting it. Second, you need to think about how much data you need to collect. Third, you need to think of how often you want to collect data. Fourth, you need to think if you want to collect data at regular intervals or just once. Finally, you need to think how long you want to wait between each collection period.

4. Determine Delivery Time

Delivery time refers to the amount of time it takes for your message to reach its destination. You need to determine how long you want to send your messages to your customers. In addition, you need to think whether you want to deliver them immediately or later.

5. Consider Only One Variable at a Time

You need to decide whether you want to test one variable at a time or all of them together. Testing one variable at a time is easier than testing all of them together. However, if you test one variable at a given time, you will not get a complete picture of how they affect each other.

Email Components You Can Test

An email is a great way to communicate with customers and prospects. However, not all emails are created equal. There are many different types of emails that you can send out to your audience. Each type of email requires a specific set of components to make sure it reaches its intended audience.

There are many different email components that you can test to determine how they perform. These components include subject lines, preview text, and body text. Each of these components have their own purpose and are tested differently.

Subject Line Testing

The first step in testing any email component is to make sure that it performs well. To do this, you need to know what type of content you want to send out. If you are sending emails about products or services, then you should use the product/service title as the subject line. If you’d rather not mention the product or service at all, then you can still use the title as the subject line and add a call-to-action (CTA) to encourage people to click through.

Preview Text Testing

Next, you’ll want to test the preview text. This text appears before the recipient opens the email. When testing preview text, you want to ensure that it does not look Spammy or pushy. In addition, you want to make sure that it is not too long or short. Longer previews tend to get cut off, while shorter ones don’t give enough information.

Body Text Testing

Finally, you’ll need to test the body text. This is the actual message that you are sending out. Make sure that it is clear and concise. Also, try to avoid using too much jargon or industry terms. Instead, stick to simple language that anyone can understand.

Testing length automation

Automating the length of your email is a huge benefit to your business. When you automate the length of your emails, you can save money and increase conversions. By automating the length of your emails you can test different lengths of content and determine what works best for your business.

Timing

Timing is another factor that you need to consider when sending out emails. If you send out emails at the wrong times, they may get lost in the shuffle. Sending out emails at the right times can help you reach your target audience. You can test the timing by sending out emails are different time using A/B test method.

Buttons vs Text

Buttons are great for converting visitors into buyers. But if you have a lot of buttons, it can be overwhelming for users. On the other hand, text is easier to read and understand than buttons. So, if you want to convert visitors into buyers, use text over buttons. You should test this feature before starting the email campaign.

Long copy vs short copy

Long copy is great for getting attention. However, long copy can be difficult to read. Short copy is easier to read and makes it easier for people to digest information. It depends on the type of email you are sending, which type of email copy you need. You can test short copy and long copy using A/B testing to make sure which will work best.

Images

Images are great for grabbing attention. But, if you don’t know how to use them, they can be distracting. Make sure to choose the right image size and placement. Using A/B testing you can try out different images or no images in different emails to check which email performs better.

Call-to-Action

Call-to-actions are the final step before someone converts into a buyer. Your call-to-action should be clear and concise. Also, make sure to add a link to your website or landing page. You can try different call to actions in your email and try them out. You will get a good idea about the call to action you need to put in your email campaign.

Email Testing Softwares

Spam Filtering

Spam filtering is a critical component of any email system. Without spam filters, we would have no way of knowing if our emails were being sent to the right people or not. There are many different types of spam filters out there, including those that are free and paid options. Paid options tend to be more effective at stopping spam than their free counterparts, but they do cost money. In order to test your current spam filter, you need to know what type of spam you receive. If you don’t know what kind of spam you get, then you won’t be able to effectively test your spam filter.

Email Client

An email client is a piece of software that helps you manage your email messages. These clients range from simple text editors to full featured applications. To test your current email client, you’ll want to send yourself an email message. Then, open the email in your client and read it. Does the email look correct? Do you notice any formatting issues? Is it easy to navigate?

Tips for Running More Effective Email Tests

Choose the right test

A/B testing is a great way to find out what works best for your audience. However, if you don’t choose the right test, then you may end up wasting time and money. So before you start any A/B testing campaign, make sure you have chosen the right test. Here’s how to do it:

• Make sure the content is relevant to your target audience. If you are trying to sell a product to people who live in California, then you should probably not run an A/B test comparing two different products that only appeal to people in New York.

• Make sure the content is not biased. You want to avoid having one version of the content be positive about your brand while the other version is negative.

• Make sure the content does not violate Google’s guidelines. If you’re using AdWords, then you need to make sure that the keywords you use aren’t banned.

• Make sure the content fits your budget. If you’re doing A/B testing for free, then you’ll want to keep your budgets low. But if you’re paying for ads, then you’ll want them to be targeted at the right audiences.

• Make sure the content has enough variation. If you’re going to compare two versions of the same piece of content, then you need to ensure that they vary significantly enough to be considered separate pieces of content.

• Make sure the content gets seen. If you’re going after a specific keyword, then make sure that you’re targeting that keyword well. And if you’re going after a broad term, make sure you’re reaching out to people searching for that term.

• Make sure the content converts. If you’re going for conversions, then make sure that the content is converting well.

• Make sure the content performs well. If you’re going with a conversion goal, then make sure that your content is performing well enough to convert.

• Make sure the content works across devices. If you are going mobile-first, then make sure that both desktop and mobile versions of the content perform equally well.

Conclusion

Email marketing cannot be successful without testing. You just cannot ignore the importance of A/B email testing. A/B email testing helps you to get the best version of your email.

A/B testing is neither mysterious nor difficult, you just need to understand the method. We have mentioned all the important steps to do A/B testing more effectively. We hope you will implement all these steps to test your email marketing campaign and tell us about the outcome. We will be waiting to hear from you. You can comment below or send your feedback at info@massmailservers.net.

Message

HTML Source Code

Remember, that technically incorrect emails are often identified as spam. HTML source code is another important moment. Never use applications that create awful HTML code. If your message has too complicated structure, or contains empty or invalid tags, email servers may regard such emails as spam. Try not to overload your message with illegible fonts, bright backgrounds, or other elements that will distract the readers’ attention.

Layout

Make the layout of your email as simple as possible. Keep in mind that your newsletters should be informative. The information you share has more value than the outlook of your message. The outlook should be simple and function. So, try to avoid complex formatting.

Pictures

You will considerably increase the deliverability, if about 80% of your message is text.  Try to substitute picture buttons with HTML. Don’t include large images. The same goes about active components such as JavaScript and ActiveX. The less graphic you employ the better will be the results.

Message Personalization

It’s a good tone to personalize your messages. Avoid generic salutation such as “Dear User” or “Dear Customer”. With Bulk Mailer email personalization becomes an automated process. With Bulk Mailer email personalization becomes an automated process which allows sending unique messages to every recipient. If your mailing list contains various data, such as first name, last name, company name, tile, country, etc, you can add all these fields into your message.

Subject line

Never mislead your recipients by writing deceptive subject lines. The subject line should reflect the content of the message. It’s a bad idea to include a note that your message is not spam.

Message content

It is advisable to use normal conversational language. There’s no need to use redundant spacing, capitalization, or uncommon spelling or characters.

HTML vs Plain text

There’s no need to give preference either to messages created in HTML-editors or plain text. Instead, it is advisable to include a plain text version in addition to your HTML email. Plain text version is necessary mainly for anti-spam filters rather than for your recipients. It will give you a guarantee that your messages won’t be regarded as spam. Bulk Mailer creates plain text version by default, no matter whether you’ve created the message in Bulk Mailer’s editor or imported the template from the file.

Sending Process

Disable Scanning of Outgoing Emails

To improve Bulk Mailer’s performance the first thing to do is to disable checking outgoing emails for viruses. In fact, you should scan the message for viruses only once. If you scan every outgoing message, it will not only slow down the process, but also prevent the majority of your messages from being delivered.

Sending Mode

Bulk Mailer works via direct and relay sending modes.

Direct sending mode is the default mode. It works as a built-in SMTP server. It is of utmost importance to have access to 25 port. If it is blocked by your ISP, you won’t be able to send messages via direct sending mode, only via relay mode.

Relay sending mode uses your ISP’s SMTP (or any other you specify). Every ISP has its own limits on the number of messages sent. Please, contact your ISP to find out if there is such limit. If the limit is 200 messages per day, the program will fail to send 5000 messages via relay sending mode.

Below are a few suggestions on how to best ensure your messages are delivered and your IP does not get blacklisted.Send email only to those that want it. Spammers write to many people who don’t want their mail and anti-spam filters are designed to identify that behavior. To avoid being perceived as a spammer, use an opt-in method of subscription for your mailing list. Better yet, utilize a confirmed opt-in process where subscribers actively verify their intent to receive your mailings by clicking on a confirmation email before being added to your list.

Use email authentication such as DKIM. This will help identify that the email is legitimately from you and, if you sign all your email, it will help identify forgeries, too. In addition, using dedicated domains for different mail streams (e.g., transactional messages vs. marketing emails) is also a recommended practice. DKIM.org is a good resource for information about DKIM.

Consistently manage your lists by paying attention to hard and soft bounces as well as inactive recipients. Persistent emails to these addresses are a surefire way to get your connections deferred. If your messages are being blocked, look closely at any SMTP Reply Codes the mail servers are returning to you and be sure you’re addressing the problem.

• Don’t retry 5xx messages. If a mailserver rejects a message with a 5xx error, they will not accept it anytime soon. Retrying the message wastes both resources and makes you appear to have a dirty list.
• Do retry 4xx messages. If a mailserver sends a 4xx error, this is a temporary error and you shuold try sending the message at a later time.
• Do refresh your list periodically. Consider removing or sending a reconfirmation email to inactive subscribers, i.e., users who have not opened or clicked on your emails for a period of time. Sending your mail to users who are not reading them, or who may even mark them as “spam,” will almost certainly hurt your delivery metrics and reputation.

Respect the user’s mailbox

• If a customer entrusts an email address to you during a transaction or for a particular newsletter, they do not expect to receive unrelated messages, such as extraneous marketing emails, in the process. Adding email addresses to other lists without their explicit opt-in is a guaranteed way to lose a customer’s trust.
• Honor the frequency of the list’s intent. If customers believe they are signing up for a monthly newsletter but they start receiving messages on a weekly basis, such a practice will likely prompt users to label your messages as spam.

Use a consistent “From:” header address. Your domain name is an important element of your brand. Using it consistently helps your recipients to distinguish your email from spam. Additionally, using a static From: address helps users who have set up filters to route messages to a specific folder.

Pay attention to your email’s content

• Test your email’s look and feel with image placeholders. Many users won’t see images in your email by default.
• Link to domains, not IP addresses. Many mail clients warn users that IP address links are dangerous.
• Use standard ports. Many main clients  warn users that links containing ports are dangerous.
• Don’t include HTML forms in emails. Many main clients warn users that submitting forms in email can be dangerous.
• Don’t include Javascript in emails. Usually javascript is stripped out and will not run.
• Don’t include embedded objects in emails (like flash or ActiveX). The objects usually are stripped and will not run.

Honor unsubscribe requests as fast as you honor subscribe requests. When a user unsubscribes, they don’t want to receive that mail anymore. Promptly removing them from the list should help prevent users from marking your messages as spam in the future.

Be CAN-SPAM compliant. Regardless of where in the world you’re sending your mail, make sure that you adhere to the requirements stipulated by the CAN-SPAM Act.

Publish reverse DNS (PTR) records for your sending IPs. If there is no reverse DNS entry for your IP address, or if it looks like a dynamically-assigned IP instead of a static mail server, spam filters are more likely to downgrade its sending reputation. you can setup a reverse DNS in the control panel. This is automatically done for shared hosting clients.

Secure your mail servers. If you are running your own mailserver ensure they are not open to abuse. If your servers act as “open proxies” or “relays,” spammers may attempt to send their own mail from your systems. Keep your software up to date with the latest security patches, and always filter user-generated content before sending it out, to help prevent spammers from using your resources and tarnishing your reputation.

Use common-sense settings. While we have not published guidelines for numbers of connections you can concurrently use, we ask that you treat our resources with respect. The more you take, the fewer there are for others, which may force us to defer your connections.

Introduction
Email marketing, as many of us know, can be a powerful, inexpensive method of reaching our most active potential or existing customers. It can boost not only our direct sales, but also our credibility and referrals.

One of the major benefits of email marketing is that email is free, but obviously this is the same reason why spam has become so popular and so frustrating. With spam comes spam filters and with spam filters comes the blocking of legitimate email.

In this article I’ll try and describe the basic steps that can help reduce the number of emails you send out that get blocked by spam filters — hopefully resulting in a more rewarding marketing effort.

The right selection of words
Many spam filters work by analyzing the email based on its content and the words used. Many words — such as free, sex and so forth — are very heavy spam trigger keywords. Your priority should be to avoid such words while keeping your newsletter as professional as possible.

Later in this article I will show you a technique that I use to help me detect words that could trigger spam filters that I may have missed.

Email content and design

• Don’t include too many images in your email (1-2 per email should suffice) or some filters will flag it
• Don’t use images or send attachments that are too large. 300Kb should be the absolute max size of the email so be sure to make sure the total for images and attachments comes in well under that (adding a file to an email will increase the size by about 30% just in encoding it) so you need to take that into consideration too
• Don’t include a single large image in your email if you can avoid it since this is how most spam nowadays seems to operate

Pay attention to your formatting
When formatting your email, keep it simple and professional. Excessive use of different colors, fonts, sizes, images and so forth will result in a higher spam filtering rate. Keep your email as clean as possible, and try to stick to a maximum of 2 or 3 different font types and sizes. Overly large sized fonts will surely add to an email being flagged as spam, as will too many images (or not enough text).

Try and use a short and simple stylesheet rather than using font tags excessively. Most spam filters don’t appreciate a multitude of font tags and inline formatting, and the more primitive filters can’t detect stylesheets so they will not penalize as easily.

Consistency is king
Use a template if you plan on sending newsletters consistently. This will make sure that all your newsletters look and feel the same. It will also add a touch of professionalism and branding to your newsletters.

Whilst not directly affecting spam filters, this will enable your readers to distinguish your newsletter instantly, thus not reporting it as spam accidentally. Some spam filters work by querying a spam server, whereas others report individual emails as spam. If your email gets reported as spam, then more than likely multiple spam filters will flag your email.

Being consistent with your timing of the newsletter also helps. For example, if you send a newsletter once per month (I personally don’t recommend you send out any more than this, unless you’ve got something really interesting to say), then aim to send it out at the same time, on the same day each month.

Once again, your potential readers will learn to expect your email, adding professionalism and often improving open rates, also reducing accidental spam flagging as well.Keep your lists as clean as possible

Honor all unsubscribe requests (your email marketing software should generate an unsubscribe link for all emails you send) and process bounced emails frequently. Sending to email addresses that have bounced repeatedly can result in a blacklisted IP address. You want to make sure that each bulk email you send does go out to legitimate, working email addresses.

You should also make sure your abuse@ and postmaster@ emails are valid and working. The Internet Engineering Task Force (IETF) has recommended these email addresses for complaint spam reporting and you may receive emails from users or ISP’s if they have a complaint or spam report about your mailings.

Unsubscribe and Contact Information
Every newsletter you send out should contain a way for the reader to unsubscribe. Not doing so is illegal in some countries and is an instant sign of spamming. You should also display your contact information (Phone, Fax and Address) clearly, as this greatly increases confidence in your email and your company, as well as conforms to spam laws in the United States. Contact information also allows a potential customer to contact you if need be.

Test, Test, Test
The key to avoiding spam filters is testing. The first method of testing I use is to send the newsletter to multiple email accounts with existing spam filters. For example, I have a Gmail (http://www.gmail.com) account and a Hotmail (http://www.hotmail.com) account that I make sure I send my newsletter to. If the newsletter ends up in the junk folder, then I’ve got some work to do.

I also have a couple of email accounts with different web hosts that have spam filters in place. In particular, they mostly use spam assassin — a popular piece of spam filtering software. Spam assassin is useful because every email that it flags as spam is given a report and a list of why that email was considered spam.

I also use a powerful spam filtering tool, which detects any spam related “triggers” based on Spam Assassin’s scoring system. The scoring system is constantly updated so you will always stay on top of the latest spam filter changes. By using the spam assassin checking system — gives me feedback as to why my email may have been flagged. If I’ve used words or formatting that I shouldn’t have, or if I’ve included too many images, etc.

More Ways to Avoid the Spam Filters

• Watch your attachments
  Many people don’t realize that the type of attachment you send with your email can cause different spam filters to block your email and even go so far as reporting your I.P to a black list database.You should avoid using script or any type of attachment besides PDF. Many corporate mailboxes as well as virus filters block attachments that end in .exe, .avi, .swf, .zip, etc.
• Use double opt-in lists
  I’ve mentioned this in my previous article but once again to make your mailing list as clean as possible always use double opt-in strategies. That is, when someone signs up they should receive an email which contains a link they must click to verify that they do indeed want to be on your mailing list. This stops illegitimate email addresses from being added to your mailing list.

Conclusion
Although there’s no fail-safe way to absolutely guarantee all of the emails you send will reach the intended recipients, simply by applying the techniques described in this article can dramatically help to improve your email deliverability and hopefully your click thru rate and bottom line.

In conclusion, I hope that you’ve found the above tips useful and I wish you luck with your email marketing endeavors!