Lever InteractiveLever Interactive Partners
Pin Blog Categories Pin Pin Blog RSS Pin Follow Us On Twitter Pin
Lever SEMPO Certified
Lever CIMA Certified


Pulling Levers Blog

Words of Wisdom

10 Quick Tips for Running a Website Optimization Test

Posted by on January 21, 2010 within Web Site Optimization. | 1 Comment

Running website optimization tests can be a very rewarding process. You are able to see firsthand the fruits of your labor. You may be surprised how the smallest of changes can have such large impacts in the overall outcome of a website. Many times, once you get started testing different elements of a website it is hard to stop. You will keep looking for things to test and ways to improve your website. This is great but it is also important to make sure that every test you conduct is worth your time and effort; not simply testing for the sake of testing.

Next time you set out to start a website optimization test consider the following tips:

1. Think About the Test
Before you jump head first in to rolling out a test make sure you really know what you are doing, what type of test are you going to run, what you are testing, and what you are trying to achieve from the test. The more thought you put into the test beforehand, the better crafted the test will be. The better crafted the test, the better results you will see. It is worth your while to put a little extra time and thought into the test before launch.

2. Test on High Traffic Pages
The more traffic you send to a test the quicker you will be able to see results. It is not a bad idea to run your first test on a low traffic website or page in order to get the hang of it but generally speaking the more traffic you are able to direct at a test the better.

3. Start with Small Test
With multivariate and split testing, you are able to test a wide variety of elements on a page, however, it is important to keep in mind that the more variations you are testing the longer the test will take. When running an initial test, start small by testing just a couple of elements on a page, 4 items maximum. From there you can take those findings and move on to a larger, more dynamic test.

4. Test on High Conversions
Ideally you want to run a test on something that will produce a high amount of conversions so that you are able to quickly gain insights from the test. For example, start by testing on how often a user adds a product to the cart rather than how often a user completes a purchase. Understandably, you want users to make the purchase but taking your test one step at a time will eliminate a lot of the guessing work involved when trying to figure out what is and is not working. Once you have users adding items to the cart, then worry about them successfully making a purchase.

5. Be Courageous
When you are running a test, make sure to test against variables that are distinctively different from one another. Testing a red headline against a light red headline probably is not going to yield much of a difference. Instead test a red headline against a green or blue headline. Make the differences notable and then, if need be, you can always come back and test slight variations later.

6. Analyze Your Test Combinations
Creating different test combinations takes a fair amount of time and effort. In creating these combinations you are likely to whip up some odd ball combinations; combinations that will most likely clutter the test and impair the results. As an example, using white text on a light gray background is not a great idea. Keep a close on your test combinations and make sure they are credible and do not run off track.

7. Inspect User Impact
The last thing you want to do is run a test in which your conversions tank, but it is possible. Look into the impact a test will have on your users, good or bad. Maybe you only want to run a test on a certain part of a website or perhaps even run a small series of tests. Make sure you are not flipping a switch for better or worse and always recognize the potential outcomes.

8. Keep Your Head Up
If your first test falls flat on its face, do not give up. It may take a few tests for you to see any improvements and finally catch on to how your visitors are reacting. No worry, optimizing a website is a long process, of which you need to be committed to in order to become successful. Keep your head up and be patient. If you are persistent you will run a successful test and find the results you have been looking for.

9. You Be the Judge
When it is all said and done, you know and understand your website better than anyone else. Make educated decisions based on what you believe to be the most beneficial for the website. Doing so can put your website on the map and ahead of your competitors.

10. Get Started!
Setting up and running an effective test might seem like quite a bit of work, and to an extent it is, but in the end it is all worth it. Do not put off testing because you feel like it is too much work or you are simply content with the conversion rate you have. Realize that there is always, always room for improvement and the test you start today could pave the way for greater success further down the road.

Written by Shay Howe

As a web and user interface designer at Lever Interactive, Shay believes good design can change the way users interact and utilize the web. Shay is a designer because he has a passion for solving puzzles while creating pleasing visual solutions at the same time. Please feel free to say hello to Shay on Twitter.

Please Share


Share on Facebook Share on Digg Share on Delicious Share on Stumble Upon Share via Email

Comments

One Response to “10 Quick Tips for Running a Website Optimization Test”


  • Sam – January 23rd, 2010 at 2:59 pm

    I would suggest give Clicktale a try. I’ve been using it for 2 months and it is neat to watch what your users do, I learned a lot.

Leave a Reply

Going Up