3. Pay attention to language.
“Whenever we do an audit for a new client, one of the things we look at is the language that they’ve been using,” Dennison says. “We want to understand what kind of language works, and when it connects with the customer.”
Sometimes, it can be difficult to tell which variables cause some emails to succeed and others to fail. But at the heart of good email is writing and language. As you’re running an audit for a new client, make sure you’re not only looking at subject lines and CTAs—also take a careful look at the body content.
4. Every great idea needs great data.
“When we can show a client tangible evidence about how people are engaging, we get better traction,” Dennison says. “We have enough experience to know what’s good strategy, but being able to generate a report or comparison with the client’s own data makes the argument for us.”
The data available with each of your clients will vary, and sometimes a client might not have much for you to sift through. But as you develop strategy, using client data to support your case will help to affirm your recommendations.
5. Make a habit of testing.
“Trial and error is a good thing. With email—and marketing in general—you can’t be afraid to test things,” Dennison says. “But it has to be a habit. It has to be something you do with every email you send.”
Every email you send is an opportunity to collect information on your subscribers. Are you taking advantage of it? Like any habit, it takes a little repetition to get this one to stick, but once you’re used to regularly using every email to test a variable, you’ll see that it’s actually pretty easy to keep up with. And for a relatively small time investment, the payoff is huge: accurate, actionable data.
6. Take advice with a grain of salt.
“Every subscriber is different. There’s a lot of advice out there—a lot of articles detailing best practices for when to send emails, how to write subject lines. But that’s only ever a start. Your testing is what’s most important,” Dennison says.
Yes, we recognize the irony here, but it has to be said: Articles dispensing advice on email can only take you so far. Hundreds of thousands of words have been written on how to get the most out of your email campaigns, but the best intelligence you’ll ever get is what you test and measure with your actual subscribers.
7. Don’t make big assumptions from small numbers.
“You want to split test a few times, but you don’t want to keep split testing the same thing forever,” Dennison says. “You need enough information to say, yes, this data is correct. But you always want to make sure you’re testing multiple variables over time.”
Let’s say the first time you used an emoji in your subject line, the open rate spiked. Is that enough information to decide how you’ll run campaigns from now on? Certainly not. But by the same token, that emoji can’t be all you test—100 split tests on that 1 variable won’t yield much useful data.
Find the happy middle. Repeat a few tests, but also test multiple variables. The data—and your conclusions—will be the better for it.