If you don’t have anything relevant to send then don’t send an email!

If you're sending marketing emails but skipping out on the analytics. You’re missing one of the biggest and simplest ways to understand your customer base. Not to say that email analytics doesn't have its challenges.

First major difference between email analytics and web analytics is, did the receiver ever get the email or did they simply delete and ignore? All you really do know is how many emails you sent, and hopefully you know how many links inside the email were clicked, and the bounce rate.

According to Anil Batra, the number 1 email marketing mistake is sending irrelevant emails to users that were kind enough to give the email address to you. Making the assumption that emails are free and there is no cost for a miss-sent email. "However, this is a big mistake because what they don’t realize is that ultimately they will lose the privilege to this “Free” marketing [1]."  

I don't mind getting marketing emails that are for what I already wanted to buy but was waiting for a better price. Who doesn’t? If I have to go through 10 emails or more to get 1 good email, I'm already counting the emails till you get the spam button. If enough people react this way on Gmail, Yahoo, or Hotmail you’re going to get blacklisted.

Wrong Spam

So how can analytics help us? First we have a few metrics:

Open Rate - Rate at which users open the emails.

CTR - A ratio showing how often people who see your email end up clicking it. CTR can be used to gauge how well your emails are performing. CTR is the number of clicks that your link receives divided by the number of times your email was sent.

ConversionRate - Rate at which your users start from an email and do whatever you were asking them to do in the email, donate blood, purchase a product, sign a petition etc.

Bounce Rate - User opened the email, clicked a link, but moved on immediately. Typically a sign that the email offer didn't match the expectation of the landing page or product page. 

How we can use these metrics to improve -

Use past email click-through behavior to determine what peaks user interest. If a customer has shown interests in certain products/categories/offers in past then they are very likely to be interested in similar products/categories/deals. Someone who repeatedly clicks on deal offers is most likely to open an email that says so and also click on a product that is on discount. Use that information to target. discounts are not the only thing, maybe they just didn't find the exact product they were looking for in a given category, emails for most popular items in a category will be helpful.

New hot items might be another avenue, but sending an email about a new fishing pole to a woman who has never shopped for fishing poles is not the best idea. In other-words use browsing history or purchase history to look at what users have been looking at but has not bought yet – Target the correct promotion [2]. Use checkout funnel abandonment history for the individual to determine what to send in an email. Use the frequency and recency of visits to the site to determine when to send him emails. Then look into coupons and type of promotions. If you do have to send something unrelated to persons interest, to provide an opportunity to see what else you sell, then send it along with something relevant.

[1] http://anilbatra.com/analytics/2013/01/number-one-email-marketing-mistake/

[2]http://webanalysis.blogspot.com/2008/10/relevancy-matters-in-email-marketing.html#ixzz2KonLSdlN