Split Testing Like Crazy
Okay… This one reads like a SAT Math Question.
(But don’t worry, it’s not hard.)
Pop quiz hot shot!
Bob and his brother Jerome areco-owners of a retail store specialize in Widget. Just recently, they each came up with a new packaging design for their beloved Widget (in case you didn’t know, packaging has a significant impact on how well a product sells) and both swear by their new designs.
Being the competitive types, Bob and Jerome got into a heated argument about whose design is better and therefore should be used for the packaging of their Widget.
Bob claims that the color he chose is brighter and therefore will attract more attention while Jerome claims that his design is transparent which will attract more interest. It became such a big deal and for days they couldn’t settle it that instead of letting it get in the way of their brotherhood and hurt their relationship, they decided to call upon their mom to decide.
What do you think is the logical choice for her mom to make:
- Bob’s
- Jerome’s
The Answer
If you said a. Bob’s, I hate to tell ya but you are wrong.
And if you said b. Jerome’s, you are also wrong…
See, the answer isn’t that simple. As a matter of fact, the answer isn’t even dependent on Bob, Jerome, OR their mom!
The answer actually depends on their market - and the answer is simply this…
Let their market tell them what works better. Run a split-test where on one shelf the customers will find Widget in bright packaging while on the shelf immediately adjacent they will find Widget in transparent packaging.
And at the end of each day, take heed of which one sells better. And that becomes the answer.
Lesson #1:You let your market decide for you rather than just second guessing. What you might think would sell better may oftentimes turn out to be dead wrong. And when you turn out to be dead wrong, you lose money. Period. |
Split Testing in AdWords
You need to practice this very same concept of split testing in your AdWords campaign. You have to constantly tweak your campaign and let the market tell you which ad performs best, which keywords gets you the more relevant reader, which URL attracts the most active buyers etc.
Split testing in AdWords encompasses many factors on various levels.
And it is most easily done on the ad level - trying it figure out what ad gets you the most relevant click. This is done so through adding more than one ad per ad group and making sure your “Ad Serving” option is set to “Rotate” (found in “Campaign Settings”).
Then going to your ad groups and adding in more than one “Ad Variation” (write another ad).
Lesson #2:Start by split testing your ads, this is often the easiest place to start since AdWords itself offers tracking of your ads’ CTR. Just be sure to change your “Ad Serving” to “Rotate“. |
Some factors you want to split test on the ad level include, but are not limited, to the following:
- Titlecasing (would it improve your CTR if you capitalized all the important words?)
- Adding in symbols such as quotes, ©, …, !, ?, etc.
- Change your punctuation marks (have a comma instead of a dash, etc.)
- Changing your URL (what if you add an extra “/Keyword” to your display URL such as going from www.YourWebsite.com to www.YourWebsite.com/Widget)
- Changing up the theme of your ad (perhaps go from focusing on the pain to focusing on the benefit)
- Change the tone of your ad (be more “in your face” as opposed to being passive)
The list goes on… Once you made your changes, let your ads run side by side for a bit and observe the discrepancies in click thru rates. Dump whichever got the lower CTR over a period of time/number of clicks, take note of it, and make a new split test against the winning ad.
Always try to outdo yourself, this is the secret to getting high conversions - letting the market tell YOU what works.
(Now… if you want to abide by the concept of “standing on the shoulder of giants”, you can get a head-start by checking out 43 PROVEN split-tests with results you can implement right away, click here.)
Lesson #3:Always be split testing and trying to “beat” your best. This is how you get better and become the most efficienty and effective AdWords marketer you can be. You can get a head-start with your split testing by checking out these killer 43 split test results: |
More AdWords Split Tests
Split testing also goes beyond just on the ad level. For example, you can split rest from a settings perspective. These factors include, but are not limited, to the following:
- Geographical location (are the English speak population in Australia as prone to buying your product?)
- Day, night, all day of advertising (should you run your campaign ONLY during the day, night, or run it all day?)
- Weekday/weekend (what about running your ads on the weekdays versus the weekends?)
- Position (would you save money if you opted for going for a lower ranking ad such as at positions 4 through 7 or would it make more sense to be # 1 through 3?)
- Broad, phrase, exact matching, or all three types (which match type of a keyword is profitable?)
- Standard versus accelerated (you have the option of spreading out your ads throughout the day depending on your budget or have them show up as often as possible… which one nets you the most profit?)
- Negative words (what negative words will have a positive impact on your campaign?)
I’ve listed just a few of the split tests you can possibly do to fine tune your campaign and let your market tell you what’s your best marketing piece. These split tests are on more of a campaign level and aren’t as easy to test as testing for the best ad, but they are still well worth testing - you just have to get a bit more creative in setting up the test.
Lesson #4:Split testing goes beyond just the ad level. Look to the campaign level also. |
So when it comes to testing your campaigns and ads, don’t pull a Bob and Jerome…
Leave your mom out of it.
To the Top,
 |
| Lawrence Talent |