Mobile CPA Bootcamp Preview -- Full course coming soon. <br />Read Post: https://ppcmode.com/campaign-optimization/ <br /> <br /> <br />VISIT PPCMode.com FOR TONS MORE TIPS & TRICKS <br />https://ppcmode.com <br /> <br />SUBSCRIBE & NEVER MISS A VIDEO <br />http://www.youtube.com/channel/UC6DuyxNTs79lhZsUoMmYCiQ?sub_confirmation=1 <br /> <br />Twitter ⇨ http://twitter.com/brentrdunn <br />Linkedin ⇨ https://www.linkedin.com/in/brentrdunn <br /> <br />A/B Testing <br />Allows you to test one variant vs. another to see which performs the best. <br /> <br />This can be anything from a banner, landing page, offer and even different price points. <br /> <br />Usually, A/B testings consist of sending 50% of your visitors to one variant a and sending the other 50% to variant b. <br /> <br />ABTest <br /> <br />Pros: <br /> <br />By using A/B testing, you’re able to more quickly identify which of the two variants perform the best. <br /> <br />Since your traffic is only split 50/50, you will be able to reach statistical significants fairly quickly. <br /> <br />Usually, I only use A/B testing once I have a campaign that is already running profitably. <br /> <br />I use A/B testing always to improve the campaign by always trying to beat the current winner. <br /> <br />Cons: <br /> <br />Though it’s quicker to identify a winner, you have less of a chance to identify a variant that is out performing the rest of the group. <br /> <br />Multi-Variant Testing <br />Similar to A/B testing, multi-variant testing allows you to test more variants at a time. <br /> <br />Instead of sending 50/50 to A and B, multi-variant allows you to test as many variants as you want. <br /> <br />In the example below we’re testing four different variants: variant a, b, c & d each of which get’s 25% of our traffic distributed evenly. <br /> <br />Multi-Variant <br /> <br />Pros: <br /> <br />Multi-variant testing is best when you are first starting a campaign. It allows you to test drastically different marketing angle and styles. <br /> <br />This will allow you to identify which angle or style your audience responds to the best by comparing them all at once. <br /> <br />When I am first testing out a particular offer, landing page or banner I usually will use multi-variant testing. <br /> <br />Then once I identify a clear winner I will then revert to A/B testing to try and beat the current best performer. <br /> <br />Cons: <br /> <br />Since you’re splicing up your traffic so much with multi-variant testing. It’s going to take a lot longer to get statistical significance. <br /> <br />Optimization Through Testing <br />Though there is no perfect way for you to optimize your campaign, there is one thing that does matter. <br /> <br />Consistency <br /> <br />If you just launch a campaign at random with no rhyme or reason and then make changes to that same campaign a few hours later…. <br /> <br />…The data you did collect is useless. <br /> <br />No matter how bad a test appears to be performing, if you want to have accurate data you need to allow yourself to gather enough data. <br /> <br />This means that when you’re preparing to test something… <br /> <br />No matter if it’s a new banner, landing page, offer, etc <br /> <br />You need to setup the test and then leave it alone. <br /> <br />Start researching your next campaign launch and stop refreshing the stats like a maniac. <br /> <br />If nothing else takes a step away from the
