Follow

How To Efficiently A/B Test with AdParlor on Facebook

What is A/B testing?

A/B testing in advertising is simply a technique used to compare the effect of a particular variable (e.g. gender) on the performance of your campaigns. It works by holding all other variables constant and allowing the variable of interest to hold different values (e.g. male, female) so you can fairly distinguish any differences in the performance outcomes.

AB Test on Facebook with AdParlor

Why should I A/B test?

A/B testing allows you to identify which features of your targeting or creative are most likely to generate conversions for your advertising objectives. This enables you to spend your advertising budget efficiently.

 

How should I set up an A/B test?

If you want to do A/B testing on your targeting groups (e.g. males vs females) Facebook recommends that you split these up across different Ad Sets, not Campaigns. The Campaign entity will house all Ad Sets that have the same objective (e.g. get users to visit my website), and the split testing for targeting is done at the Ad Set. Testing for creatives is done at the Ad level. This rule of thumb holds for other things you might want to test, such as age, location, interests etc. 

This is a sample campaign I created for illustrative purposes:

My campaign objective: Offsite Conversions

My ad sets are targeted at:

  • New Zealand Males 18-65
  • New Zealand Females 18-65
  • Canada Males 18-65
  • Canada Females 18-65
  • UK Males 18-65
  • UK Females 18-65
  • USA Males 18-65
  • USA Females 18-65
  • Australia Males 18-65
  • Australia Females 18-65
 

Efficiently create your campaigns using AdParlor

Libraries are really useful for saving time if you are re-using many lists e.g. Interests, Geographies, Messages etc - it's best to get these set up before you start uploading.

Click on the +f button on the top left when you're ready to start creating your ads. 

During the Ad Creation process:

  1. Save targeting groups and use in future uploads to save you time and effort. It also enables you to divide up the work between your team members into targeting and creative responsibilities if required. 
  2. Save time by importing values for each targeting criterion from the AdParlor Libraries e.g. list of countries.
  3. The splitting feature allows you to rapidly create many targeting groups – it will create combinations for you. You can split across anything where you see the split icon.

Effortlessly generate multiple variations of your Post creatives

Use our bulk creatives tool to generate multiple unique variations of your creative aspects to perform a fair split test. The tool iterates over each creative parameter e.g. 4x images, 2x message text, 2x call to action buttons to generate 16 posts which are unique combinations of images, text etc. This makes it very easy to split test creative aspects (e.g. the image used) and easily generate many variants of posts to keep the material fresh for your users. 

Group your creatives together for easier management

Create and save groups of your creatives for re-use in other campaigns or for organizational purposes. For example, you can use this for separating bilingual posts - a group for posts with English copy, and a group for posts with French copy. This helps ensure you are always targeting the right people (French speaking) with the right posts (French copy). Another common example is to create groups of posts that have copy or images tailored for each broad targeting group (e.g. USA, Canada, UK). 

Upload your Ads to multiple Ad Sets quickly and easily

Perform a multi-Ad Set upload to launch your entire campaign in a single workflow. Simply pick the Ad Set(s) you want to upload to and pair with the creative groups you want to upload into those Ad Set(s). Repeat this process until you have queued all your uploads. Hit 'Create' and you’ve quickly launched a variety of powerful split targeting tests matched with the most appropriate creatives.

 

Analyze Campaign Performance and Optimize

The AdParlor all-in-one reporting and management tool, aptly named "Actionable Reporting", makes optimization easy. 

Analyze your data

Choose from predefined reporting templates to effortlessly compare performance between targeting or creative differentiators (e.g. post image or gender) across your Campaigns or Ad Sets or both. These 'Presets' are tailored depending on your overall advertising objective: 

  • Ecommerce & Direct Response Leads
  • App Installs & Engagement
  • Brand Awareness & Engagement

You can also choose your own grouping criteria, pick your metrics, and filter your results to generate a fully customized report that best suits your needs.

For example, by clicking 'Analyze Genders' I am able to easily analyze my data across all my Ad Sets to compare performance of males vs. females. This allows me to quickly and effortlessly gain insights to the trends that may be driving my results.

Whether I am looking at a custom or preset report, I can now generate a user friendly URL snippet so I can share this same view with clients, colleagues and any other users of the platform. These are referred to as 'Shared Actionable Reports'.

Optimize your campaigns

Using your findings, you can leverage the bulk actions functionality to perform actions across one, some, or all of your Campaigns/Ad Sets/Ads. This saves not only time and effort, but reduces the likelihood of user error when implementing modifications to optimize campaign spend. 

Examples of common modifications for optimization include:

  • Pausing poorly performing Ads
  • Increasing bids for all Ad Sets that are getting low delivery
  • Increasing budgets for all Ad Sets that are generating good results 

The actions available are specific to each level (Campaign/Ad Set/Ad) and can be applied to as many of each entity as you wish. 

Using the example from above, I can select the row that corresponds to female-targeted Ad Sets and click on 'Bulk Actions' to immediately modify all those Ad Sets. In this case I am choosing to reduce the Ad Set budgets as I am getting a higher CPC than desired. I can make a counter-change by increasing the budget for the male-targeted Ad Sets so that my overall campaign budget stays the same but I am allocating more budget to my better performing Ad Sets. This allows me to get a better 'bang for my buck'. 

 

To quickly summarize

  • Save time with re-usable targeting and creative groups
  • Split your target groups with a simple click
  • A/B test these split groups for deeper insight into which users are driving the best results
  • Bulk generate creatives
  • Advanced budget management 
  • Sophisticated Actionable Reporting now with Presets

Get started in seconds with our intuitive user interface and simple how-to guides.

Check out our website for more information on our other services.

Comments

Powered by Zendesk