How to Choose the Right Elements to Test

With multi-variate testing, you can sift through 100 ideas a lot faster and in a more effective manner than with traditional split testing methods. That's why Taguchi testing is most valuable when many elements are tested at the same time.

Your job is thus to determine which factors you think are most critical when it comes to a buying decision with your visitors, so you need to pick elements that have the most influence in the conversion process. You could think of just enough elements to plug in to the test on your own, but you probably won't see a significant improvement if you do that.

Brainstorming potential elements to test with a group of diverse people is the key to running a successful test. The more people you get to help you, the better. If possible, recruit 15-20 people (or as many as you can) to attend a brainstorming session together. Provide each with a copy of your landing page as well as some advertisements for your product. Then ask each person to take five minutes to write ways he or she thinks you can make your page sell more.

Once the timed writing is done, go around the room and ask each person to share one new idea. Write them down as you go, and continue to ask for new ideas until no one has anything else to share. Don't pause to discuss any of the ideas, and certainly don't criticize anyone even if the idea seems a bit ridiculous. Just keep the ideas flowing, and you should be able to reach 100 without a problem. Other ways to stimulate ideas include reading books on marketing and studying your competitors.

However you choose to brainstorm elements to test, keep these two rules in mind. One, don't choose elements that interact with each other. You want them to be as independent from each other as possible so they will still make sense when combined with each other during the testing process.

And two, pick elements that have the greatest possibility of significantly improving your conversion rate. Some, like the headline, offer and price, should be highly significant. Some, like the header graphic, placement of testimonials and the pre-headline might make a difference. And some, like font style, order button and style of order form border, would be interesting things to try.

What elements can you think of to test? The "right" answer for your product in your niche may not be the "right" answer for another marketer in another niche. You basically need to experiment with a lot of different ideas to find what results in the best conversion rate for your business.

Next, create alternatives for each element selected. This is where you utilize your copywriting skills. If one of the elements you are testing is your headline, you need to write six versions of your headline for the test. Actually, you need to brainstorm at least a dozen different headlines, then choose the best six from those options.

Also, make the alternatives of each headline-and all the elements-vastly different from each other. By only changing a word or two in each headline, you probably won't see much difference in the way the headlines perform in the test.

A small group of 3-4 people could greatly help you with this process. By combining your knowledge and expertise, you come up with more effective alternatives than you could have developed on your own.

Another option is to hire a professional copywriter. You can have the copywriter compose just the original letter, and that would provide you with a springboard for your own ideas.

Or you can have the copywriter create the original letter and all the alternatives of the elements you intend to run in your test. Then set up your multi-variate Taguchi test in a program such as the Split Test Accelerator.

Before you begin your test, preview it. Visit each page. Verify that each loads properly and shows what you want it to show. Also make sure that you are able to generate an impression and action on all the pages.

When you are sure everything is in working order, reset your test and let it run. Track the results during the testing phase. As soon as you determine the test has run its course by generating enough traffic and conversion, examine the overall results.

Which elements are impacting sales? Which aren't? What is the best theoretical combination of elements based on the results? And what hidden elements can you add together to gain massive improvement?

Determining the answers to those questions should lead you into the next test, a confirmation test. Take the "winner" of all the pages and test it against your control. The more you test, the more you'll know about how to improve your conversion rates and make more money.

Seed Newsvine

Bookmark this article in your preferred program

Comments

Add Comment

Popular Articles in this cathegory

This article is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.