I’m sure you would agree that split testing and affiliate marketing go hand in hand.
It can also be difficult to figure out what to split test first.
In this article, you will learn how to go from newbie to a split testing ninja so that your campaigns have a fighting chance at becoming profitable.
Why Does Split Testing Work?
Because split testing will help you identify which ad will give you the cheapest cost per 1000 impressions (CPM), highest CTR (click through rate) and lowest CPC ( cost per click).
Let’s consider what factors impact whether Facebook shows your ad:
- Who you’re targeting
- How you’re targeting your audience (demographics, interests, custom audiences, geo’s, etc)
- How many advertisers are bidding for the same audiences you are
- The campaign objective for your ads
- The quality of your URLs on your bridge page
- The quality of the Facebook Page that you are running your ad’s from
- Your Ad text and creatives
- How much the ad set spends on a daily basis
Creating a process where you strategically analyze multiple variants in your campaign can help you work with the algorithm for results instead of trying to outsmart it.
Staying organized and having a clear goal will help you make sure you are tracking everything correctly is the first step..
The data will tell why or what you should start split testing in order to increase the particular KPI ( key performance indicator) that is the bottleneck of your campaign and influence Facebook’s decision to let your ad win the auction against your competition.
In the end, you will put forward the best ad that will help you get the cheapest CPM for your campaigns.
Split-Testing Allows Data Take The Steering Wheel
The first mistake you can make is being too ambitious and wanting test everything you possibly can. What you don’t want is a pile of data that is hard to understand as a result of throwing 50 ads in one ad set to start a new campaign.
Testing too many variables too soon will do more harm than good.
Let’s assume that the ads that publish are approved and running. Now is your chance to dive deep into your data and figure out what makes your audience tick.
- What exactly are you looking for?
- What makes one ad different from the other?
- What exactly are you testing?
Remember, the signs will be in your data.
So what do you analyze first?
For starters, you should be keeping an eye on what variables can improve your CPC.
Is your CPC Is higher than expected?
Then that means that you need to focus on designing more appealing ads. Your offer and the audience that you are targeting may have good, but your ads aren’t getting any action.
Focus on creating ads that get the click. That should be your main goal.
To make your audience click on your link and take action you can:
- Make sure your call to actions on your landing pages are clear and more assertive
- Test different headlines
- Rotate images and ad copy to find the right pattern interruption to pull your audience in
CPM higher than expected?
Then your offer is good but the audience who is seeing it is not interested. If your offer doesn’t appeal to certain genders or ages in your audience then you test a new one. You can also play with targeted interests or even custom audiences.
On the other hand, having a high CPM can also mean that your audience targeting is correct, but you may not have presented the offer to them in proper context. That’s why you should have multiple variables of ad copy or images to see what works best on your ads.
The CTR is the key to making sure you have cheap CPM’s and low CPC’s. High CTR’s imply that the ad’s performance can be potentially good for your campaign.
Even though having more clicks means more leads to your offer, you can also risk the chance of having a high CTR where your ad spend can go towards clicks that don’t convert.
Split testing to see what makes your audience click on the ad and the landing page will help you optimize your campaigns for more conversions.
December CPA Marketing Mastermind Now Enrolling!
Sign up now while spots last!
How to Split Test Like a Boss
Keep It Simple
Again, too many variable cause too many headaches. You’ll have data , but you wont have any data you can organize and visualize clear patterns from your campaign.
The simplest thing to do is to test 1 variable at a time , find the winner, and the move on to the next set of variables.For example, let’s say you have one offer. You want to test 5 different images. Use the same ad text and headlines for each image in the ad set instead of having different copy.
Facebook will optimize and test which image creates the best performing ad.
From there you can test your winning image and then test that one image with different ad text in another ad set.
After the testing for this new combination is complete, you will have now found your winning image and copy. Then you can also move on to the Headline and Description that goes at the bottom of the Facebook ad.
Automate Your Split Tests for Accuracy
You can kick your split testing into high gear with some of your tools.
Using third party performance tracking software will give you additional automation functionalities. For example, in Voluum, you can rotate offers directly with your campaign.
Sure, you can set up 2 campaigns in Facebook ads manager and just analyze your data later. But think about how much time you can save if you automate the process and let your tracker filter out your traffic for you.
For instance, you can set up flows in your tracker you can rotate landing pages for different traffic sources. Once you have your Facebook ads campaign all set up in your tracker, to test your different landing pages and set up the offers strategically and you can make the entire analytics process a piece of cake.
Be Obsessed with Organization
According to Google dictionary, the word system means a set of things working together as parts of a mechanism or an interconnecting network.
To make sure you’re entire campaign ecosystem is good to go, you’re going to need a system.
But not just any system. You will need a way to keep track of your ads in your tracker and also in your campaign planning.
You will need one that is organized and can be accessed easily. One that will save you time, keep you on track, and help you identify patterns faster.
Dynamic URL Parameters
This gives you the ability to track your campaigns directly from Facebook ads into your tracker. When you are testing mulitple ad variants, you can label your ad sets and ads accordingly. In return, you can glance at your data and easily tell where your clicks and conversions may or may not be coming from.
The dynamic tracking parameters can be built inside of Facebook’s ads manager or within your tracking software. The great thing about tools like Voluum is they have tons of templates for traffic sources already uploaded into the platform.
You can add your own dynamic tracking tokens to your URL in Facebook manually if you want. But Voluum makes things 20x’s faster and easier.
For Facebook, all of the dynamic tokens are already inserted for you. When you create your campaign URL within Voluum to insert into your ad, it will push back the the information added from the Traffic Source settings and it append to the URL for you.
Sign up for Voluum and create a Facebook Business manager account to run your ads. Set up your accounts, create your campaigns in each platform.
Get a discounted monthly Voluum subscription or save up to 45% with an annual plan
Add Facebook as a Traffic Source in Voluum. Use the template added into the system for you. The dynamic tracking parameters will be passed back from the traffic source to the tracker.
Complete the campaign in Voluum and paste your Campaign URL into your Facebook ad. It should look like the following.
After your ad gets published and starts getting clicks, the data will be sent back to Voluum for you to analyze. You simply pull up the campaign, access your Reports, and choose the appropriate variable that you want to review.
Create a Launch Schedule
Keeping a launch schedule and or a spreadsheet organizing what copy you’re testing is your ticket to the bank.
Knowing exactly what is working when you are ready to scale is going to bring in the profits that you have seen in my YouTube videos.
Organizing exactly what you are testing helps in case you may have multiple winners that you can use in your campaigns. Likewise, if you have plenty of losers to choose from, just cross them off the list and take them out for the ad set.
Split Test Based on Your Goals
As you are split testing make sure you know where true interests lie. The more in tune you are with your audience, the better your results will be.
Sometimes its as simple as just testing an image, color, or ad copy. But factors like page speeds, translation, angles can make or break the success of your campaign.
Variables You Can Split Test:
- Load Time
- Questions / Angle
- Colors / Images
- Facebook Ad
- Ad Text
Launch your campaign and make sure that all landing pages and URLs have dynamic tracking parameters. This will help you drill down your conversion data in your tracker.
Use your data to figure out which specific part of your campaign could use some additional improvement.
Optimize your campaign with your new information and continue to monitor your campaign’s performance.
Take Your Split Testing to Another Level
By breaking your campaigns down like they are mad science experiments and embrace the testing process. Analyze every major component that impacts large volumes of traffic.
Hopefully you gained some ideas on where to start first. The effort that you put in to keeping track of what you are testing will pay off in the long run.
You can reinforce everything you have learned in this guide and combine it with additional techniques that are taught in the in the AFK forum or the CPA marketing mastermind.