One critical part in your pay per click management campaign that we see most people not doing right is A/B testing of ads. A/B testing, or split testing is when you have baseline data, and then you do one change at a time to improve conversion rates. You only do one change at a time, so you know if your results were positive or negative. By doing too many changes, you will not be able to pinpoint which change made the result. To start A/B testing, you can either create two different ads going to one landing page, or two ads that are the same going to two different landing pages. Now make sure you edit the settings, so the ads rotate, and show evenly, that way you have data that is 50-50. Let these ads or landing pages run for at least 100 clicks each for optimal results, but if it is not highly searched let if go for several weeks. Now let’s analyze the results.
The amount of data in the results may seem a bit overwhelming for some, but Windy City Strategies internet consulting helps our customers understand what each part of the data specifically means to the navigability of the website so that the client can adjust its internet marketing where it’s most needed.
First, let’s take a look at the two different ads going to one landing page. The ad with the highest click through rate is the one that potential customers like, but that does not mean that it is the best. If you are an informational or service site that does not have conversion tracking, the click through rate is the most important, and if that is the case, take the ad with the lowest CTR, and create a new one to compete against the ad that performed the best. Now if you have conversion tracking, the highest click through rate does not mean that it brought you the most conversions. You have to take this into account while implementing pay per click management strategies, because conversions make you money.
Now the scenario with two of the same ads going to two different landing pages is about the same as the last scenario, except now you are analyzing the landing pages. Which one performed the best? Again, look at the click through rates if you don’t have conversion tracking, and use the best performing landing page, or create another page to compete against the better performing page. If you don’t have another page to analyze, create another ad going to the better performing landing page, and now do A/B testing for ads. If you do have conversion tracking, use the best converting landing page, and create another ad, and start doing A/B testing again. I hope these scenarios and tips help you with your internet marketing.