Tuesday, December 8, 2009

How hard is your newsletter working for you?


It's been a while since my last blog. I just finished up with the Idaho Hoops Preview, a project that I've done each of the last four years. Check it out by CLICKING HERE

This post is about some research I did around email marketing. Specifically our electronic newsletter. A big part of Idaho Select's relationship building is through our newsletter. Producing it is not easy, requiring several hours each week. To make sure that my efforts in email marketing campaigns are maximized, I dug into the newsletters that we've sent out since August. My goal was to find what worked best and what wasn't working. 

The following analysis is based on seven newsletters sent out every second and fourth week of the month (CLICK HERE to learn a bit more about the newsletter). I use Constant Contact to track my emails. I can see who is opening them, where they are clicking, etc. It's not rocket science and never 100% accurate but the results do help me gauge how the campaigns are doing.

Any successful newsletter knows its readership and builds trust. The Idaho Select Newsletter is finishing up year number two so we've been building that trust for a while now. The results presented here are based only on our numbers.

How does our newsletter measure up to others? There is no true apple to apple comparison with industries since calculations and definitions vary. And, Idaho Select doesn't nicely fall into a standard category such as an accounting company or a law firm. But, we are in the top 20% of most standard categories, no matter which one you look at. You can look at your own industry by clicking here for Mail Chimp and here for Constant Contact. If I was forced to put us in a category, we probably fit best in Sports & Recreation or Marketing/Event Planning.


Getting more people to open your email
Headline vs. Sender vs. Timing

When I started the newsletter two years ago, the data that was out there all suggested you send emails Tuesday-Thursday between 8:30am-11:30am to get the most people to open them. This seems pretty accurate except for the time of day. All of the emails were sent on either Tuesday or Wednesday between 9:30am-3:30pm. There was no real difference in open rates based on day or time.

The next factor was the subject line. All of them had "Newsletter" in them along with other details. This didn't seem to factor into open rates either. The lowest rate we got was when we used the word "free" and "Twitter". Likely "free" activated spam blockers. I'm not sure on "Twitter" other than it appears my subscription list just isn't interested in that form of social media.

The one big factor appears to be who the email was from. In an effort to make the newsletter scalable, I switched the From: line to newsletter@idahoselect.org from vince@idahoselect.org. This long term strategy hurt the open rate initially, dropping it nearly 7%. I believe that the new address triggered spam filters and wasn't recognized by our readers as a worthwhile open. Though we have continued to get better open rates (the latest was 28.5%) we still haven't gotten back to anything over 30%.

Call to action
The Idaho Select Newsletter has two basic goals: create credibility and drive traffic to our website. The credibility comes through the consistent delivery of relevant content. We create web traffic by peaking interest and providing a link to a page.

Obviously, the more interesting the content the more likely it will be clicked. Here are a couple of things that help in our newsletters.
  1. Make the link a "CLICK HERE" link or a "check it out here" link. It tells them what to do and where to do it.
  2. Using the word "preview", "iconic", "winners" and "rankings" in or near the link drove up the click through rate.
  3. Put the link you most want clicked in the first paragraph. The first link in our newsletter gets clicked on 17.6% of the time. The second drops down to less than 6%.
  4. Build on the popular. Our section "Top 5 Links From The Last Issue" tend to get a great repeat click through rate. These are obviously interesting but I think people are always curious as to what other people are looking at.
  5. Combining any of these methods increases the effectiveness by more than an additive factor. In other words, put a link that has "CLICK HERE for rankings" in the first paragraph and your click through rate jumps way up.
Some things to avoid
  1. Don't just list the URL. For example, http://mysportsaffair.blogspot.com was not obvious enough to get anyone to click on it. They either didn't know it was a link or didn't know what to do with it.
  2. The last link rarely gets any clicks. The last link in our newsletters gets less than a 2% click through rate. There's really nothing you can do about this since one of your links has to be last. Just be aware that unless it's also the first link, it may be lonely.
What I didn't expect
  1. Links to video alone doesn't create a higher click through rate. The only way is if the reader thinks the video connects to them.
  2. Feel good headlines don't appear to move the needle. Those headlines generate about the same click through rate as a coach or player blog, camp or event. Not bad, but also not special.
Now what?
First, future newsletters will reflect the items that work. Second, Constant Contact provides a list of everyone's email that opens each email, who clicks where and who is getting bounced. If I had an intern I would have them track down everyone that was not opening the email or being bounced and find out what was going on. Third, there is a monetary value associated with these stats. They can tell who is going to product pages and sponsor pages. Since these are voluntary clicks they show the value of the newsletter. The goal here is understanding the best way to use them. Can they drive up the value of a sponsorship? Can they be used to determine an ROI on the newsletter? Ultimately, it is the answer to these questions that will determine where Idaho Select goes with the newsletter. When I find out, I'll be sure to post those here as well.