We've sent more than 1.5 billion emails since launching five years ago. We've made a few mistakes--and learned some vital lessons.
As an e-commerce company with more than 4 million subscribers to our daily fashion-sale alerts, we keep a close eye on what makes emails most effective.
We've sent more than 1.5 billion emails since starting our business five years ago and admit to our fair share of mistakes. We've gotten better, but along the way, we've learned a lot about what not to do.
Here are our top seven takeaways:
1. DON'T make your email one-size-fits-all.
Most department stores cater mainly to women--but if you were a guy walking into that store, no salesman would point you first to the women's beauty section. Yet close to half of companies make a similar mistake online, sending email promotions that couldn't possibly be relevant for the recipient.
Don't be one of them. Ask users what they like and who they are when they sign up—keep it quick and explain your reason, and almost everyone will comply. Then send different emails based on gender, age and other preferences.
Even if you can't ask, be creative. You can often make good educated guesses about your subscribers' gender and age based on what they have clicked on, which previous emails they've opened ... or even their email domains. (For example, Gmail users tend to be younger than AOL subscribers.)
2. DON'T skip A/B testing.
Marketing by email is a perfect A/B testing ground because you can immediately measure click-through to see what's working and what's not. Every time you want to change an email, send both the old and new variations to different random samples at the same time.
In our case, we spent a lot of time making our emails look beautiful. But when we tested the new designs, we found that some users responded a whole lot better to simple, plain text. Lesson learned.
3. DON'T ignore getting dumped.
We thought at first that when a user unsubscribed, she was in effect breaking up with us. But you can learn a lot by asking subscribers why they're leaving—and that can lead to potential improvements in your business.
Unsubscribers are often candid; they'll let you know what is not working for them. For example, we learned that some women were unsubscribing from Shop It To Me when they became pregnant. Our solution: We created a "take a break" option, which lets members stop receiving Salemail for up to nine months; we also boosted our maternity and baby clothing selections.
4. DON'T ask too much of the user.
If your email has more than one simple call to action, that's one too many. I know; you want your subscriber to take any of 10 actions. Every company does. But that does not mean you should put them all in one email.
Serving up too many options will make the email recipient think--and once they think, they're less likely to act. Instead, set one primary action for each email. You can vary the goals, but keep them simple.
Then, go back to them later for the other nine actions. Separately.
5. DO focus on benefits, not features.
When we launched new products or features, it took a while for us to figure out how to frame them for our customers.
For instance when we introduced a new brand recommendation service, our first instinct was to say that directly in the subject line: "We have brand recommendations for you." But we learned we could get much better results when we explained how the new feature would benefit our subscribers. ("Make your emails better by adding these 5 brands!")
If you are an enterprise software company that is launching a new reporting tool, don't just announce it. Focus on what that new tool will do for your customers, such as "Increase your sales by 20% with our new reporting tools"--and lead with that information instead.
6. DON'T become a pest.
This is a hard one. If your email marketing is successful, then with every additional email you send, you'll see a net increase in sales and actions ... in the short run.
But sender beware! Adding an occasional extra email is fine, but add new emails too frequently and that sales increase will subside--or even turn negative--over the long run. Users will fatigue quickly and put less value on your emails, or, worse, unsubscribe completely.
If you do want to add emails, run a test. Isolate a random control group and don't send them the additional emails. Then compare their behavior against the group receiving more emails over a, say, six-month period. You might be surprised to find who spends more money with you.
7. DON'T overlook mobile.
We all know that mobile is penetrating every walk of life. Recent reports from Pew and Nielsen suggest that more than three-quarters of smartphone users are using their phones to check email regularly.
Use your analytics tools to figure out how many mobile subscribers you have, and always consider mobile devices when building your emails. For example, remember that people click with their fingers and not a mouse--so make clickable "calls to action" big, and don't put tiny links next to each other.
Frustrate your users with a poor mobile email experience and they might not only ignore that email, they may start thinking poorly of your entire brand.
More from Inc.com: