There’s a saying: “If you’re not measuring, you’re not marketing.” In
other words, successful marketing requires the ability to measure the
effectiveness of your programs and make adjustments to improve results.
By its nature, e-mail marketing is one of the easiest to measure
because e-mail and Web systems provide key statistics such as open
rates, click-through rates, and conversions. Some basic testing of your
e-mail campaigns against these metrics can lead to changes that will
make your campaigns more effective.
To test e-mail, you must send e-mail in HTML format and use a third party e-mail marketing service or your own internal solution that tracks statistics such as e-mails opened and links clicked. Virtually all e-mail marketing services offer this capability.
What Metrics Count
Open rate refers to the percentage of delivered e-mails that are opened. The factors that you can test that impact open rates are:
- Subject line
- From line
- Day of week sent
- Time of day sent
Click-through rate is the percent of recipients who click on a link in your e-mail. Those who click through are either interested in your offer or interested in reading more about a specific subject you are introducing. The factors affecting click-through rates are:
- Creative, such as layout and graphics
- Copy, including headlines
- Offer, such as downloading a white paper or registering for a Webinar
Conversion rate refers to the percent of recipients who accept your offer. These are actual downloads of your white paper or registrants for your Webinar. The factors affecting conversion rates are:
- All of the items listed above
- The Web site landing page
While the landing page is not itself part of the e-mail, any e-mail campaign with an offer should include a landing page for capturing conversions, so we include landing pages here as something to test.
Develop Your Test Plan, Prepare Your List
In developing your e-mail test plan, start with an objective. If your campaign is a simple newsletter or announcement, your objective may be to increase open rates. If you want to drive more traffic to your Web site, click-through rates are important. If your goal is to increase downloads of a white paper, then open rates, click-through rates and conversion rates are all important.
After deciding what you will test for, segment your list to perform the testing. The easiest way to segment your list is to divide it in half, and send half the list one e-mail and the other half another e-mail, then measure which one performs better. This is called A-B testing.
If you have a larger list, you may want to take a subset of it and perform A-B testing, then send the better performing e-mail, either A or B, to the remainder of the list.
One question that often comes up is how many responses you need from testing to feel confident the results are valid. No matter the size of your list or how many test e-mails you send out, a good rule of thumb is that 30-50 responses will give you an accurate trend; 100 responses give about a 95 percent accuracy rate, according to EmailLabs.
Test One Factor at a Time
One of the important rules of e-mail testing is to test only one factor at a time. Keep everything else the same, so you know how this one item impacts results. This doesn’t mean you have to send out dozens of test e-mails before launching your campaign. Rather, test something new with each campaign, and apply what you learn to subsequent campaigns.
Here are some ideas for testing. Use the ones that apply to your business and objectives:
- From line: From an individual in your organization or from the “brand” itself, such as the branded name of your e-newsletter.
- Subject line: Shorter or longer. Negative or positive. For example: “Your hydraulic pump might fail today" vs. "Keep your hydraulic pump operating."
- Time of day: first thing in the morning vs. lunch time vs. end of day.
- Day of week: Friday and Monday typically don’t perform as well as other days, but each business is unique so it’s worth testing.
- Creative: placement of images (or use of images at all) and headlines. One or two columns. Colors. Type size.
- Copy: Long or short. Different headlines. Bulleted lists vs. paragraphs.
- Offer: Where it’s placed on the page. Button vs. text links (or both). Repeating the offer several times. Hard sell vs. soft sell.
- Landing page: A single page or multiple pages. Placement of the form (to capture conversions). Use of phone numbers or e-mail. Creative and layout.
Learn From Your Testing
E-mail testing is an ongoing learning experience. You should keep testing with each e-mail, and with each test you perform, you will make incremental discoveries about what works best. Over time, as you optimize the different elements, your e-mail campaigns will deliver better results.