Last Updated: 02 Nov 2023
|
Differences
This shows you the differences between two versions of the page.
Both sides previous revision Previous revision | Next revision Both sides next revision | ||
marketing-tech:html-email:best-practices [Apr 22, 2010 04:43 AM] dordal |
marketing-tech:html-email:best-practices [Nov 21, 2020 10:39 PM] 127.0.0.1 external edit |
||
---|---|---|---|
Line 1: | Line 1: | ||
+ | = Email Marketing Best Practices - April 2010 = | ||
+ | This is a list of best practices for doing HTML email marketing. It is focused mainly on overall strategy; there are separate articles on [[content-best-practices|email content best practices]] and [[design-best-practices|email design/ | ||
+ | |||
+ | This information is current as of April 2010. | ||
+ | == General Tips == | ||
+ | |||
+ | * **Build A Template.** Rule #1: Build an email template for your organization. This may seem obvious, but I can't tell you how many people don't do it. In terms of effort, doing an HTML email is a bit like designing a mini website: you have to go through a few rounds of creative, build out some messaging, program everything, and then test it in a million email clients. If you're planning to send more than one email (and who isn' | ||
+ | |||
+ | * **Use an Email Service Provider (ESP).** You may be interested in creating your own mailing software and/or setting up a package on your server, such as [[http:// | ||
+ | |||
+ | * **Worry about Deliverability.** One of the big reasons to use an ESP is that they handle ' | ||
+ | |||
+ | * **Test Extensively.** Email development is a mess. Now that IE6 is (almost) gone, standards compliant web development is starting to come into its own, but email development is still a decade behind. The world actually took a step // | ||
+ | |||
+ | * **Start thinking about mobile.** Mobile email is becoming increasingly prevalent; as of March 2010 the numbers I've seen thrown around say about 10% of email in the USA is checked on a smartphone type device (iPhone, Android, etc.). I'm still looking for hard data, but one way or another it is clear that you should consider mobile in your email strategy. | ||
+ | == A/B Testing == | ||
+ | |||
+ | One of the great things about email, and the Internet in general, is that you can test extensively. With email, A/B testing is easy, particularly if you have a large list. | ||
+ | |||
+ | * **The key to proper A/B testing is controlling variables.** You can't use Design A on your newsletter today, and Design B next week, and expect a valid A/B test. If Design B did better, was it because it is truly better? Or is it because the contents in the second week were more interesting? | ||
+ | |||
+ | * If you're sending emails to a large volume of people (>100K), **you should consider doing an pre-test**. Make two versions of the email, send one to the first 5K people and the other to the second 5K, and track how both do. Send the one that does better to the other 90K. Obviously, you can get more sophisticated than this if you want. | ||
+ | |||
+ | * If you don't have a large list of people, **you can perform a test over multiple sends**. For example, I once tested whether prefixing the subject line with 'Hey, NAME!' would increase open rates. My list was less than 5,000 people, and I was sending once a week. Each week, I'd randomly remix the list, and send half with the prefix, and half without. It took several weeks, but I was eventually able to determine that the prefix increased open rates by about 18%. | ||
+ | |||
+ | * **Try to establish rules & best practices that you can always follow.** The 'Hey, NAME!' I mentioned above is an example of that: now that I know it is better, I'll always use it. Establishing rules like this for your list can be hard, but if it increases your open rates or click rates it can be very beneficial. | ||
+ | == Types of Email == | ||
+ | |||
+ | There are several different //types// of email you can send. | ||
+ | |||
+ | * **Newsletter/ | ||
+ | |||
+ | * **Transactional Emails**: If your customers do any kind of transactions with you, the transactional emails (e.g. ' | ||
+ | |||
+ | * **Triggered Emails**: One of today' | ||
+ | * If a customer abandons his shopping cart without checking out, you might send an email an hour later inviting him to complete his purchase. | ||
+ | * If you notice a customer hasn't logged into a site in a while, you might send him an invite to come back. | ||
+ | * If a customer signs up for your service but isn't using it, you might email him to check how they like it or if they are having any problems. | ||
+ | * If your customer is looking at some specific content on a website (say ski trips to Utah), you might send an email the next day with a special invitation (10% off your next ski trip to Utah if you book in 24 hours). | ||
+ | == Getting Permission == | ||
+ | |||
+ | If you're doing email marketing, it is absolutely critical to make sure you //actually have permission to email your users.// | ||
+ | |||
+ | * **Make sure your users know they' | ||
+ | * Have them signup online (by entering their email address), and then **send them a confirmation email** to let them know they' | ||
+ | * You may want to do a **double opt-in**, which means you opt-in via the website and send them a 'click to confirm your email address' | ||
+ | * You can present **a very clear checkbox** during your order/ | ||
+ | * One technique I've seen used presents **a Yes/No radio button with //neither// option checked** by default. If the user tries to checkout without selecting either option, the validation routine forces him to select one or the other. | ||
+ | * **Setup a subscription management center,** or ' | ||
+ | * Be sure to **give your users a ' | ||
+ | |||
+ | == Other Resources == | ||
+ | |||
+ | Here are some excellent articles on email creation that I've found: | ||
+ | |||
+ | * [[http:// | ||
+ | * [[http:// | ||
+ | * [[http:// | ||
+ | * [[http:// |