Ad Rotation & AB Split Testing (part 1)
Remember in high school science experiments when your teacher would correct you for changing more than one variable at a time? It makes sense: changing more than one variable at a time can make it difficult to determine which one had an effect on the end-result.
“You are an alchemist; make gold of that.” -William Shakespeare
Rotating and split testing ads in AdWords is very similar to a typical high school science experiment. It’s important to keep in mind that changing too many variables at once will end up giving you inconclusive data. “But my campaigns aren’t bringing in any conversions; I need to change SOMETHING, right?” Have a certain level of patience so that your decisions are based on conclusive data. You won’t be able to fix everything at once, so focus on improving the variables within your campaigns one-at-a-time rather than trying to juggle too many things at once. Panicking and making decisions on a whim will likely cause your advertising campaign to flop. First and foremost, only test two ads at once. Too many businesses try to run 3+ ads then just go with the one that performed best. There’s quite a few problems with this thought process. Running more than two ads tests way too many variables, so it’s nearly impossible to determine whether certain phrases such as “Free shipping!” or “Call today!” or “Best prices & service since 1988” performed better than others. If you end up testing 3+ ads against each other, there is likely to be ad copy that didn’t get a chance to stand out because it was buried when competing against so many other ads. Let’s take my dog training business as an example. Here are two ads I created: Dog Training in Oregon One-on-one dog behavior specialists; 17+ Years Experience. Call Today! www.AssertingAlpha.com Need Help With Your Dog? One-on-one dog behavior specialists; 17+ Years Experience. Call Today! www.AssertingAlpha.com Here’s a great example of testing a single variable at a time. Notice how only the headlines within these two ads are different; everything else is the same. If I let these ads run for a month and discover that my click-through rate or conversion rate in one ad is significantly higher than the other, I’ll be able to determine (with a high level of confidence) that one of those two headlines is more effective than the other at bringing in good traffic. However, if I tested two ads such as this: Dog Training in Oregon One-on-one dog behavior specialists; 17+ Years Experience. Call Today! www.AssertingAlpha.com Need Help With Your Dog? One-on-one dog behavior specialists; Obedience, leash-training, & more! www.AssertingAlpha.com With these two examples it’s difficult to determine whether the headline or the second description line had an effect on results between these two ads. I’ve seen accounts with 14+ ads running at the same time. You can imagine that it becomes increasingly difficult to determine exactly which phrases and headlines work best in your ads when you start increasing the number of variables in each test.
Adorable Puppy Selfie!
I’ll write a couple follow-up blogs that give a more in-depth look at what kinds of words, phrases, and headlines you should using in your ads to catch your customer’s eye. I will also be going over which pieces of data hold the most weight in determining whether an ad is performing ‘better’ than another ad. Stay tuned for more information on how to properly test and analyze your ads within AdWords. Cheers.