Mastering e-mail metrics: How to test your messages for responsiveness

The problem with e-mail

While having a list is great, there’s a problem we’ve noticed. Because e-mail is generally quite effective, and also largely free, many people don’t take the time to really study their numbers (also known as metrics) and track their ads for responsiveness.

In this article, we‘ll give you the low-down on which things are important, and tell you how to measure, calculate, and analyze the various campaign elements that effect profitability.

It doesn’t matter if you mail every day, or every three months; you can use what you learn to gauge interest, find hot buttons, and ultimately improve response throughout your entire sales process.

Tracking chart

Let’s start with the basics, what to track and how to track it. With an e-mail campaign, it’s a pretty straightforward process, with only 6 elements that need to be measured. Below you’ll find an easy to understand chart:

• If you want to know how good your subject line is, you’ll compare the number of pieces sent with the number of people who opened the message (open rate).
  
  
• If you want to know how good your teaser ad is, you compare the number of opens with the number of clicks (clickthrough rate).
  
  
• If you want to know how good your NameSqueeze™ page is, you compare the number of clicks with the number of opt-ins (opt-in rate).
  
  
• And if you want to know how good your sales letter is, you compare the number of clicks and opt-ins with the number of sales (sales conversion rate).
  
  

Reverse engineering a poor performing campaign

When you understand this process, you can use it to pinpoint problems in ad campaigns quickly. For example, if you send 50,000 pieces, get 2,000 clicks, and nobody opts-in, you can bet there’s something wrong with either the tracking link you used in the e-mail, or the form on the page.

If you get a 60% opt-in rate, but no sales, you probably have something wrong with either your redirect page (in the form), the link to the order form, or the form itself.

And so on and so on.

The nice thing is, because e-mail is so fast, you can see the sales process in action in real-time! Discover errors and weaknesses in just a few hours, and respond accordingly.

Below are four important techniques you’ll need to implement to stay on top of your e-mail testing.

Technique one: Use html mail

One of the key metrics you’ll want to measure is your open rate (how many people open the message) since that determines how well your subject line created interest and generated action.

A lot of marketers have an aversion to using html, so let’s deal with that objection right away. Maybe you think it’s more likely to get filtered, or it looks less personal. And all that may be true, but…

• Sending an html mail piece is the ONLY way to check open rate: Using the tools built-in to your e-mail provider (AWeber, GetResponse, etc.), simply choose the appropriate option to enable open rate tracking on your broadcasts.
  
  
• You don’t have to use html forever, just throughout the test: I’m not asking you to change all your mail pieces forever, just until the test results are over.
  

Actually, you might be surprised with the results. One technique that has done very well for us is to use a “fake” text piece. In other words, we’ll create an html e-mail designed to mimic a text message, right down to the hard line breaks and the 10-point courier font.

The two versions look almost exactly the same to a recipient, but the benefit is that you’ll get to see how many people open the html version. And you may just discover that “fake” text actually outperforms regular text, as we have ourselves many times;)

Technique two: Use MIME

Whenever you mail an html piece, you should also load a text message into your e-mail program, in the event the recipient can’t read html (about 10% of the Internet world). With many providers, this automatically enables a technology known as MIME, which will “auto-detect” the users’ capabilities, and deliver the correct version to their inbox.

10% doesn’t sound like much, but if you are sending 100,000 pieces, that’s 10,000 people who need the text version in order to read your e-mail. If they receive a garbled message, they’ll likely unsubscribe, even initiate a spam complaint, so it’s well worth enabling this feature.

Some providers require you to enable MIME manually. If that’s the case, make sure to check the appropriate box before hitting send.

Technique three: Use multiple tracking systems simultaneously

Many people ask what we use to track e-mail responsiveness. Because tracking programs are notoriously unreliable, we generally use two or three systems (simultaneously) to track a particular campaign, and then compare the results. For example…

• When we launch a new e-mail promotion, we’ll create an html and a text version and use AWeber’s Open Rate Calculator to measure opens.
  
  
• To track clicks, we’ll use both AWeber’s click tracking, as well as a third party tracker (like HyperTracker). This gives us double tracking.
  
  
• To track opt-ins, we’ll check both the HyperTracker stats, as well as manually count the number of subscribers in our AWeber account. Again, this double tracking means we aren’t relying on any one particular source for our data.
  
  
• To count sales, we’ll check HyperTracker stats and the numbers from our sales system (ClickBank, EasyOnlineSales, etc.).
  

Then we can compare the totals and contrast with one another, looking for trends, problems, and consistency with past campaigns.

Now listen, it doesn’t matter if you use different tools then we do, the process works the same way.

• Some people use Topica, GetResponse, or EmailAces as their e-mail provider.
  
  
• Some use AdMinder, ProAnalyzer, or TinyURL.com as their ad tracker.
  

Use what you both like and understand. The idea is to collect intelligence, for example, which subject lines stands out, or which NameSqueeze. page generates the highest opt-in rate, and use that information to improve results in future campaigns.

Technique four: Look at unsubscribe rate

Your open rate is important, but remember, if you upset, confuse, or surprise your list, they’ll only be opening your message so they can unsubscribe;)

When you are sending anything even remotely controversial (political, religious, profanity, sexual, etc.) always check your unsubscribe rate, especially if your open rate seems unusually large. This allows you to determine your message’s “annoyance factor.”

If your unsubscribe rate is over 2% consider toning it down in the future!

Four things that can screw up your metrics,
and what you can do about it!

There are a variety of snafus that occur when people test, and they can really screw up your results. Here are a few easy ways to overcome the most common ones:

• Keep the same return address and from field the same throughout the testing period: Your subscribers expect you to use the same mailing details consistently, and they set up their filters and white lists accordingly. If you change the mailing details, that alone can lead to more opens, more clicks, and especially more spam complaints! While you should definitely test the impact of different return addresses and from fields, those need to be separated from ad-copy and offer tests, which is what we are talking about here.
  
  
• Use the same e-mail provider throughout the testing period: Like the return address and from field, the provider you are using has a big effect on deliverability. Changing your ISP or e-mail provider can lead to big changes in response (both positive and negative). While you are testing teasers, subject lines, and other aspects, use the same e-mail company.
  
  
• Use link tracking (masking) consistently throughout the testing period: Many e-mail providers offer a “tracking link” option in place of the normal URL. This is done to track the click-through rate, and reduce spam complaints you might get if you send people directly to your own domain. Because a URL can trigger a spam filter, we recommend ALWAYS using link tracking when testing
  
  

Setting benchmarks

One question we hear a lot is, “What’s a good click-through rate?” And the answer is, of course, it depends. We’d love to be able to say a 2% or 3% click through rate is average, but then again, we have lists that regularly get 25%. And we also have lists that get less than 1%, but they happen to generate over six figures a year;) In other words, there is no “normal.”

The important thing to do is set your own benchmarks. That means keeping a running average of your totals, opens, clicks, opt-ins and sales, and comparing them each time you run a promotion.

Because you are mailing to your own list, presumably one you collected yourself, your response rate should be higher than a list you rent. But you definitely can (and should) use these techniques track third-party mailings through e-zines and newsletters as well.

Conclusion

As a marketer, you need to know what moves a particular list to take action. And the only way to do that is to study what they do when you send them a mailing.

Using these testing methods, you’ll be able to identify key metrics and benchmarks, compare different approaches and copy, and ultimately, improve your profitability.

Respectfully Submitted,


Jonathan Mizel