How can I optimize emails?
I'm totally new to Optimizely and for what I see most of the information provided here is about websites.
I'm asked to help a client out with optimizing their email campaigning. But we aren't the one sending the mails out to the audience. We only deliver the mailings in HTML and they send everything out. Is there a way to implement a snippet in the HTML of the mailings so that I can track the goals?
Thanks in advance!
Solved! Go to Solution.
Optimizley is built to work with websites and mobile apps, but not with emails. Even with HTML emails, our project code (or snippet) will not be able to make changes or track goals.
However, if the email contains a hyperlink to a website, you can run Optimizely on that page, and target people who are coming from the email based on query parameters. Check out this article titled Optimize based on query parameters to find out more.
Hope that helps! Please let me know if you have any follow up questions.
Ehsan provides great advice in how to track clicks/conversions using query parameters within email.
Optimizely does not yet integrate with an ESP, however that doesn't mean you can't improve your email campaigns' performance!
A/B testing is the simplest way to measure effectiveness of a campaign, and if you have the Optimizely testing formula nailed down, you can do it even if you don't have optimization tools at your disposal.
If you A/B test your emails, you can provide two HTML emails to your client, each containing different link tracking parameters as Ehsan mentioned, and look at results that way.
When planning our email campaign, we set a conversion goal for that email (usually # clicks to the CTA). We then create a *fullproof* hypothesis, and isolate one variable in an email to test.
Once we set a hypothesis, we clone the email, to test one variation against the other. Using our Marketing Automation system, we then send one variation to a certain percentage of our audience, and the control to the rest of the group.
As a B2B company, and because we are already personalizing our email blasts to a certain persona, we aren't as concerned if the test doesn't reach statistical significance.
What we do look for is any leading indicator that shows why one email performed better than the other. If one did, you then can iterate on that variation, and you begin gaining invaluable knowledge about your customers, and what makes them want to read your emails!
Let me know if that helps, of if you have a more specific question about setting this up operationally.