Test, Test, Test

It’s a challenge writing and laying out all the elements for an email campaign: articles, links, images, calls‑to‑action, subject line and landing pages. When all is brain‑stormed and discussed, arranged and re‑arranged, said and done—you’re most likely to send it out to your entire list, lock, stock and barrel.

Then, once your campaign is distributed, the assumptions float to the surface: Were the links prominent? Was the image appropriate? Was the call to action eye-catching? Did the subject line influence open rate?

A/B Split Testing will reveal many answers to these types of questions. With consistent testing and within time the cumulative results deliver a solid knowledge base that positively impacts the long-term success of your email campaigns.

Let’s take a look at some simple split test protocols.

Test 1: Subject Lines

I live and breathe email. So, naturally I’ve seen every type of subject line there is; from the sales-driven to humorously absurd. Many clients ask: What do I think makes a great subject line? But the real question is: What do your subscribers think? Here’s food for thought.

Personalized subject lines work well in general, but you won’t know until you try!

Is there such a thing as over branding? If your newsletter subject line is an equivalent to name, rank and serial number, there’s an obvious element of recognition, but is it working in your favor?

A utilitarian believes that everything serves a purpose, or just does not belong. Is short and sweet better than any other approach?

Who are your subscribers, anyway? Consumers, distributors, constituents, prospects, proctologists? Each group speaks a slightly different version of the mother tongue. You speak their language!

Subject line A/B testing is undeniably the easiest split test to manage. Your campaign content remains unchanged while you split the distribution of your email into equal portions, randomly selected, and release with separate subject lines.

Picking the subject lines is the kingpin. Subtle or sassy. Generic or specific. Classic or contemporary. Test, test, test! Like a mad scientist, eureka - you’ll find revealing trends over a 3 to 6 month testing time period.

Test 2: Content

A good marketing manager will revamp content over time as a direct result of open rates and click thru metrics. What content resonates with your subscribers?

Your stats are key here, numbers rarely lie. If you see a downward slump in open rates, check back to the better performing campaigns. What made them tick?

To learn which stories best target your readers, prepare similar but separate features. Include read more links to help keep tabs on interest and watch the click-thru stats.

If you’re looking to improve goal conversions, review link placement and color, wording, and graphics. Too many calls-to-action dilute your strategy, so focus on the your campaign’s action items and A/B test two distinct versions.

Last word: don’t test subject lines and content simultaneously. It’s just too much information to digest.

Testing Protocols

A test should include a control page and your best effort to optimize performance. If you are getting started with A/B testing, your control page will be your current editorial content.

Establish your test goals and determine an amount of time sufficient enough to gather good data. Identify unique clicks and/or conversions. Collect your metrics and confidently declare a winner.

A/B testing provides an opportunity to resolve problems, increase conversions and goals, and challenge assumptions. Over time, testing your content will provide demonstrable results.