Dan MoyleAlways be testing. 👈 It’s become a marketing motto. But it doesn’t tell the whole story.
Yes, we should test tactics and strategies in marketing. But should we test each content piece we create?
Here’s a hot take: You don’t need to test everything, all the time.
Instead, let’s ask why. Why do we want to test an email subject line? Why do we need to test every email?
Look, HubSpot makes it easy to test myriad marketing methods in its tool. Call to action buttons, landing pages, website pages and emails all have easy-to-deploy test functionality.
But it’s important to know why you’re testing.
What do you want to know? Email clicks? Opens? Landing page performance?
Once you know why you’re testing, it’s time to set up a good test. HubSpot has a great A/B testing checklist to get you started.
Come up with a hypothesis and identify your goal.
“I think using personalization in email subject lines will lead to a higher open rate. We’ll add ‘first name’ to the subject line and measure open rates between the two versions.”
Then, set up the test, run it for as long as it takes to get a statistically significant amount of data.
Finally, be sure to actually look at the data and analyze it.
Many marketers fall short here. We set up an email test like that above and forget to go back to see what works. Then we just keep running A/B tests on our emails and personalizing the subject line, when in reality it doesn’t make a difference (perhaps).
Once you’ve chosen a winner, it’s time to ensure it’s a best practice at your company. So, document the process and the results for all to see.
But here’s where the hot take comes back: You don’t have to test everything all the time. Pick a metric you know will make a difference for your business. Test it. Analyze it. Document it. Then use that tactic.
When you’re tempted to do another test, ask yourself why. What are you wanting to test and improve on? If it makes sense to test, go for it. If it’s not something that will move your business forward, skip it (for now) and focus on efforts that will give you a return.
One thing to keep in mind when testing regularly: Just because something worked a year ago (or even 6 months ago) doesn’t mean it’s still right. Test it again from time to time. Always be learning.
