What works better on our website: the red button or the yellow button? Does a video of me work better than a photograph of an attractive woman? Should we put our phone number at the top or bottom of the page?
I’m all about split testing. I like to take an existing page on our site and contrast it with a new page and see what happens. We use Unbounce to do most of our tests. It’s effective and reasonably priced.
Test Accuracy Usually Depends on Web Traffic
We can conduct effective tests because we have substantial traffic at our site North Carolina Divorce. Traffic is an essential element for getting accurate results with these tests.
We get thousands of visitors per day, so we can easily conduct reliable tests in a few days and verify the results.
When a site has low traffic, under 1,000 unique visitors per week, it becomes very difficult to conduct tests with a high degree of reliability. There just isn’t enough data to be sure the results are accurate.
Unfortunately, many law firm sites are low on traffic. In part, they don’t get much traffic because they aren’t well designed based on scientific testing. We’ve got to approach testing differently when we’re dealing with minimal traffic.
You Can Still Get Accurate Testing Results Despite Low Traffic
Visual Website Optimizer, another excellent split testing tool, posted “How to Do A/B Split Testing on Low Traffic Sites” (http://visualwebsiteoptimizer.com/split-testing-blog/ab-split-testing-low-traffic-sites/) on its blog. It’s a detailed piece on the challenges and necessary procedures for testing on sites like those created by many law firms.
The article suggests a number of approaches that dramatically improve the reliability of testing low traffic sites.
It suggests, among other things, the following techniques:
- Go big or go home. Test something high impact. Don’t screw around with button color. The big changes are more likely to make a drastic impact, which is easier to measure on low traffic sites.
- Stick to A/B testing. The company offers multivariate testing. Unfortunately, this sophisticated software is wasted on low volume sites. Keep it simple, and you’ll be able to rely on the results you get. You need much more traffic to employ the more complex approach.
- Run sitewide tests. Test something that shows up all across the site. Many sites have sidebar content that runs sitewide. Test those elements first so you’ll gain the benefit of exposing the test to more visitors.
- Run qualitative tests. Rather than running A/B tests, try observing some users using your site and get their feedback. It’s a different approach, but it may move you forward faster than split testing when you have minimal traffic. Check out Feedback Army as an option. It offers great pricing.
The article offers a number of other great suggestions.
Testing is essential. Every site presents testing challenges and they can all be overcome. As you run successful tests, you’ll increase your traffic and you’ll move on to new challenges in testing your site. Try some of these approaches to speed your way to the next challenge.
